You are on page 1of 175

Conference 6736: Unmanned/Unattended Sensors and Sensor Networks IV

Tuesday-Thursday 18-20 September 2007 Part of Proceedings of SPIE Vol. 6736 Unmanned/Unattended Sensors and Sensor Networks IV

6736-02, Session 1
Sensor deployment on unmanned ground vehicles
G. R. Gerhart, U.S. Army Tank-Automotive Research, Development and Engineering Ctr. (USA) TARDEC has been developing payloads for small robots as part of its unmanned ground vehicle (UGV) development programs. These platforms typically weigh less than 100 lbs and are used for various physical security and force protection applications. This paper will address a number of technical issues including platform mobility, payload positioning, sensor configuration and operational tradeoffs. TARDEC has developed a number of robots with different mobility mechanisms including track, wheel and hybrid track/wheel running gear configurations. An extensive discussion will focus upon omni-directional vehicle (ODV) platforms with enhanced intrinsic mobility for positioning sensor payloads. This paper also discusses tradeoffs between intrinsic platform mobility and articulated arm complexity for end point positioning of modular sensor packages.

approach can be extended for multiple starting points to reach multiple goals. The simulation gives the time taken by the convoy to reach a goal from its initial position. In the battlefield environment, commanders make various tactical decisions depending upon the location of an enemy outpost, minefields, number of soldiers in platoons, and barriers. The simulation can help the commander to make effective decisions depending on battlefield, convoy and obstacles to reach a particular goal. The paper describes the proposed algorithm and gives the simulation results. The paper also gives a number of problems for future research in this area.

6736-05, Session 1
On the reliability of a convoy of unmanned intelligent vehicles and their collaboration and coordination
H. Singh, L. Hua, Wayne State Univ. (USA); G. R. Gerhart, U.S. Army Tank-Automotive Research, Development and Engineering Ctr. (USA) There is an increasing interest in the use of a convoy of unmanned intelligent vehicles for defense and security. These vehicles have a number of sensors associated with them. It is very important to have a highly reliable sensor network so as to determine the dynamic reliability of the intelligent vehicle system for a safe battle field environment. The mobility, path planning and navigation of such convoy of vehicles are in the state of infancy. However, it is considered important to develop the reliability techniques so that a commander in the battle of field can predict the reliability of the various stages of the movement of the convoy and then can take decisions depending on reliabilities determined at varies places and time. In this paper a combination of intelligent techniques like fuzzy and neural fuzzy and Boolean algebra techniques are exploited to determine the reliability of the wireless sensor network in the battle field. The branch of reliabilities are determined using intelligent techniques like fuzzy and neural fuzzy while terminal reliabilities are determined using Boolean algebra techniques. Based on this technique, a new algorithm is proposed in determining the dynamic reliability of convoy of unmanned intelligent vehicles. Such a vehicle will help in the collaboration and coordination of the convoy of vehicles.

6736-03, Session 1
Integrated multi-sensor package (IMSP) for unmanned vehicle operations
E. C. Crow, Jr., K. M. Reichard, C. M. Rogan, J. Callen, The Pennsylvania State Univ. (USA) This paper describes recent efforts to develop integrated multisensor payloads for small robotic platforms for operator situational awareness and ultimately for improved robot autonomy. The focus is on enhancements to perception through integration of electro-optic, acoustic, and other sensors for navigation and inspection. The goals are to provide easier control and operation of the robot through fusion of multiple sensor outputs, to improve interoperability of the sensor payload package across multiple platforms through the use of open standards and architectures, and to reduce integration costs by embedded sensor data processing and fusion within the sensor payload package. The solutions investigated in this project to be discussed include: improved capture, processing and display of sensor data from multiple, non-commensurate sensors; an extensible architecture to support plug and play of integrated sensor packages; built-in health, power and system status monitoring using embedded diagnostics/prognostics; sensor payload integration into standard product forms for optimized size, weight and power; and the use of the open Joint Architecture for Unmanned Systems (JAUS)/ Society of Automotive Engineers (SAE) AS-4 interoperability standard. This project is in its first of three years and will discuss the applicability of each of the solutions in terms of its projected impact to reducing operational time for the robot and teleoperator.

6736-06, Session 1
Algorithms and evaluation framework for uninhabited vehicles
M. Bernhardt, C. R. Angell, Waterfall Solutions Ltd. (United Kingdom); P. K. Kimber, SELEX Sensors and Airborne Systems Ltd. (United Kingdom) The nature of co-operating Uninhabited Vehicle (UV) systems is such that performance enhancements are likely to be a result of greatly increased system complexity. Complexity emerges through the interaction of multiple autonomous UVs reacting to their current surroundings. This complexity presents a fundamental challenge to the specification, design and evaluation of such systems, and drives the need for new approaches to the systems engineering. For applications involving multiple autonomous UVs, research into collective and emergent behaviour offers potential benefits in terms of improved system performance and the utilisation of individual UVs with lower processing complexity. This paper reports on the development of a new simulation framework that addresses the systems engineering issues and allows novel algorithms to be created and assessed. Examples are given of how the framework has been used to develop and assess the performance of individual and multiple UVs, as well as unattended ground sensors. Furthermore, a variety of novel algorithms developed using the framework are described and example results are provided. These include co-operative UV missions requiring improved detection performance and the improved management of unattended ground sensors to minimise power usage.

6736-04, Session 1
Simulation of convoy of unmanned vehicles using agent based modeling
S. Sharma, H. Singh, Wayne State Univ. (USA); G. R. Gerhart, U.S. Army Tank-Automotive Research, Development and Engineering Ctr. (USA) There has been an increasing interest of unmanned vehicles keeping the importance of Defense and Security. A few models for a convoy of unmanned vehicle exist in literature. The objective of this paper is to exploit agent based modeling technique for a convoy of unmanned vehicles where each vehicle is an agent. Using this approach, the convoy of vehicles reaches a specified goal from a starting point. Each agent is associated with number of sensors. The agents make intelligent decisions based on sensor inputs and at the same time maintaining their group capability and behavior. The agents reach the goal by avoiding the obstacles on their path through the use of sub goals. The simulation is done for a battlefield environment from a single starting point to a single goal. This

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6736: Unmanned/Unattended Sensors and Sensor Networks IV

6736-07, Session 1
Bio-inspired motion planning algorithms for autonomous robots facilitating greater plasticity for security applications
S. V. Desai, U.S. Army Research, Development and Engineering Command (USA); Y. Guo, Stevens Institute of Technology (USA); M. E. Hohil, U.S. Army Research, Development and Engineering Command (USA) Proposed are techniques toward using collaborative robots for infrastructure security applications by utilizing them for mobile sensor suites. A vast number of critical facilities/technologies must be protected against unauthorized intruders. Employing a team of mobile robots working cooperatively can alleviate valuable human resources. Addressed are the technical challenges for multi-robot teams in security applications and the implementation of multi-robot motion planning algorithm based on the patrolling and threat response scenario. A neural network based methodology is exploited to plan a patrolling path with complete coverage. Also described is a proof-ofprinciple experimental setup with a group of Pioneer 3-DX and Centibot robots. A block diagram of the system integration of sensing and planning will illustrate the robot to robot interaction to operate as a collaborative unit. The proposed approach singular goal is to overcome the limits of previous approaches of robots in security applications and enabling systems to be deployed for autonomous operation in an unaltered environment providing access to an all encompassing sensor suite.

stop when proximity sensor value reaches a limit) while maintaining teleoperations. This architecture involves a server which communicates commands and receives sensor inputs via a wireless modem to the ODIS platform. This server can take requests from multiple client processes which have prioritized access to on-board sensor readings and can command the steering. The clients would include the teleoperation soldier unit, and any number of other autonomous behaviors linked to particular sensor information or triggered by the operator. For instance, the behavior of certain tasks can be controlled by low-latency clients with sensory information to prevent collisions, place sensor pods precisely, return to preplanned positions, home the units location or even perform image enhancements or object recognition on streamed video. This architecture is well-suited for sensor-enabled advanced mobile platforms.

6736-10, Session 2
Autonomous laser accelerometer for platforms and systems
B. V. Melkoumian, Baghron Co. (Russia) New laser accelerometer is presented. Proved method consists in the using of standing wave of coherent radiation in the linear resonator as the sensitive element of the accelerated movement measurement. There are not in the autonomous resonatory device (ARD) any matter parameters altering during movement of the resonator, attached with object to be measured. It provides the best accuracy in the measurements. The devices can be applied as measuring transducers for control and stabilization of movement of the objects with measuring of accelerations and coordinates in demanded operating regions. ARD provides autonomous measurement of acceleration of actual motion at free falling of object as against to all other accelerometers.

6736-08, Session 1
Biomimetic approach for coordination algorithms utilized in distributed multi-robot systems
Y. Meng, Stevens Institute of Technology (USA); M. E. Hohil, S. V. Desai, U.S. Army Research, Development and Engineering Command (USA); J. Gan, Stevens Institute of Technology (USA) This paper presents a synergy of Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO) into a novel hybrid coordination algorithm for distributed multi-robot systems. The intended multi-robot systems are composed of relatively simple, expendable agents with highly decentralized, self-organized behaviors; which as a whole achieve global optimization over a set task. Basically, two coordination processes among the robots will be established. One is a stigmergy-based algorithm using the distributed virtual pheromones to guide the agents movement, the other one is interaction-based algorithm, where a global maximum of the attribute values can be obtained through the interaction between the robots. The proposed algorithm has been implemented on an embodied robot simulator Player/Stage under different dynamic environments. The simulation results demonstrate that the proposed hybrid swarm intelligence based architecture is feasible, efficient, and robust to coordinate a distributed multi-robot sensing system that can be outfitted to handle a specific set of missions with a given sensor suite.

6736-11, Session 2
Preparation and characterization WDM technique for linear disturbance localization in fibre optical sensor
M. Zyczkowski, W. M. Ciurapinski, M. Szustakowski, Wojskowa Akademia Techniczna (Poland) Ordinary perimeter security systems consist of many individual sensors with detection range 200-300 meters. These limitations are connected with physical phenomena that are used in microwave and infrared barriers as well as in ground and fence cable sensors. On the contrary, fiber optic perimeter sensors can be applied in the range of many kilometers and zone length 200-300 meters is degradation of their possibilities. This paper presents investigation results of a new generation of the fiber optic perimeter sensor using WDM technique. These systems can detect a potential intruder and determine its position along a protected zone. The work presents a method of linear localization of disturbance point in the two-interferometer and modalmertic fiber optic sensors in one optical fiber. These sensors are in one fiber optic cable with two- interferometers (Sagnac) and in transmissive configuration. Using WDM technique makes it possible to obtain new technical properties that can broaden effectiveness and scope of application. Arrangement of a laboratory model of the sensors and its signal processing scheme is presented. During research of a laboratory model, it was possible to detect the position of the disturbance in one optical fiber with resolution of about 50m along a 10-km long sensor.

6736-09, Session 1
An architecture to allow hybrid autonomous and teleoperated control of ODIS T2
S. T. Hunt, Wayne State Univ. (USA) and Turing Associates, Inc. (USA); Y. Li, Wayne State Univ. (USA); G. Witus, Turing Associates, Inc. (USA); R. D. Ellis, G. Auner, A. Cao, A. K. Pandya, Wayne State Univ. (USA) A fully teleoperated strategy is employed to control mobile platforms used for surveillance such as the ODIS-T2. In this situation, there is no on-board intelligence, and the robot will perform drive and steer and camera functions on demand from the remote operator. As the ODIS platform evolves to accommodate functions beyond surveillance, a more robust hybrid method of control is needed. It is envisioned that as sensor pods are deployed, and articulated arms and employed, and the system is sensor-enabled, the simple master-slave strategy will not be sufficient as the navigation and sensory inputs will become complex. In this paper we provide a tested software/hardware framework that allows the ODIS-T2 platform to offload operator tasks to autonomous behaviors (for instance,

6736-12, Session 2
Classical theory of autonomous laser accelerometer
B. V. Melkoumian, Baghron Co. (Russia) Classical theory of autonomous laser accelerometer based on Galileo transformations and Maxwell equations is proposed. Theory based on quantum operators for laser accelerometer had been published before with same results.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6736: Unmanned/Unattended Sensors and Sensor Networks IV

6736-45, Poster Session


Generalized ambiguity function in the linear canonical transform domain and optical systems
Y. M. Kozlovskii, Institute for Condensed Matter Physics (Ukraine) In this work we combine two concepts the linear canonical transform (LCT) and generalized ambiguity function (GAF). The GAF is proposed as generalization of usually ambiguity function (AF) and it is shown that Wigner distribution function and ambiguity function are only particular cases of generalized distribution. In this work we introduce canonical representation of GAF which is generalization both generalized Wigner function and the GAF. The GAF is characterized by two parameters: t the form representation parameter and p - the FrFT parameter. At we can reconstruct intensity distribution which is registered experimentally in the output plane of optical cascade. As we know in the case of fractional Fourier transform we have rotation matrix with restricted coefficients. The advantage of this method is that in the LCT theory coefficients of matrix are arbitrary. Consequently combination of the LCT and the GAT allows us propose new methods for optical systems describing. The GAF for rectangular impulse is obtained. The scheme of signal intensity distribution reconstruction by the GAF is proposed for rectangular impulse. The results of numerical calculations of rectangular impulse distribution in the fractional Fourier transform domain are presented. It is shown the Wigner function forms as the rotational displacement of ambiguity function on informational diagram. In particular cases of the GAF such method allows us calculate images for rectangular impulse which forms in the output plane of single and double optical stages in the LCT domain.

use of the relatively high bandwidth of such pulses by modulating different spectral portions independently but retaining the phase locking between them. The encoding method is based on special optoelectronic devices (multichannel electro-optic and acousto-optic modulators) designed for this special task. We elaborated prototypes of such devices and planned their integration in complete communication systems. The secure nature of the encoded information lies on the large number of possible phase-amplitude combinations usable for the same information at the same transfer rate. The basic phase relations must be calibrated for the given optical path, so a recalibration is necessary each time the sender or receiver units are moved. This feature excludes unwanted decoding with an arbitrarily placed third party receiver. We performed simulations of the pulse shaping necessary to perform the purposed task. The number of independently modulated spectral channels, N, and the resolution of the phase and amplitude modulation in each channel determine together the number of possible encoding combinations, M. The minimum hardware may allow the transfer of 2^N bits in one pulse, or equivalently, this value of security code deepness. The present technology imposes pulse bandwidth between 10100 nm to process 3-30 spectral channels independently.

6736-13, Session 3
Sensor networks: future technology challenges
J. A. Parmentola, U.S. Army (USA) No abstract available

6736-14, Session 3
Miniature, low-cost unattended sensors and networks for ocean and coastal monitoring
A. Bargnesi, US Navy, Naval Underwater Warfare Ctr. (USA); E. M. Carapezza, B. Javidi, Univ. of Connecticut (USA) No abstract available

6736-46, Poster Session


Application of Bessel beams to wireless optical communications
O. Wilfert, Z. Kolka, Brno Univ. of Technology (Czech Republic); Z. Bouchal, V. Kollarova, R. Celechovsky, T. Medrik, Univ. Palackho V Olomouci (Czech Republic) The paper deals with application of nondiffracting Bessel beams to wireless optical communication. Nondiffracting beams possess unique physical properties originating from a specific form of their spatial spectrum. Up to now, a particular attention has been focused on the nondiffracting beams with microscopic dimensions. Such beams have been successfully utilized in metrological applications, particle manipulations and electron acceleration. In this paper, the nondiffracting beams with a long propagation range are examined and their properties potentially applicable to wireless communications are discussed. In the case of the nondiffracting beam, there is a possibility to control the beam propagation length even if its transverse dimension remains unchanged. The best known zero order Bessel beam with the wavelength lambda remains propagation invariant in a region whose length L depends not only on the diameter of the beam central spot d but also on the diameter of the transmitter D. These parameters are simply related as d = lambda*L/D, so that a narrower beam can reach the same distance if the diameter of the transmitter is enlarged. Furthermore, the intensity profile of the nondiffracting beam can be shaped into a demanded form and can reconstruct itself if the beam passes through objects distorting its amplitude and phase. Our paper is focused on experimental determination of the maximal propagation range of the nondiffracting beam, on an efficiency of the energy transfer and on the beam behavior in atmospheric turbulences.

6736-15, Session 3
Failure prediction for satellite monitoring systems using Bayesian networks
S. Bottone, C. J. Stanek, DataPath, Inc. (USA); M. Spivack, Univ. of Cambridge (USA); M. OSullivan, San Diego State Univ. (USA); J. Rosalia, DataPath, Inc. (USA) Predicting failure in complex systems, such as satellite network systems, is a challenging problem. A satellite earth terminal contains many components, such high-powered amplifiers, signal converters, modems, routers, and generators, any of which may cause system failure. The ability to estimate accurately the probability of failure of any of these components, given the current state of the system, may help reduce the cost of operation. Probabilistic graphical models, in particular Bayesian networks, provide a consistent framework in which to address problems containing uncertainty and complexity. Building a Bayesian network for failure prediction in a complex system such as a satellite earth terminal requires a large quantity of data. Software monitoring systems have the potential to provide vast amounts of data related to the operating state of the satellite earth terminal. Measurable nodes of the Bayesian network correspond to states of measurable parameters in the system and unmeasurable nodes represent failure of various components. Nodes for environmental factors are also included. A description of Bayesian networks will be provided and a demonstration of inference on the Bayesian network, such as the calculation of the marginal probability of failure nodes given measurements and the maximum probability state of the system for failure diagnosis will be given. Using the data to learn local probabilities of the network will also be covered.

6736-47, Poster Session


Improving free space optical communication security using femtosecond laser pulses
P. A. Mak, A. Barcsi, L. Jakab, P. I. Richter, Budapest Univ. of Technology and Economics (Hungary) In our paper we present a new method for increasing communication security based on hardware codes. This method uses ultrashort (femtosecond) pulses in optical communication over free space. The basic principle is that a given amount of information is encoded in single optical pulses, using amplitude and phase modulation simultaneously. The encoding makes

6736-16, Session 3
Renewable energy for sustainable ocean sensors and platforms
E. M. Carapezza, A. Fagri, T. M. Molter, Univ. of Connecticut (USA) No abstract available

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6736: Unmanned/Unattended Sensors and Sensor Networks IV

6736-17, Session 4
Seismic augmentation of acoustic monitoring of mortar fire
T. S. Anderson, U.S. Army Engineer Research and Development Ctr. (USA) The US Army Corps of Engineers Engineer Research Directorate participated in the joint ARL-NATO TG-53 field experiment at Yuma Proving Ground, AZ in early November 2005. Seismic and acoustic signatures from muzzle blasts and impacts of small arms fire and mortars were recorded using 7 seismic arrays and 3 acoustic arrays. Arrays were located from 700 m to 18 km from sources. Location and characterization of these impulsive events is performed. Preliminary analysis of 60mm, 81mm, 120mm and 155mm mortar is presented. Initial results from seismic sensors co-located with acoustic arrays indicate that the seismic surface wave is a robust classifier for mortar fire at ranges of 1-3 km from sensor location.

6736-18, Session 4
Mortar and artillery variant classification by exploiting characteristics of the acoustic signature
S. V. Desai, M. E. Hohil, A. Morcos, U.S. Army Research, Development and Engineering Command (USA) Feature extraction methods based on the discrete wavelet transform and multiresolution analysis facilitate the development of a robust classification algorithm that reliably discriminates mortar and artillery variants via acoustic signals produced during the launch/impact events. Utilizing acoustic sensors to exploit the sound waveform generated from the blast for the identification of mortar and artillery variants. Distinct characteristics arise within the different mortar variants because varying HE mortar payloads and related charges emphasize concussive and shrapnel effects upon impact employing varying magnitude explosions. The different mortar variants are characterized by variations in the resulting waveform of the event. The waveform holds various harmonic properties distinct to a given mortar/artillery variant that through advanced signal processing techniques can employed to classify a given set. The DWT and other readily available signal processing techniques will be used to extract the predominant components of these characteristics from the acoustic signatures at ranges exceeding 2km. Exploiting these techniques will help develop a feature set highly independent of range, providing discrimination based on acoustic elements of the blast wave. Highly reliable discrimination will be achieved with a feedforward neural network classifier trained on a feature space derived from the distribution of wavelet coefficients, frequency spectrum, and higher frequency details found within different levels of the multiresolution decomposition. The process that will be described herein extends current technologies, which emphasis multi modal sensor fusion suites to provide such situational awareness. A two fold problem of energy consumption and line of sight arise with the multi modal sensor suites. The process described within will exploit the acoustic properties of the event to provide variant classification as added situational awareness to the solider.

suite is the system will provide this information on the move while the chemical sensor will have adequate time to determine the contents of the event from a safe stand-off distance. The CB/HE discrimination algorithm exploits acoustic sensors to provide early detection and identification of CB attacks. Distinct characteristics arise within the different airburst signatures because HE warheads emphasize concussive and shrapnel effects, while CB warheads are designed to disperse their contents over large areas, therefore employing a slower burning, less intense explosive to mix and spread their contents. Differences characterized by variations in the corresponding peak pressure and rise time of the blast, differences in the ratio of positive pressure amplitude to the negative amplitude, and variations in the overall duration of the resulting waveform. The discrete wavelet transform (DWT) is used to extract the predominant components of these characteristics from air burst signatures at ranges exceeding 5km. Highly reliable discrimination is achieved with a feed-forward neural network classifier trained on a feature space derived from the distribution of wavelet coefficients and higher frequency details found within different levels of the multiresolution decomposition. The development of an adaptive noise floor to provide early event detection assists in minimizing the false alarm rate and increasing the confidence whether the event is blast event or back ground noise. The integration of these algorithms with the TDOA algorithm provides a complex suite of algorithms that can give early warning detection and highly reliable look direction from a great stand-off distance for a moving vehicle to determine if a candidate blast event is CB and if CB what is the composition of the resulting cloud.

6736-49, Session 4
Multi-frame filtering techniques for the detection and recognition of moving objects
A. Mahalanobis, J. C. Perez, Lockheed Martin Missiles and Fire Control (USA) No abstract available

6736-21, Session 5
Cargo identification algorithms enabling unmanned unattended inspection at high throughput portals
A. L. Chalmers, American Science and Engineering, Inc. (USA) Software algorithms are described that attempt to characterize specific cargoes found in cargo containers crossing security checkpoints at national borders. Sample processing of simulated and actual cargos is included to highlight the expected operational issues such processing will encounter. A theoretical architecture of processing cabapabilites is presented utilizing existing networks and deployed inspection systems to potentially provide automated and unmanned verification of cargo with its declared manifest.

6736-22, Session 5
Area protection network (APN): a concept for autonomous perimeter surveillance and protection including a demonstrator
P. J. B. Lindquist, Saab Bofors Dynamics AB (Sweden) Area Protection Network (APN) is a concept for autonomous surveillance and perimeter protection. It is a network of sensors, software and a command & control unit. The purpose is to protect military and civilian objects and installations with an all-weather, 24-hour, modular, low-cost and mobile system. The system shall primarily be able to detect and track human activity in the area of interest and also detect and classify behaviour in order to trig appropriate actions. In order to show the concept in real life, a fully operational demonstrator was developed during 2006 for the Ground Combat Days in Skvde, Sweden. The APN demonstrator currently consists of the following components. Two optronic (infrared) sensors: PDES One radar: SIRS77 An image processing and data fusion software: Impress

6736-19, Session 4
An algorithm suite to provide detection, localization, and identification of potential chemical/biological threats resulting from airburst events on the move
S. V. Desai, M. E. Hohil, B. U. Peltzer, A. Morcos, U.S. Army Research, Development and Engineering Command (USA) Integrating a sensor suite with ability to discriminate potential Chemical/Biological (CB) events from high-explosive (HE) events employing a standalone acoustic sensor with a Time Difference of Arrival (TDOA) algorithm we developed a cueing mechanism for more power intensive and range limited sensing techniques. Enabling the event detection algorithm to locate to a blast event using TDOA we then provide further information of the event as either Launch/Impact and if CB/HE. The added information is provided to a range limited chemical sensing system that exploits spectroscopy to determine the contents of the chemical event. The main innovation within this sensor

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6736: Unmanned/Unattended Sensors and Sensor Networks IV


A Command and control unit: WCU One high definition Pan-Tilt-Zoom video camera: Sony EVID70 A network that connects all units The APN demonstrator can currently detect and track multiple humans and/or vehicles in limited 3D. The tracks are presented on a map in the operators graphical user interface. The operator can easily define multiple alarm zones in the map. If any track enters an alarm zone an alarm is triggered. The system also includes blue force tracking. Due to the successful demonstration and the great external interest, the APN team has performed several additional demonstrations in different places in Sweden and abroad.

6736-25, Session 5
Trends in optoelectronic security sensors
M. Szustakowski, W. Ciurapinski, M. Zyczkowski, N. Palka, Wojskowa Akademia Techniczna (Poland) Terrorist threats around the world are accelerating the need for new generation security technologies that combine new physical aspects of sensors with digital signal processors and internet protocol (IP) network. In many case sensors used in security are base an optoelectronic technology as chemical and biological sensors, video surveillance, television, infrared and thermovision sensor, fiber optic sensors, biometric terahertz and radar sensor. Integration of optoelectronics sensor with information technology by introduced inside theirs scheme modern algorithm open the door to applying computer based analytical tools, to create intelligent, proactive security systems capable to recognizing threatening behaviors and patterns and create actionable alerts and alarms to the appropriate responders. In this paper we show trends of technology development different kind of optoelectronics terahertz and radar sensor in their streaming to name of intelligent sensors. The intelligent sensors are now needed to build physical integrated security systems for secure importance society structure as airport, town transport and another. New generation intelligent optoelectronic sensor is playing an increasingly important role in the physical security and video surveillance industry. This in turn is creating new opportunities for optoelectronics and photonics professionals to apply their knowledge to security applications.

6736-23, Session 5
Towards detection of marine vehicles on horizon from buoy camera
S. Fefilatyev, D. Goldgof, L. Langebrake, Univ. of South Florida (USA) This paper presents a new technique for automatic detection of marine vehicles in open sea from a buoy camera system using computer vision approach. Users of such system include border guards, military, port safety and flow management, sanctuary protection personnel. The system is intended to work autonomously, taking images of the surrounding ocean surface and analyzing them on the subject of presence of marine vehicles . The goal of the system is to detect an approximate window around the ship and prepare the small image for transmission and human evaluation. The proposed computer vision-based algorithm combines horizon detection method with edge detection and post-processing. The dataset of 100 images is used to evaluate the performance of proposed technique. We discuss promising results of ship detection and suggest necessary improvements for achieving better performance.

6736-26, Session 5
Consistent detection and identification of individuals in a large camera network
J. M. Orwell, V. Leung, S. Velastin, A. Colombo, Kingston Univ. (United Kingdom) In the wake of an increasing number of terrorist attacks, counterterrorism measures are now a main focus of many research programmes. An important issue for the police is the ability to track individuals and groups reliably through underground stations, and in the case of post-event analysis, to be able to ascertain whether specific individuals have been at the station previously. While there exist many motion detection and tracking algorithms, the reliable deployment of them in a large network is still ongoing research. Specifically, to track individuals through multiple views, on multiple levels and between levels, consistent detection and labelling of individuals is crucial. In view of these issues, we have developed a change detection algorithm to work reliably in the presence of periodic movements, e.g. escalators and scrolling advertisements, as well as a contentbased retrieval technique for identification. The change detection technique automatically extracts periodically varying elements in the scene using Fourier analysis, and constructs a Markov model for the process. Training is performed online, and no manual intervention is required, making this system suitable for deployment in large networks. Experiments on real data shows significant improvement over existing techniques. The content-based retrieval technique uses MPEG-7 descriptors to identify individuals. Given the environment under which the system operates, i.e. at relatively low resolution, this approach is suitable for a short timescales. For longer timescales, other forms of identification such as gait, or if the resolution allows, face recognition, will be required.

6736-24, Session 5
A multi-sensor approach for coastal surveillance
B. van den Broek, B. van den Broek, J. C. van den Heuvel, P. B. W. Schwering, TNO (Netherlands) Maritime borders and coastal zones are susceptible to threats like drug trafficking, piracy, undermining economical activities, etc. The Single Picture Integration for Territorial water Surveillance (SPITS) initiative at TNO Defence, Security and Safety aims at improving situational awareness in a coastal environment by detecting anomalous behaviour of surface and air targets. Within this context we focus here on multi-sensor surveillance of the coastal environment. We present a study on improving classification results for small sea surface targets using an advanced sensor suite and a scenario in which a small ship is approaching the coast. A next generation sensor suite mounted on a tower has been defined consisting of a maritime surveillance and tracking radar system, capable of producing range profiles and ISAR imagery of ships, an advanced infrared camera and a laser range profiler. The radar system detects and tracks ships, while simultaneously information from the sensor suite is extracted and fused. This information is first used to select suspected ships. Next, information for the selected ships is compared with information about ships in a database. Using a priori information, classification probabilities are calculated for the ships on basis of simulated sensor and target data. The classification probabilities are used to evaluate the capabilities for recognizing and identifying non-cooperative ships in coastal waters. We have found that the different sensors give complementary information. Each sensor has its own specific distance range in which it contributes most. A multisensor approach reduces the number of misclassifications and reliable classification results are obtained earlier compared to a single sensor approach.

6736-27, Session 5
3D as a foundation for intelligent underwater autonomous mine countermeasures systems
M. P. Strand, Naval Surface Warfare Ctr. (USA) It is proposed that high resolution three dimensional (3D) imaging sensors be exploited to form the foundation for a systematic approach to intelligent underwater imaging sensor systems. The particular domain of interest is mine countermeasures. This systematic approach consists of a set

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6736: Unmanned/Unattended Sensors and Sensor Networks IV


of inter-related components, each of which contributes significantly to the solution of the overall problem. The first and primary component is to use high resolution 3D as the basis for a robust approach to autonomous object detection, classification, and especially target identification. Using three dimensional images for autonomous object identification (AOI) introduces a number of advantages. The first, and most important, advantage is that by using 3D imagery it is much easier to deal with the viewing aspect dependence of the problem. A second key advantage is that 3D images have a much weaker dependence on the local environment than 2D images have. Both of these advantages mean that autonomous object identification based upon 3D is likely to be markedly more accurate and more robust than AOI based upon 2D imagery. Other key components of the systematic approach to intelligent underwater mine countermeasures systems based upon 3D include exploiting 3D for (1) intelligent bottom characterization (which impacts, e.g., mine burial probabilities and characterization of local clutter), (2) determination of intelligent search patterns (which may depend, for example, on local geographical patterns), (3) the basis for intelligent navigation (singularly or in groups), and finally (4) intelligent target reacquisition. (Romania); O. D. Iancu, I. Cristea, N. Grosu, Univ. Politehnica Bucuresti (Romania); C. N. Toader, M. S. Mihalcea, NBC Defense and Ecology (Romania) The paper reports the experimental results for polymer-coated surface acoustic wave (SAW) sensors that were exposed to chloropicrin (CCL3NO2) and hydrogen cyanide (HCN). The two chemical agents were chosen due to their different vapor pressure. The sensors are delay line type, made on quartz piezoelectric substrates. Quartz (ST-cut, propagation axis X) was chosen for its temperature stability. Two different polymers were used in our investigations. The air brushing technique was selected to deposit the sensitive film onto SAW devices. The nominal frequency was 70 MHz. The sensor stability was studied, and 2 Hz noise limit was determined. Sensor sensivity for chloropicrin (CCL3NO2) and hydrogen cyanide (HCN) was compared with other common gases such as (Kr, He, Ar, N2). The minimum detectable level (MDL) was evaluated for 10 Hz oscillation frequency deviation. MDL assesments indicated 0.08-0.13 ppm for chloropicrin, and 0.08-0.18 ppm for hydrogen cyanide, while for Kr, He, Ar, N2 it was in the range of 33-230 ppm.

6736-31, Session 6
Zero False Alarm Seismic Detection and Identification Systems
T. Goldburt, A. Pakhomov, General Sensing Systems LLC (USA) General Sensing Systems (GSS) has achieved outstanding and verifiable results in the design and development of various seismic detection and identification systems. These results include a new seismic miniature sensor design as well as seismic signal recording and research for many traditional and nontraditional targets - walking, running and jumping persons, heavy and light vehicles, helicopters and aircraft, ships, trains, etc. These results also include the hardware design for up-todate unattended seismic detection and identification systems. The main outcome of our effort is detection and identification algorithms and corresponding software for personnel and vehicle detection and identification which were tested in real environment conditions. These algorithms provide a zero false alarm rate with no target missing and can be used for many real and important military and homeland security applications. We will report on future seismic detection and identification systems for various military and civil applications.

6736-28, Session 6
A methodology for analyzing an acoustic scene in sensor arrays
M. E. Hohil, S. Quoraishee, U.S. Army Research, Development and Engineering Command (USA); H. Man, Stevens Institute of Technology (USA); S. V. Desai, U.S. Army Research, Development and Engineering Command (USA) Acoustic sensor arrays have been frequently deployed in surveillance and monitoring operations. Acoustic scene analysis in such sensor array environment is to achieve high level situation awareness through event specification, identification, association and anomaly detection. Introduced herein is a new acoustic scene analysis method which comprises of two statistical machine learning components. The first component is sensor signal modeling and correlation. Acoustic signals from all sensors are collected and partitioned into short time segments. Utilizing Hidden Markov Models (HMMs) to describe the signal segments inter and intra sensor signal correlation is extracted based on a Generalized Probability Product Kernel (GPPK), which provides reliable distance measures between HMMs. A fast GPPK algorithm is used to expedite the calculation. The second component is event association and anomaly detection. Employing association rule based data mining method to capture the pattern of occurrences of correlated data readings. Observation vectors for the data mining algorithm consist of GPPK scores and the current most likely HMM states of all sensor signals. Frequent episodes and association rules are used to create normal profiles of acoustic environment. Anomaly is declared when the new rules extracted from a test observation vector violate the normal profile rule set. Two datasets are used in this study, one is collected from an underwater hydrophone array for diver detection, and the other is from a ground microphone array for pedestrian detection. The proposed method not only can reliably detect the intended targets, but also provide a framework for high level data integration and situation analysis, with abstract events and profiles instead of sensor readings. The full description of this method and the complete experimental results will be presented in the final paper.

6736-32, Session 6
Changing Requirements and Solutions for Unattended Ground Sensors
G. Prado, SenTech Inc. (USA) Unattended Ground Sensors (UGS) were first used to monitor VietCong activity along the Ho Chi Minh Trail in the 1960s. In the 1980s, significant improvement in the capabilities of UGS became possible with the development of digital signal processors. This led to their use as fire control devices for smart munitions (for example: the Wide Area Mine) and later to monitor the movements of mobile missile launchers. In these applications, the targets of interest were large military vehicles with strong acoustic, seismic and magnetic signatures. Currently, the requirements imposed by new terrorist threats and illegal border crossings have changed the emphasis to the monitoring of light vehicles and foot traffic. These new requirements have changed the way UGS are used. To improve performance against targets with lower emissions, sensors are used in multi-modal arrangements. Non-imaging sensors (acoustic, seismic, magnetic and passive infrared) are now being used as activity sensors that cue imagers. The availability of better imaging technology has made imagers the preferred source of actionable intelligence. The imagers are equipped with sophisticated image processing capabilities (image enhancement, moving target detection and tracking, image compression). Infrared cameras are now available with uncooled detector arrays and better resolution at a lower cost. Visible light imagers are more sensitive extending their utility well into the night. Various commercial satellite services now provide relatively inexpensive long-range communications and the Internet provides fast worldwide access to the data.

6736-29, Session 6
Novel optical sensor system for ocean monitoring
J. Butman, Coherent, Inc. (USA); E. M. Carapezza, B. Javidi, Univ. of Connecticut (USA) No abstract available

6736-30, Session 6
Surface acoustic wave sensors for detection of ccl3no2 and HCN
C. Viespe, C. Grigoriu, D. Dragulinescu, C. Blanaru, C. Sima, National Institute for Lasers, Plasma and Radiation Physics

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6736: Unmanned/Unattended Sensors and Sensor Networks IV

6736-33, Session 6
Nanomechanical chemical sensors based on functionalized MEMS arrays
S. Rajic, Oak Ridge National Lab. (USA) No abstract available

6736-50, Session 6
Wide-angle laser rangefinder
P. D. Yankov, Univ. of Sofia (Bulgaria) No abstract available

the method applied to the reconstruction of digitally holograms recorded at CO2 laser wavelength of 10.6 micron are reported. It is show that good reconstructions can be obtained.even with the lower spatial resolution of IR recording detectors compared to visible CCD array. The results show that new prospective can be exploited by using high power CoO2 laser sources in optical metrological applications

6736-37, Session 7
Integral imaging with increased depth of field by using phase masks
A. Castro, Instituto Nacional de Astrofsica, ptica y Electrnica (Mexico); Y. Frauel, Univ. Nacional Autnoma de Mxico (Mexico); B. Javidi, Univ. of Connecticut (USA) Integral imaging systems permit the capture of threedimensional scenes. However they usually suffer from a limited depth of field that severely reduces the depth range that can be used in practice. In order for such a system to be able to capture large-depth scenes without any adjustment, we propose to include in the integral imaging system an array of phase masks in order to increase the depth of field of the images. We consider both the pickup and the reconstruction stages.

6736-34, Session 7
Local track repair for video tracking on small UAVs
S. P. DelMarco, M. Antone, A. Reiter, BAE Systems Advanced Information Technologies (USA); T. Jenkins, Air Force Research Lab. (USA) Persistent aerial video surveillance from small UAV (SUAV) platforms requires accurate and robust target tracking capabilities. However, video tracks can break due to excessive camera motion, target resolution, low signal-to noise ratio, video frame dropout, and frame-to-frame registration errors. Connecting broken tracks (video track repair) is thus essential for maintaining high quality target tracks. In this paper we present an approach to track repair based on multi-hypothesis sequential probability ratio tests (MHSPRT) that is suitable for real-time video tracking applications. To reduce computational complexity, the approach uses a target dynamics model whose state estimation covariance matrix has an analytic eigendecomposition. Chi-square gating is used to form feasible track-to-track associations, and a set of local hypothesis tests is defined for associating new tracks with coasted tracks. Evidence is accumulated across video frames by propagating posterior probabilities associated with each track repair hypothesis in the MHSPRT framework. Global maximum likelihood and maximum a posteriori estimation techniques resolve conflicts between local track association hypotheses. The approach also supports fusion of appearance-based features to augment statistical distributions of the track state and enhance performance during periods of kinematic ambiguity. First, an overview of the video tracker technology is presented. Next the track repair algorithm is described. Finally, numerical results are reported demonstrating performance on real video data acquired from an SUAV.

6736-38, Session 7
Super-resolution enhancement of flash ladar range data
G. Rosenbush, Univ. of Maryland/College Park (USA); T. Hong, National Institute of Standards and Technology (USA); R. D. Eastman, Loyola College in Maryland (USA) Flash LADAR systems are becoming increasingly popular for robotics applications. However, they generally provide a lowresolution range image because of the limited number of pixels available on the focal plane array. In this paper, the application of image super-resolution algorithms to improve the resolution of range data is examined. Super-resolution algorithms are compared for their use on range data and the frequency-domain method is selected. Four low- resolution range images which are slightly shifted and rotated from the reference image are registered using Fourier transform properties and the superresolution image is built using non-uniform interpolation. Image super-resolution algorithms are typically rated subjectively based on the perceived visual quality of their results. In this work, quantitative methods for evaluating the performance of these algorithms on range data are developed. Edge detection in the range data is used as a benchmark of the data improvement provided by super-resolution. The results show that super-resolution of range data provides the same advantage as image super-resolution, namely increased image fidelity.

6736-35, Session 7
Increasing the depth of field of imaging systems with numerically optimized phase masks
Y. Frauel, Univ. Nacional Autnoma de Mxico (Mexico); A. Castro, Instituto Nacional de Astrofsica, ptica y Electrnica (Mexico) Nowadays, for surveillance and security applications there exist imaging systems with a 360-degree field of view. It would also be highly desirable to have imaging systems with a large depth of field in order to obtain sharp images over a large depth interval along the optical axis. The depth of field can be increased by stopping down the pupil aperture of the optical system but this solution results in a loss of light gathering power. Another option that avoids this problem consists of placing a specially designed phase mask in the pupil plane of the system. In this work, we propose to numerically optimize phase masks in order to increase the depth of field of imaging systems.

6736-39, Session 8
Data harvesting using optical wireless communication
D. Kedar, S. Arnon, Ben-Gurion Univ. of the Negev (Israel) Networks of sensors are an emerging technology for real-time data gathering in a wide variety of civilian and military applications. Sensor networks comprise a large number of miniature nodes with sensing, computing and wireless communication capabilities and are randomly deployed in an area or volume of interest that may be remote or hazardous especially in defense applications. Using optical wireless communication (OWC) in the form of an illuminating laser beam a population of sensor nodes can be located and mapped with no power expenditure on the part of the nodes and data can be encoded onto the retroreflected beam of light by the sensor node at a minimal energy cost. In this paper we review some theoretical and experimental work in this area and underline some of the challenges and possible solutions.

6736-36, Session 7
Imaging and phase measurement of 3D objects at 10.6 microns by digital
P. Ferraro, S. Grilli, L. Miccio, P. Buahbassuah, R. Meucci, S. De Nicola, Istituto Nazionale di Ottica Applicata (Italy) Digital holography in the mid infrared range is shown to be a feasible technique for optical motrological applications. The technique allows to reconstruct both amplitude and phase of wavefronts scattered by a 3D object. Experimental results of

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6736: Unmanned/Unattended Sensors and Sensor Networks IV

6736-40, Session 8
Ground to survey aerostatic platform bidirectional free space optical link
F. J. Lpez Hernndez, M. A. Geday, A. Carrasco-Casado, G. del Campo, Univ. Politcnica de Madrid (Spain); P. Munuera, Ingenieria y Servicios Aeroespaciales (Spain) The mass and the power consumption are the most constrictive restrictions in an Aerostatic Platform. We present in this work a bidirectional link based on a liquid crystal retro modulator (LCRM). The use of this device displaces to the terrestrial station the tasks of powering the laser, and the pointing mechanism. Also, due to the non conductive characteristic of the LCRM, the power requirements are low. The beam aperture of the upward link can be tailored with the compromise between the aperture losses, and the tracking mechanics. The range of the distances for the system presented is from 2 to 20 km, being the lower limit for the tests and checking of the system, and the higher the final system for a stratospheric survey Aerostat. The system will be modified to fulfill the requirements associated to the increasing distances, although the main characteristics (modulation schemes, modulator, etc.) will be maintained. The system uses two modulation schemes for each direction: in the uplink a carrier constant envelope FSK modulation is used. For the down link, part of the optical signal is polarization modulated and reflected to the earth. For increasing the data rate, a multi polarization modulation is used. Actually, 32 polarization levels are used, increasing the data rate in a 5 fold factor. The use of differential scheme, i.e. the State of Polarization (SOP) change defines the data, not the SOP, avoids the necessity of SOP maintenance in the transmitter.

thermal cycling. The laser radiation of two diode laser bars consisting of six emitters is coupled in a 100 m NA=0.22 fiber. The coupling efficiency as well as the mounting accuracy of the optical components have been analyzed by raytracing. To predict the effect of different heat sink materials and to calculate the temperature drop over the module, a 3D finite element analysis has been performed. The rated optical output power of the module described in this paper amounts to 20 W, and it measures 50 mm x 50 mm x 20 mm.

6736-43, Session 8
A high-speed modulated retro-reflector communication link with a transmissive modulator in a cats eye optics arrangement
J. hgren, F. Kullander, L. J. Sjqvist, Swedish Defence Research Agency (Sweden); K. Wang, Q. Wang, S. Junique, S. Almqvist, B. Noharet, Acreo AB (Sweden) An optical communication link with a modulated retro-reflector in a cats eye optics arrangement has been evaluated. A transmissive multiple quantum well modulator with a modulation bandwidth in excess of 10 MHz was placed in a wide field-ofview cats eye optic system. A laser transceiver with a wavelength around 1550 nm was used to interrogate the modulated retro-reflector and detect the reflected and modulated signal. The bit error rate was evaluated for different experimental configurations and the signal-to-noise ratio was measured as a function of range and angle of incidence.

6736-44, Session 8
Phase compensation considerations on coherent, free-space laser communications system
A. M. Belmonte, A. Rodrguez, F. Dios, A. Comern, Univ. Politcnica de Catalua (Spain) The same turbulence-induced optical phase perturbations that limit astronomical seeing destroy the spatial coherence of a laser beam as it propagates through the atmosphere. It restricts the received coherent power levels in atmospheric optical coherent systems and degrades their overall performance by exhibiting severe temporal fading associate with the turbulenceinduced optical phase and amplitude fluctuations. Consequently, it increases the error in any coherent free-space communication link. Certainly, we need to quantify the performance limitations imposed by atmospheric turbulence on specific communication and to develop system configurations that are immune to atmospheric fluctuations. In this work, it is our intention to elucidate how the addition of phase compensation systems to the transmitter or receiver can reduce the effects of atmospheric propagation and, in so doing, to quantify the improvement on the performance of optical communications systems regarding coherent detection. Phasecompensated optics offers the potential for overcoming these limitations by adaptive tracking of the beam and correction of atmospherically-induced aberrations.

6736-41, Session 8
High-speed information systems
A. R. Pirich, C. B. Pirich, ACP Consulting (USA) It seems like yesterday that 10Gb/s (OC-192) transmission was the norm in data rates. The particular bit-rate jump to 40Gb/s (OC-768)...or higher...is not a trivial task! Electro-optic (electronic and photonic) designers now face serious complexities as they push the technology to the limit. Issues such as chromatic dispersion and polarization mode dispersion have placed more stringent requirements on fiber optics and associated components. The requirement to gather and transfer data at faster and faster rates has spurned an evolution in the thought process of electro-optic engineers. It appears that to handle the ever increasing bandwidths, electro-optic(photonic) yechniques are the wat to go! Though optical fiber exhibits low-loss and high bandwidth, bit-rate increases in communication networks become increasingly sensitive to polarization impairments. Only time will tell if electro-optic designs (along with photonics) can handle the voluminous amounts of data generated from the now present information sources.

6736-42, Session 8
Spaceborne fiber coupled diode laser pump modules for intersatellite communications
M. Traub, H. Plum, H. Hoffmann, Fraunhofer-Institut fr Lasertechnik (Germany); T. Schwander, Tesat-Spacecom GmbH & Co. KG (Germany) Compared to traditional microwave systems for intersatellite communication, optical intersatellite links offer numerous advantages like lower weight and power consumption as well as higher bit rates. In addition, the smaller diameter of the laser beam allows a higher data security. The optical terminal developed by TESAT Spacecom offers a bandwidth of 5.5 Gbps with a range of 10,000 km and a bit error ratio below 1e-9. In 2007, the coherent optical terminal will be verified in orbit. The transmitter consists of a ring oscillator and an optical amplifier. For both components, a high reliable diode laser pump module is essential to meet the demanded lifetime requirements of 10 years with a reliability of 0.9998. Built-in cold and hot redundancy as well as a truly hermetically sealed housing increase the reliability of the pump module. To proof its suitability for space applications, the module has been successfully qualified in terms of radiation, vibration, mechanical shock and

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6737: Electro-Optical and Infrared Systems: Technology and Applications


Wednesday-Thursday 19-20 September 2007 Part of Proceedings of SPIE Vol. 6737 Electro-Optical and Infrared Systems: Technology and Applications IV

6737-44, Poster Session


Infrared device for defence based on polycrystalline silicon
D. Milovzorov, Fluens Technology Group Ltd. (Russia) We proposing the infrared (IR) device based on polycrystalline silicon layers what differ in sizes of crystals. Such crystals incorporate into different spatial structures: multilayer structure with several layers of silicon crystals and conglomerates of large microcrystals surrounded by very small nanocrystalline layer. The differentiation in spatial structure results in different electrical signal propagation and photon detection. It can be applicable for sensor and microscale spectroscopic devices design. Hierarchical structures of grown thin silicon film with small and large nanocrystals we can create new photon detector with redistribution of electrical signal according to applied potentials to various silicon layers. The ratio surface/volume for small nanocrystals is high, but the surface area is small, but for large crystals is opposite situation: the small value of ratio surface/volume and large area of surface. Because, there are many small silicon nanocrystalls are bonded with one large silicon crystal. One such node of polycrystalline silicon film can be used for nanoscale device making. Such kind of device is combined as photon detection by nanocrystals and electrical signal distribution by single structural node according to the famous logical rules.

Bad pixels are spatial or temporal noise which arise from dead pixels by fixed signal level and blinking pixels by variable signal level that go beyond the bounds of normal pixel level at the temperature. Because bad pixels are the false targets over infrared imaging system for tracking, those must be replaced. Main contribution to the number of bad pixels is fixed pattern noise(FPN) according to increasing array size. And it is more simple to establish whether FPN is or not from accumulated frame. But it is needed to calculate with complex implementation such standard deviation from frame to frame in case of temporal noise. Both cases it is very important to establish the threshold level for identifying at variable operating temperature. In this paper,hardware is implemented to identify and replace bad pixels in dominance of the temporal noise. First, threshold levels of FPN are selected through three temperature sentions experimentally. Second, threshold level of temporal noise are selected at current operting environment adaptively. Bad pixels from both are classified into clusters and replaced with DSP and FPGA.

6737-01, Session 1
Ranging and three-dimensional imaging using time-correlated single-photon counting
G. S. Buller, A. M. Wallace, A. McCarthy, Heriot-Watt Univ. (United Kingdom); R. Lamb, SELEX Sensors and Airborne Systems Ltd. (United Kingdom) Time-correlated single-photon counting techniques have been applied to time-of-flight ranging and imaging. This paper will describes recent progress in photon-counting systems performing surface mapping of non-cooperative targets. This includes systems designed for short ranges of the order of 150 metres, as well as measurements on distributed targets at longer ranges of the order of 100 metres up to ten kilometres. We describe the measurement approach, techniques used for scanning, as well as the signal analysis methodology and algorithm selection. The technique is fundamentally flexible: the trade-off between the integrated number of counts (or acquisition time) against range repeatability or depth resolution allows its application in a number of diverse fields. The inherent time gating of the technique, allied to the spatial filtering provided by small active area single-photon detectors, can lead to operation under high ambient light conditions even with low average optical power pulsed sources. We have demonstrated three-dimensional imaging of metredimensioned objects where reverse engineering methods using cooperative targets cannot be routinely employed: e.g. mechanically delicate objects, or objects with more than one reflective surface. Using more advanced signal processing algorithms, we have been able to improve the system performance markedly, as measured by the depth resolution at short and long ranges. Furthermore, the application of these methodologies has allowed us to characterise the positions and amplitudes of multiple returns. Hence, the approach can be used for characterisation of distributed non-cooperative targets at kilometre ranges, even in environments where lowlight level and and/or eye-safe operation is necessary.

6737-45, Poster Session


Optical nonlinear switches based on nanocrystalline silicon
D. Milovzorov, Fluens Technology Group Ltd. (Russia) SHG spectra from silicon films with different average size of nanocrystalls was studied as possible material for active channel by nonlinear optical switches design. It is seen the spectral peak with energy 3.26 eV is related to defects appeared in interface area silicon-silicon dioxide. For films with small silicon crystals (less than 20 nm) the nonlinear optical response contains two spectral peaks. The second peak is caused by optical response from nanocrystal grain boundary that contain oxygen atoms incorporated in silicon as dipoles inside film. The optical nonlinear switch device based on the nonlinear optical response of SiOx media inside film was proposed. Also, the silicon film with quartz micro-clusters were investigated as material for making the nonlinear optical transmitter device. The Raman spectra of films were, also, studied to observe the various silicon and silicon dioxide fractions. The efficiency of transmission of radiation is sufficient.

6737-46, Poster Session


Securing information using multi-lock singlestep digital holography
C. Chang, Ming Dao Univ. (Taiwan) This work demonstrates a fully phase encryption approach that adopts modified single exposure scheme digital holography. The input image to be encrypted is hidden by random phase encoded in secrecy. An encryption hologram is recorded digitally in a single step by interference with a reference wave from a random phase modulation. The decrypted keys are also recorded as the digital holograms, called the key holograms. Multiplication of the stored encryption hologram is performed by the numerical reconstruction of the decrypted keys, and convolution of the result with the impulse response function for decrypting the encrypted information. This approach rapidly achieves the decrypted procedure once by using a personal computer. Results of this study demonstrate that the image quality of the decrypted image is satisfactory.

6737-02, Session 1
IR-dual-band-camera demonstrator: experimental assessment, practical applications
U. Adomeit, R. R. Ebert, Forschungsfesellschaft fr Angewandte Naturwissenschaften e.V. (Germany) The IR-Dual-Band-Camera demonstrator collects simultaneously infrared data in the 3-5 m (mid-wave infrared, MWIR) and 8-12 m (long-wave infrared, LWIR) atmospheric windows. The demonstrator is based on a two-layer QWIP focal plane array with 384 x 288 x 2 detector elements. Images are typically acquired with a frame rate of 100 Hz at 6.8 ms integration time and are stored as 14-bit digital data. Two different dual-band optics were designed and developed. First

6737-47, Poster Session


Implementation for temporal noise identification using adaptive threshold of infrared imaging system
I. Lim, Samsung Thales, Ltd. (South Korea)

10

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6737: Electro-Optical and Infrared Systems: Technology and Applications


a 86 mm and 390 mm focal length, F# 2 dual field of view optics based on refractive and reflective components and second a pure refractive 100 mm focal length, F# 1.5 optics. We present the performance of this IR-Dual-Band-Camera and demonstrate fusion techniques to the pixel-registered dual-band images which show in laboratory tests and field trails promising results with respect to image improvement. The development of a compact and high performance MWIR step zoom camera based on the 640x480 staring focal plane array (FPA) is described. The camera has a 20 magnification step zoom ranging between 24x20 for the wide field of view up to 1.2 x 1 for the narrow field of view and an aperture of F#4. The processing electronics is based on a flexible and expandable architecture. Special emphasis is spent on the solutions adopted for the design of this high zoom ratio and fast optics FLIR and on the electronic architecture and algorithms for image processing. An overview of the performances is given.

6737-03, Session 1
New 3-5 micron wavelength range hyperspectral imager for ground and airborne use based on a single-element interferometer
D. Cabib, A. Gil, R. A. Buckwald, CI Systems (Israel) Ltd. (Israel) The mid wavelength range (3-5 micron) hyperspectral imagers, based on different types of gratings for the collection of the spectral information need cooling of the optical elements in order to avoid background signal which significantly limits the dynamic range of the instrument. In addition, similar commercial instruments based on interferometers are not suitable for pushbroom airborne use. With the present work we present a solution to both requirements. The present interferometric configuration combined with proper 2D array detector cold shielding avoids the need for cooling the optics, and at the same time insures the possibility of airborne pushbroom operation. The latter is achieved with a single element interferometer producing an optical path difference (OPD) between the two arms, which is a function of the angular position of a pixel in the field of view along the scan direction. As the field of view is scanned each pixel of the image is measured through all the OPDs from zero to OPDmax, so that its interferogram can be stored and Fourier transformed to yield its spectrum. This design, combined with properly synchronized optical head rotation is suitable also for use on the ground. In this paper the instrument design and its advantages are described.

6737-06, Session 1
Electro-optics technology for a new generation of military and law enforcement small equipment
C. Giunti, A. Cocchi, R. Bardazzi, L. Calamai, E. Torniai, M. Maestrini, M. Livi, M. Sabatini, N. Santini, C. Toccafondi, Galileo Avionica S.p.A. (Italy) In the framework of the SOLDATO FUTURO program, supported by Italian Army, Galileo Avionica (a Finmeccanica company) has developed a family of small equipments based on suites of elctro-optics sensors. These modules, designed and built by GA, ranges, from uncooled V0x 25 nicron thermal imagers, small and very compact laser rangefinders, CMOS Visible sensors up to visual units based on the last generation of color OLED microdisplays. All the EO assemblies are integrated to form a very small and lightweight Integrated Weapon Sight, a Multi Function Target Locator, and a Fire Control System. Even if the equipments have been developed for military applications many other applications such as law enforcements or surveillance can be envisaged.

6737-07, Session 1
Performance of compact intensified camera unit (ICU) with autogating based on video signal
A. W. de Groot, P. Linotte, D. van Veen, M. de Witte, Photonis-DEP B.V. (Netherlands); N. Laurent, Photonis S.A.S. (France); J. van Spijker, Photonis-DEP B.V. (Netherlands) High quality night vision digital video is currently required for many observation, surveillance and targeting applications, including several of the current soldier modernization programs. We present the performance increase that is obtained when combining a state-of-the-art image intensifier with a low power consumption CMOS image sensor. Using an integrated FPGA for the histogram analysis of the video signal, the autogating and gain of the image intensifier are optimized for best SNR. Some critical issues regarding the optional interfacing with a separate laser in the application for range gated imaging are discussed.

6737-04, Session 1
Near to mid-infrared, broadly tunable, active hyperspectral imaging system for the detection of gaseous and liquid species
D. J. M. Stothard, C. F. Rae, M. Ross, M. H. Dunn, Univ. of St. Andrews (United Kingdom) We describe a compact mid-infrared active spectroscopic imaging system for the rapid, stand-off detection ofhydrocarbon vapour and deposited organic liquids, developed for the UK Ministry of Defence. Based upon the back-scatter absorption gas imaging (BAGI) technique, the system utilises a miniaturised, extremely efficient all-solid-state intracavity optical parametric oscillator (OPO) as the imaging illumination source. The OPO produces up to 250mW of tunable down-converted radiation over the range 1.3 - 4.5um, for a diode pump power of only 3W. Due to the nature of the nonlinear crystal employed within the OPO, the system can be tuned across its spectral range in ~1 second. We obviate the very high cost and complexity of a cooled MCT or InSb video array by raster-scanning the collimated illumination beam over the area of interest and reconstructing the image by sampling sequentially the backscattered radiation at each pixel point with a single element, thermo-electrically cooled MCT photo-detector. Video-like frame rates of 10 f.p.s. have been demonstrated via this technique. The range limit of the detector is currently <10 meters, but in the near future we hope to significantly improve this by implementing a superior detector exhibiting lower noise, the results of which will be presented. We will demonstrate how the system has been used to detect, in real time, leaks of multispecies hydrocarbon gases and the presence of, and discrimination between, selected liquid-phase organic chemicals.

6737-08, Session 1
Multi-aperture imaging device for airborne platforms
L. C. Laycock, V. A. Handerek, BAE Systems plc (United Kingdom) Imaging devices are very attractive as sensors in small airborne platforms and there is a continuing trend toward widespread employment of imaging either alone or in combination with complementary technologies. In the civil domain, modern silicon CCD and CMOS image sensors are becoming extremely small, so that the package size of commercial miniature cameras is becoming dominated by the image forming optics, even if the latter is only a structure supporting a pinhole. Recently, there have been demonstrations of ultra-flat, extremely light weight cameras working in the visible region of the spectrum. Similar ideas for cameras developed to operate in the infra-red could help to drastically reduce the size, weight and cooling requirements of imaging, also offering substantial cost reductions. In addition, designs providing wide field-of-view can potentially eliminate the need for sightline steering hardware. This paper describes work on a biologically inspired imaging system offering a wide field of view, thanks to the use of a multi-aperture sensor based on micro-optics, which can be used to observe simultaneously in different directions. Results from a near-infrared, narrowband demonstrator will be reported.

6737-05, Session 1
ERICA PLUS: compact MWIR camera with 20x step zoom optics and advanced processing
A. Porta, P. Lavacchini, M. Olivieri, D. Torrini, Galileo Avionica S.p.A. (Italy)

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

11

Conference 6737: Electro-Optical and Infrared Systems: Technology and Applications

6737-09, Session 1
Hadamard camera for 3D imaging
E. Romasew, H. D. Tholl, J. Barenz, Diehl BGT Defence GmbH & Co. KG (Germany) One development branch of future 2D and 3D digital imaging will lead us from classical imaging to computational imaging. In classical 2D digital imaging spatial information is registered electronically by image plane sensors (CCD or CMOS arrays) which imitate conventional analog films. A third dimension can be captured using illumination techniques like gated viewing. This classical approach combines spatial data sampled in pixels and arranged in an image format with digital signal processing to extract spatial information. Computational imaging, on the other hand, feeds pixelated data into a digital signal processor in a format subjected to processing or hardware constraints. The result of the computation is either a classical image or any other form of spatial information. Computational imaging is the right approach to construct large format 3D images with a small number of detector elements or to capture images in spectral ranges for which electronical focal plane arrays do not exits, as for example T-ray imaging. A new method for fast 3D image generation employs a single detector in combination with a Hadamard multiplexing technique, a high resolution 2D sampling micro mirror device (DMD) and a time of flight measurement method (TOF). In a conventional scanning process, the micro-mirror of the DMD can be tilted consecutively towards the scene and a TOF measurement is performed for each pixel separately. The combination of scene sampling and TOF measurements provides angle-angle-range information which allows the computation of 3D images. The signal-to-noise ratio of the of capturing 3D images can be improved using Hadamard encoding techniques. Several 3D scene elements (voxels) are captured at once during a single scanning event. The DMD provides the spatial pixel structure while the TOF is digitally sampled at high rates. After N frames of the DMD (where N is number of 2D pixel elements) a data cube is constructed. A Hadamard algorithm is then employed in order to extract 2D images for each time slice of the data cube. In this way, an angle-angle-range 3D image cube with improved SNR is created. The higher SNR gained with Hadamard camera can be used for range enhancement or optimization of scanning velocity. In the paper we describe in details the work that has been carried out for fusing a commercial micro mirror sampling element with TOF acquisition methods and known Hadamard multiplexing techniques for implementation of fast and SNR optimized 3D image capture. The theoretical basics of TOF and Hadamard technique are presented and will be complemented by theoretical explanation of utilizing them for 3D volumetric image generation. Finally measurement results of scene image acquisition are going to be demonstrated and discussed as well as expanded by considerations about possible applications as THz-Imaging and the following research steps.

will be located in the vicinity of each jet engine and the FMmodulated noise energy will introduce noise spikes in the missile receiver bandwidth. This will deny detection and tracking of the jet plume to the operator within the time allocated to him, which is typically 5 to 15 seconds. This system will be turned ON during the takeoff and landing operations. Once the aircraft reaches an altitude over 15,000, the pilot can turn OFF the system. Computer simulations on a commercial jet engine plume indicate that that the radiation intensity from a jet plume will be significantly less (25 to 30 %) during the roll and pitch maneuvers which are common during take-off operations. Furthermore, only 16.5 % of the total IR energy will be in the 23 micron spectral region. This and the atmospheric attenuation will reduce further the IR energy to less than 3 % at receiver, which can be easily jammed with high J/S ratio.

6737-48, Session 1
Focus-free NVG development
G. Bennett, Georgia Institute of Technology (USA) No abstract available

6737-11, Session 2
Coded aperture systems as nonconventional, lensless imagers for the visible and infrared
C. W. Slinger, N. Gordon, M. McNie, D. Payne, K. Ridley, M. Strens, G. De Villiers, R. A. Wilson, QinetiQ Ltd. (United Kingdom) Coded aperture imaging (CAI) has been used extensively at gamma- and X-ray wavelengths, where conventional refractive and reflective techniques are impractical. CAI works by coding optical wavefronts from a scene using a patterned aperture, detecting the resulting intensity distribution, then using inverse digital signal processing to reconstruct an image. This paper will consider application of CAI to the visible and IR bands. Doing so has a number of potential advantages over existing imaging approaches at these longer wavelengths, including low mass, low volume and various agile/steerable imaging modes. However, diffraction and photon flux reductions can have adverse consequences on the image quality achievable. An analysis of these benefits and limitations, along with preliminary experimental results, will be presented.

6737-12, Session 2
Low-cost wavefront coding using coma and a denoising-based deconvolution
C. Dorronsoro, Imatrics Image Technologies (Spain); J. A. Guerrero-Colon, Imatrics Image Technologies (Spain) and Univ. de Granada (Spain); M. de la Fuente, J. M. Infante, Indra Sistemas (Spain); J. Portilla, Instituto de Optica (Spain) and Consejo Superior de Investigaciones Cientificas (Spain) Wavefront coding (WFC) is a powerful hybrid optical-numerical technique for increasing the depth of focus of imaging systems. It is based on two components: (1) an optical phase element that codifies the wavefront, and (2) a numerical deconvolution algorithm that reconstructs the image. Traditionally, some sophisticated optical WFC designs have been used to obtain approximate focus-invariant point spread functions (PSFs). Instead, we present a simple and low cost solution, implemented on infrared (IR) cameras, which uses a decentred lens inducing coma as an adjustable and removable phase element. We have used an advanced deconvolution algorithm for the image reconstruction, which is very robust against high noise levels. These features allow its application to low cost imaging systems. We show encouraging preliminary results based on realistic simulations using optical PSFs and noise power spectral density (PSD) laboratory models of two IR imaging systems. Without induced optical phase, the reconstruction algorithm improves the image quality in all cases, but it performs poorly when there are both in and out-of-focus objects in the scene. When using our coding/decoding scheme with low-noise detectors, the proposed solution provides high quality and robust recovery even for severe defocus. As sensor noise increases, the image suffers a graceful degradation, its quality being still acceptable

6737-10, Session 1
IR system to provide effective IR countermeasure (IRCM) capability to ward off threats posed by shoulder-fired missiles (SFMs)
A. R. Jha, JHA Technical Consulting Services (USA) This paper presents a unique IR sensor technology capable of providing integrated deception and noise jamming IRCM capability to ward off threats from SFM or STINGER missiles operated by terrorist groups. Latest survey indicates more than 60,000 such missiles currently in existence. Even one such missile can bring down a commercial jet carrying more than 350 passengers. The proposed IRCM system deploys innovative jamming techniques to confuse the IR seeker receiver by introducing sharp FM-modulated noise spikes in the IR receiver bandwidth, thereby preventing the detection, tracking and lock-on of jet engine plume. The proposed IR equipment will provide instant IRCM capability to eliminate the SFM threat posed to passenger jets, military aircraft and helicopters. The system uses low-cost, compact, ZENON short-arc lamps each rated for 300-watt IR power enough to provide a high J/S over the IR spectral range of interest (2-3 micron). These IR sources

12

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6737: Electro-Optical and Infrared Systems: Technology and Applications


even when using highly noisy sensors, such as microbolometers. We have experienced that the amount of induced coma is a key design parameter: while it only slightly affects the in-focus image quality, it is determinant for the final depth of focus range.

6737-16, Session 2
Properties of light reflected from road signs in active imaging for driving safety
A. Halstuch, Y. Yitzhaky, Ben-Gurion Univ. of the Negev (Israel) Night-vision systems in vehicles are a new emerging technology. A crucial problem in active (laser-based) systems is distortion of images by saturation and blooming, due to strong retroreflections of the light from road signs. We measured the properties of the reflected light from the three road sign types commonly used with respect to the angle between the laser source and the reflecting plane. We found that different types of signs have significantly different reflection properties. It is concluded from our measurements that the optimal solution for attenuating the retro-reflecting intensity performs well for two of the road sign types. Unfortunately, the performance of this solution is less efficient for the third road sign type.

6737-14, Session 2
Wideband protection filter: single filter for laser damage preventing at wide wavelength range
A. Donval, B. A. Nemet, M. Oron, R. Oron, R. Shvartzer, KiloLambda Technologies, Ltd. (Israel); L. Singer, IMODSIBAT (Israel); C. Reshef, Ministry of Defence IDF (Israel); B. Eberle, H. Brsing, R. R. Ebert, Forschungsfesellschaft fr Angewandte Naturwissenschaften e.V. (Germany) High power laser radiation when impinging on imaging or detection systems can seriously interrupt the signal. Interruption may start from transient saturation and can lead to permanent damage. The problem exists in imaging sensors comprising of CCDs and other matrix detectors, human eyes or other imaging and non-imaging sensors. Protection from high power laser radiation is generally performed using fixed absorptive spectral filters used to block specific laser wavelengths. The major drawbacks of this solution are that the color impression may be affected and that the protection is limited to specific wavelengths only, whereas the threats can exist in any other laser wavelengths. We present a passive, solid-state threshold-triggered Wideband Protection Filter (WPF) that blocks the transmission only if the power exceeds a certain threshold. As opposed to fixed spectral filters, which permanently block only specific wavelengths, the wideband filter is clear at all wavelengths until it is hit by damaging light. At input power below threshold, the filter has high transmission over the whole spectral band. However, when the input power exceeds the threshold power, transmission is decreased dramatically. This decreased transmission is limited to the hitting point, where a spot becomes permanently opaque and remains so even after a long exposure to high power. We demonstrate the protection ability of the WPF at several laser wavelengths including protection behavior for single and series of pulses. The WPF can be readily used for protection of detectors, cameras, or eye safety.

6737-17, Session 2
Evolution of test and evaluation of infrared missile warning systems
S. A. Holloway, ESL Defence Ltd. (United Kingdom) Test and evaluation of missile warning systems is most efficiently performed using electro-optic stimulation systems that can simulate the launch and approach of a missile at operational ranges. Much research and development over recent years has gone into optimising the fidelity and power of these systems in the mid-infrared waveband. The work has provided a variety of solutions based generally on black body technology. Recent demands for higher power equipment have pushed development towards a laser based approach. Viable options for laser based architectures have been studied and compared against key criteria such as power input/output, modulation performance, dynamic range, environmental requirements and engineering complexity. An array of quantum cascade lasers has emerged as the most advantageous solution for the present challenges. State of the art quantum cascade lasers from a research institute have been incorporated into the design of two prototype systems. These systems have been thoroughly characterised with results matching the required performance.

6737-15, Session 2
Validation of a target acquisition model for active imagers using perception experiments
F. Lapaz, L. Canevet, Delegation Generale Pour LArmament (France) Active night vision systems based on laser diodes emitters have now reached a technology level allowing military applications. In order to predict the performance of observers using such systems, we developed an analytic model including sensor, atmosphere, visualization and eye effects. The perception task has been modelled using the targeting task performance metric (TTP metric) developed by the Night Vision and Electronic Sensors Directorate (NVESD). Sensor and atmosphere models have been validated separately. In order to validate the whole model, two identification tests have been set up. The first set submitted to trained observers was made of hybrid images. The target to background contrast, the blur and the noise where added to armoured vehicles signatures in accordance to sensor and atmosphere models. The second set of images was made with the same targets, sensed by a real active sensor during field trials. Images were recorded, showing different vehicles, at different ranges and orientations, under different illumination and acquisition configurations. Indeed, this set of real images was built with three different types of gating: wide illumination, illumination of the background and illumination of the target. Analysis of the perception experiments results showed a good concordance between the two sets of images. The calculation of an identification criteria, related to this set of vehicles in the near infrared, gave the same results in both cases. The impact of gating on observers performance was also evaluated.

6737-18, Session 2
Cooperative target identification marking materials for thermal infrared, near infrared and visible sensing
E. S. OKeefe, QinetiQ Ltd. (United Kingdom) Combat Identification is the process of identifying detected objects in the battlefield as friendly, enemy or neutral. Combat Identification is achieved using a combination of situational awareness and target identification capabilities, and is used in conjunction with doctrine, tactics, techniques and procedures to derive a combatants decision to shoot or not to shoot. Current electro-optic cooperative target identification techniques use sensors ranging from human vision, assisted human vision (binoculars), image intensifier based devices, for example, night vision goggles or the UK common weapon sight, and thermal infrared imagers TI for example, the UK thermal weapon sight. These detection and observation methods are used in conjunction with a corresponding vehicle or personnel marking, to comprise what is referred to as the joint combat identification marking system. There are a number of combat scenarios where CTI could be used: ground-to-ground including platform to platform, platform to soldier, soldier to soldier, soldier to platform; air-to-around including, fixed wing aircraft-, rotary-wing aircraft- and UAV-platform- to platform or -soldier; and ground-to-air including platform to air vehicle and soldier to air vehicle. Not all of these scenarios are currently catered for in all sensor wavebands. In this paper we compare some of the existing technologies and introduce some new technologies including lightweight flexible thermal infrared marking materials for dismounted troops.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

13

Conference 6737: Electro-Optical and Infrared Systems: Technology and Applications

6737-19, Session 2
Design and analysis of MEMS-based optical sensor for monitoring density level changes in fluid solution
A. Rahman, Polytechnic Univ. (USA) Optical sensor is playing a vital role for monitoring/detecting physical, chemical and optical properties with high accuracy. A new Micro-Electro-Mechanical System (MEMS) based optical sensor has been presented to monitor the density level changes in fluid solution. In this paper, complete design and analysis is reported to demonstrate a new micromachined optical sensor. The theoretical model is presented based on Fabry-Perot interferometer principle. The numerical model is reported with simulated results using MATLAB and commercial Finite Element (FE) software. Numerical results found in this study are very encouraging which leads to believe that newly presented sensor could effectively be used for monitoring density level in fluid solution in real time. The detailed fabrication processes have been outlined in a multimode optical fiber which is based on conventional MEMS fabrication. The sensor could also be used in the area of defense sensing, microfluidics, nano research, biomedical applications, etc.

6737-20, Session 2
Slant-path atmospheric MTF
A. Zilberman, E. Golbraikh, N. S. Kopeika, Ben-Gurion Univ. of the Negev (Israel) As modern imaging systems become increasingly sensitive and accurate, they are also increasingly affected by the atmospheric channel. It is well known that the atmospheric optical turbulence plays an important role in optical signal propagation through that medium. It is especially important to take into account turbulence characteristics while processing images of objects obtained in the visible or infrared range. The resolution, as well as MTF, is essentially dependent on the properties of turbulent media. It is usually assumed that the turbulent field is of the Kolmogorov type and, in imaging systems, the long- and short-exposure turbulence MTF characterization are based on this type of turbulence. However, experimental data indicates that in the atmospheric boundary layer and at higher altitudes the turbulence can be different from Kolmogorovs case. The change in fluctuation correlations of the refractive index can lead to a considerable change in both the MTF form and the resolution value. Atmospheric aerosol/dust particles cause attenuation and scattering of radiation that may affect the performance of the imaging system. The results indicate that aerosol small-angle forward scatter causes image blur with a noticeable attenuation of high spatial frequencies for increasing optical depth, even under non-fog conditions. In satellite imagery, the aerosol blur is considered to be the primary source of atmospheric blur (the turbulence blur is usually neglected) and is commonly called the adjacency effect. In this work, on the basis of LIDAR measurements and model calculations, the influence of turbulence and atmospheric scatterers on imaging through the atmosphere is estimated for different scenarios of vertical and slant-path propagation. Implications can be significant for optical communication, imaging through the atmosphere, and remote sensing.

wavelengths. The NV detectors are always silicon detectors that are sometimes color-filtered to mimic the spectral distribution of the irradiating source in the test-set. Also, laser sources are frequently used to calibrate a NV detector against a reference detector at the peak wavelength of the test-set LED. As the spectral product of the source distribution and the detector responsivity can be very different in these measurements, the uncertainty of the reticle-radiance calibrations can be too high which is unacceptable for military goggle applications. The spectral responsivity of the goggles can be very different as well resulting in further uncertainty increase in the viewed image which is the spectral product of the reticle-radiance, the responsivity of the goggles photocathode, and the spectral conversion of the phosphor at the output of the image intensifier. In order to perform uniform NV detector, test-set, and goggle calibrations with low uncertainty, a NV calibration facility has been developed at NIST. An integrating sphere, irradiated by the same source distribution used in the test set, produces uniform irradiance for the NIST reference detector and also uniform radiance for the NV (test) detector. The sphere radiance is derived from the irradiance responsivity of the reference detector utilizing the known geometry of the flux transfer. The NV detectors, that transfer the NIST scale to the military primary standard laboratories, were also developed at NIST to perform uniform spectral responsivity and high sensitivity. The selected silicon photodiode has an increased responsivity in the spectral response range of the NV goggles, between 600 nm and 900 nm. Also, the photodiode is thermoelectrically cooled to obtain high shunt resistance needed for low output noise and high sensitivity for the NV detector (radiometer). In the optical design, the photodiode works as a field stop and it produces a radiance measurement angle of 15 degrees. Because of the efficient outof-target blocking of an improved input optics, the distance dependence of the radiance responsivity became negligibly small during calibrations. The uniform goggle-calibration system, which is proposed here for standardized military applications, can decrease the present measurement uncertainties by about an order of magnitude which is an existing need for safe operations. The NIST NV calibration facility, the design issues of the improved NV radiometers, and the uniform calibration system is discussed.

6737-22, Session 2
Human visual performance of a dual band I2/IR sniper scope
P. S. Paicopolis, Army Research Lab. (USA); J. G. Hixson, U.S. Army Night Vision & Electronic Sensors Directorate (USA); V. Noseck, Battelle Memorial Institute (USA) Modeling results of the human visual perception performance of a dual band near (intensified) and long wave (thermal imager) are presented. The system combines an uncooled focal plane array thermal imager with an intensifier tube to provide a simultaneous fused, intensified only, or thermal only on demand image. The fused image can obtain any percentage combination of thermal or intensified imagining. The system uses a common multi-spectral aperture to provide parallax free registered images in each spectrum. A custom designed eyepiece with a flat panel display overlays the thermal channel image on the intensified image. The intensified image is viewed directly on the fiber optic output of the intensifier. Consequently the 64lp/ mm intensified resolution, equivalent to 2300 resolvable lines is maintained in the intensified channel. A system of this type combines the very high line resolution (as required by Johnson criteria for discrimination at longer target ranges) available from intensifiers with very high spot detection for targets of military interest. The results of an analysis of human visual performance using the Armys Night Vision Lab (NVL) model are presented for this dual-band common aperture Sniper Scope that can be mounting on a Sniper weapon.

6737-21, Session 2
Uniform calibration of night vision goggles and test sets
G. P. Eppeldauer, National Institute of Standards and Technology (USA) There is a two and a half orders of magnitude gap between the ~0.1 % (k=2) uncertainty of NIST reference detector calibrations and the uncertainty of field night vision (NV) goggle measurements. Frequently, different source distributions are applied in the reference and transfer calibrations of the NV detectors used to calibrate test-sets and then goggles. In different test-sets, different sources are used to irradiate the reticle-plane viewed by the test goggles. These sources are typically LEDs with different spectral distributions and peak

6737-23, Session 3
Effects of image restoration on automatic acquisition of moving objects in thermal video sequences degraded by the atmosphere
O. Haik, Y. Yitzhaky, Ben-Gurion Univ. of the Negev (Israel) Automatic acquisition of moving objects from long-distance video sequence is a fundamental task in many applications such

14

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6737: Electro-Optical and Infrared Systems: Technology and Applications


as surveillance and reconnaissance. However, the atmospheric degradations, which include blur and spatiotemporal-varying distortions, may reduce the quality of such videos, and therefore, the ability to acquire moving targets automatically. Pervious studies in the field of automatic acquisition of moving objects ignored the blur in the video frames. They usually employed simple methods for noise reduction (such as temporal and spatial smoothing) and motion compensation (registration of frames). The purpose of this work is to determine the effect of image restoration (de-blurring) on the ability to acquire moving objects (such as humans and vehicles) automatically. This is done here by first, restoring the long-distance thermal videos using a novel blind image deconvolution method developed recently, and then comparing the automatic acquisition capabilities in the restored videos versus the non-restored versions. Results show that image restoration can significantly improve the automatic acquisition capability. These results correspond to a previous study which demonstrated that image restoration can significantly improve the ability of human observers to acquire moving objects from a long-range thermal video. In this work, we propose and analyze passive strategies for focusing InfraRed (IR) cameras. We consider an autofocus system split into two main blocks: 1) computation of a focus measure, and 2) research of the best-in-focus lens position. The first block uses a criterion function whose value quantifies the quality of image focus. We analyze and compare different focus measures with reference to some important properties like symmetry, smoothness and lack of local maxima. For the second block we study the important problem of the optimization of the search process and investigate methods to make it as rapid as possible. We consider different search approaches and their implementation on IR test bed. We finally present and discuss examples of results obtained on experimental data sets acquired in outdoor and indoor scenarios.

6737-26, Session 3
Image segmentation based on level set method
Y. Ouyang, X. Qi, Q. Zhang, Institute of Optics and Electronics (China) In this paper, A segmentation model that combines techniques of curve evolution, the Mumford-Shah model and level set method was presented, to detect the contour of object in a given image, the model can detect object whose boundary is not necessarily defined by gradient and whose gray structure may be complicated. First we construct signed distance function, adopted a method which based on the times that is odd or even numbers through close curve from the point along a direction (if need, may be along several directions) to construct sign table. Then we used improved Mumford-Shah model to segment image, we consider that the object to be segmented is made up of some different gray level, it is difficult to detect the object contour using the Mumford-Shah model, for general objects, the contour of the object is piecewise-contour of along the edge, and the gray difference among the object points nearby the contour is little, so we divide the curve into finite segment, compute gray average of narrow band in and out of the curve, and compute the gray difference between the inner narrow band and outer narrow band of the curve, using improved Mumford-Shah model to segment the object. Experiment results show that the proposed algorithm can be used to segment object without edge and with complex gray structure, and the performance of the algorithm is satisfactory.

6737-24, Session 3
Methods for the visualization of high dynamic range IR images
F. Branchitta, M. Diani, G. Corsini, Univ. di Pisa (Italy); A. Porta, Galileo Avionica S.p.A. (Italy); M. Romagnoli, Galileo Avionica S.p.A. (USA) Modern thermal cameras have increased sensitivity which implies high dynamic ranges, typically in the order of 14-15 bits. This requires suitable visualization techniques to adapt the large signal variations to the low dynamic range of the display and of the human visual system (less than 8 bit). Classical dynamic range compression techniques based on simple linear mapping, reduce the perceptibility of small objects and often prevent the human operator from understanding some of the important details. Therefore, sophisticated techniques are required to adapt the recorded signal to the monitor maintaining, and possibly improving, the image visual quality. This problem has been widely investigated in the literature for the visualization of images acquired in the visible spectral domain. Such techniques may be inappropriate for IR sensors or may require important modifications. This subject has been scarcely investigated in the literature, and the definition of proper rendition techniques for IR images is, to some extent, an open problem. The aim of this work is to study new techniques for the visualization of IR images which are capable of accounting for the challenging requirements that must be satisfied by an operating sensor. The proposed techniques attempt to combine dynamic range compression and local contrast enhancement to get a better perception of the details and a pleasant display representation. We discuss the results obtained on a number of experimental data sets acquired in typical operating conditions and compare the performance of the different visualization techniques in terms of quality of the final image presented to the human operator.

6737-27, Session 3
PowerPC-based system for tracking in infrared image sequences
J. Lee, C. Park, J. Lee, Samsung Thales, Ltd. (South Korea) Several powerful algorithms for object tracking have been developed in last two decades. For a stationary camera, the frame differencing method was preferred and it can be generalized to situations where the video data can be easily stabilized. Modern appearance-based tracking methods such as the mean-shift algorithm use histogram-based object appearance models, so it is robust to non-rigid pose changes. Kalman filtering and particle filtering also contribute to enhance the tracking performance. Collins reviewed above tracking methods and emphasized that tracking success or failure depends primarily on how distinguishable an object is from its surroundings. In addition, he also noticed that tracking features need to be used adaptively since both foreground and background appearance can be changed as the target object moves from place to place. In this paper, we propose one tracking scheme based on the block matching and employ several features such as intensity, deviation over time duration, matching error, etc. to classify each pixel into the target region or the background region. Each feature is weighted individually according to separability. So it can fine occlusion occurs when doing object tracking. We implemented our algorithm on PowerPC based System for less calculation time. We use Matroxs Odyssey Xpro+ System. It has several items for vision processing, especially G4 PowerPC core. It is very helpful for real time image processing. Design and implementation of our algorithm can be divided into 6 steps 1. IR sensor image input (Using Host PC)

6737-25, Session 3
Automatic focusing techniques for IR sensors
A. Masini, M. Diani, G. Corsini, Univ. di Pisa (Italy); A. Porta, M. Romagnoli, Galileo Avionica S.p.A. (Italy) The definition of automatic focusing techniques is a crucial problem in many applications, as computer vision, microscopy and ordinary video systems. When the distance between the camera and the object is known, the autofocus process is very easy and straightforward. Therefore, many existing systems use an active sensor to measure the distance to the object. This approach has some disadvantages as the additional costs for the hardware and the fact that an active system emits energy which can be easily detected. To overcome these drawbacks, systems based on passive methods for finding the best focus have been presented in the literature. These techniques estimate the correct focus from images acquired at different positions of the lens. Existing passive autofocus approaches have been proposed with reference to visible cameras, while the application to IR sensors is scarcely treated in the literature.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

15

Conference 6737: Electro-Optical and Infrared Systems: Technology and Applications


2. Global motion Compensation (Using PowerPC) 3. Region Restriction (Using PowerPC) 4. Block Matching (Using PowerPC) 5. Target Position Filter(Using PowerPC) 6.Get the right objects position and display(Using Host PC) This proposed tracking algorithm uses the information from IR sensor for tracking objects When the perfect occlusion occurs, the proposed algorithm predicts movements of an object using the historical tracking information and it can keep the object tracking. CMOS transistors are on the same wafer there is a reduction in pixel area and an additional reduction in the parasitic capacitance effects. This leads to a significant improvement in pixel performance. Pixels incorporating 5 micron and 10 micron diameter GMAPs have been simulated. The circuits were optimised with a view to maximising the photon count rate. Results show a significant improvement in the dead time with values of 42 nanoseconds and 47 nanoseconds being observed for the 5 micron and 10 micron GMAPs respectively. These simulations together with measured results from fabricated pixels will be presented in the final paper.

6737-28, Session 3
Naval target classification by fusion of IR and EO sensors
F. Lapierre, Royal Belgian Military Academy (Belgium); R. Croci, R. Di Stefano, A. Farina, SELEX Sistemi Integrati S.p.A. (Italy); F. Gini, S. Giompapa, Univ. di Pisa (Italy); A. Graziano, SELEX Sistemi Integrati S.p.A. (Italy) This paper describes the classification function of multiple naval targets performed by an infrared camera (IR) and an electrooptical camera (EO), that operate in a more complex multisensor system for the surveillance of a coastal region. The following naval targets are considered: high speed dinghy, immigrant boat, fishing boat, oil tanker. The classification of the targets is performed by mean of the sensors confusion matrix. This is analytically computed as a function of the sensors noise features, the sensors resolution and the size of the involved image database. The model of the IR and EO sensors in the system uses the widely accepted Johnson criterion. For both the sensors, a database of images is generated exploiting a three-dimensional Computer Aided Design (CAD) of the target, for the four ships mentioned above. For the EO camera the image generation is simply obtained by the projection of the three-dimensional CAD on the camera focal plane. For the IR images simulation, we firstly compute the surface temperatures using the OSMOSIS software that efficiently integrates the dependence of the emissivity upon the surface temperature, the wavelength, and the elevation angle. It is applicable to realistic ship geometries. Secondly, we use these temperatures and the environment to predict realistic IR images. The local decisions on the class provided by the two imaging sensors are fused according to a maximum likelihood decision rule. The global performance of the classification process is considered by mean of the final confusion matrix of the system, obtained after the fusion of the local decisions. This analytical approach can effectively reduce the computational load of a Monte Carlo simulation, when the sensors described here are introduced in a more complex multisensor system for the maritime surveillance.

6737-31, Session 4
A novel SWIR detector with an ultra-high internal gain and negligible excess noise
H. Mohseni, O. G. Memis, A. Katsnelson, S. Kong, Northwestern Univ. (USA) Short wave infrared (SWIR) imaging systems have several advantages due to the spectral content of the nightglow and better discrimination against camouflage. Achieving single photon detection sensitivity can significantly improve the image quality of these systems. However, the internal noise of the detector and readout circuits are significant barriers to achieve this goal. One can prove that the noise limitations of the readout can be alleviated, if the detector exhibits sufficiently high internal gain. Unfortunately, the existing detectors with internal gain have a very high noise as well. Here we present the recent results from our novel FOcalized Carrier aUgmented Sensor (FOCUS). It utilizes very high charge compression into a nano-injector, and subsequent carrier injection to achieve high quantum efficiency and high sensitivity at short infrared at room temperature. We obtain internal gain values exceeding several thousand at bias values of less than 1 volt. The current responsivity at 1.55 um is more than 1500 A/W, and the noise equivalent power (NEP) is less that 5x10^15 W/Hz^0.5 at room temperature. These are significantly better than the performance of the existing room temperature devices with internal gain. Also, unlike avalanche-based photodiodes, the measured excess noise factor for our device is near unity, even at very high gain values. The stable gain of the device combined with the low operating voltage are unique advantages of this technology for high-performance SWIR imaging arrays

6737-32, Session 4
Linear array ZnTe/Si heterojunction photodetector for laser detection and imaging applications
R. A. Ismail, Univ. of Technology (Iraq) A linear array 1D of ZnTe/Si heterojunction photodiode for laser detection and imaging applications operated at room temperature in the visible and near infrared regions is presented. Detectors are realized by thermal evaporation of ZnTe film on p-type monocrystalline silicon substrate held at 523K. The main characteristics of the liner array photodetector including currentvoltage, capacitance-voltage, spectral responsivity, uniformity, cross- talk, rise time, and detectivity are investigated. The experimental findings revealed that the presented photodiodes are suitable for imaging applications.

6737-30, Session 4
Optimisation of a Geiger mode avalanche photodiode imaging pixel based on a hybrid bulk SOI CMOS process
N. G. Coakley, A. M. Moloney, A. T. Schwarzbacher, Dublin Institute of Technology (Ireland) Single photon detection has a wide variety of scientific and industrial applications including optical time domain reflectometry, astronomy, spectroscopy, defect monitoring of Complementary Metal Oxide Semiconductor (CMOS) circuits, fluorescence lifetime measurement and imaging. In imaging applications, the dead time is the time during which the detector is inhibited after a photon has been detected. This is a limiting factor on the dynamic range of the pixel. The rate of photon detection will saturate if the dead time is too large. Time constants generated by Metal Oxide Semiconductor (MOS) transistor bulk and sidewall capacitances adversely affect the dead time of pixels developed in conventional CMOS technology. In this paper, a novel imaging pixel configuration based on a Geiger Mode Avalanche Photodiode (GMAP) and fabricated using a dedicated hybrid bulk Silicon On Insulator (SOI) CMOS process is presented. The GMAP is fabricated in the bulk layer and the CMOS circuitry is implemented in the upper SOI layers. As a result, bulk and sidewall capacitance effects are significantly reduced. As both the diode and the

6737-33, Session 4
BIRD640: SCDs high sensitivity VGA VOx bolometer detector
A. R. Fraenkel, U. Mizrahi, L. Bikov, A. Giladi, A. Adin, N. Shiloah, E. Malkinson, T. Czyzewski, Y. Sinai, A. Amsterdam, SemiConductor Devices (Israel) In this paper we report preliminary data of BIRD640, which is a high-sensitivity (50 mK \@ F/1, 60Hz) VGA format detector with 25 m pitch. This high performance is achieved by utilizing an improved pixel design. The product is architecturally compatible to BIRD384 and contains SCDs proprietary unique features (e.g. Power-Save, Ambient drift compensation, etc.). The ROIC architecture follows the framework of the previous designs. It consists of an internal timing machine with a single clock that facilitates the system interface. Extensive effort was invested in reducing the detector and system power dissipation.

16

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6737: Electro-Optical and Infrared Systems: Technology and Applications


The ROIC supports special low power modes, where considerable power is saved with only minor performance degradation. With its superior temporal sensitivity, long-term stability and operational flexibility the BIRD640 serves as an ideal candidate for high end and high resolution uncooled VGA systems, including hand-held applications. The paper will also address SCDs future roadmap and development directions, stressing further user-interface simplification and power reduction. 1. Udi Mizrahi et al. Uncooled Development Program at SCD, Presented at the SPIE Defense & Security Symposium, Orlando, April 2005. 2. A. Fraenkel et al. Advanced Features of SCDs Uncooled Detectors, Presented at the SPIE International Congress on Optics & Optoelectronics, Warsaw August 2005. 3. A. Fraenkel et al. Recent Developments in SCDs VOx Bolometric Detectors, Presented at the SPIE Defense & Security Symposium, Orlando , April 2006. 4. A. U. Mizrahi et al. Large Format and High Sensitivity VOx -Bolometr Detectors at SCD, to be Presented at the SPIE Defense & Security Symposium, Orlando , April 2007. matching technique to model the electromagnetic response of these patterned resistive sheets to achieve the required narrowband IR response. Since the design space consists of four geometrical parameters and one sheet resistance parameter, the modified mode-matching technique is ideally suited for rapid multi-variate optimization. To verify the results of this design optimization, the spectral response is also calculated using HFSS, an industry standard full-wave finite element method EM solver. The results of these FEM simulations will be compared to those from the mode matching calculations.

6737-36, Session 5
A decade of developments of biologically inspired sensory information processing
P. L. McCarley, Air Force Research Lab. (USA) TBA

6737-38, Session 5
Efficient readout for carbon nanotube (CNT)based IR detectors
N. Xi, Michigan State Univ. (USA) The objective of this work is to develop an efficient readout for carbon nanotube (CNT) based middle wave IR (MWIR) detectors. For the traditional MWIR sensing materials, the thermal noise at room temperature is significant comparing with the signal. Hence the traditional MWIR detectors normally have to work at extremely low temperature. As a one-dimensional nano structural material, CNTs have a potential to be efficient infrared (IR) detection material due to their unique electronic properties. The ballistic electronic transport property of CNT makes the noise equivalent temperature difference much smaller comparing with other semi conducting materials. Hence, it is possible to have the CNT based IR detector work at room temperature or moderate low temperature. Due to the unique structure of the CNT based IR detector, the readout mechanism is different from traditional IR detectors. In this work, a diode based readout has been developed. By utilizing the Schottky contacts between CNT and electrodes, a semi conductive CNT works as a photodiode. The junction created by the Schottky contacts can significantly increase the photo current and reduce the noise. The proper design of the diode can significantly increase the signal noise ration of the CNT based MWIR detectors. The experimental results have demonstrated the advantages of the design. The improvement of the signal to noise ration can be improved as much as 3000 times.

6737-34, Session 5
Compact uncooled amorphous silicon 160x120 IRFPA with 25 m pixel-pitch for large volume applications
J. Tissot, O. Legras, C. Trouilleau, B. Fieque, S. Tinnes, C. Minassian, ULIS (France); J. Yon, Lab. dElectronique de Technologie de lInformation (France) The rapid mastering of the 25 m amorphous silicon technology allowed ULIS to develop and produce in volume a very compact 160 x 120 with 25 m pixel-pitch IRFPA for low end camera for imaging or thermographic IR camera. This detector has conserved all the innovations developed on the full TV format ROIC (detector configuration by serial link, low power consumption or wide electrical dynamic range ... ) and offers an advanced, highly reliable, RoHs compliant, TECless focal plane array very which is well adapted to thermal imaging. The specific appeal of this unit lies in the miniaturization of the packaging and its extremely light weight. The reduction of the pixel-pitch and the innovative package turn this 160 x 120 pixels array format into a low cost product and therefore well adapted for mass production. In the last part of the paper, we will look more closely at electrooptical performances of this TEC-less product 160 x 120 as well as the other 25 m products like the 640 x 480 and 384 x 288. We will insist on the wide thermal dynamic range and the low consumption achieved thanks to the mastering of the amorphous silicon technology coupled with the innovation and care in the ROIC design.

6737-39, Session 5
Model based on-chip 13bits ADC design dedicated to uncooled infrared focal plane arrays
B. Dupont, P. Robert, ULIS (France) This paper presents an on-chip 13 bits 10 M/S Analog to digital converter specifically designed for infrared bolometric image sensor. Bolometric infrared sensors are MEMs based thermal sensors, which covers a large spectrum of infrared applications, ranging from night vision to predictive industrial maintenance and medical imaging. With the current move towards submicron technologies, the demand for more integrated smart sensors has dramatically increased. This trend has strengthened the need of on-chip ADC as the interface between the analog core and the digital processing electronic. However designing an on-chip ADC dedicated to focal plane array raises many questions about its architecture and its performance requirements. To take into account those specific needs, a high level Matlab model has been developed prior to the actual design. In this paper, we present the trade-offs of ADC design linked to infrared key performance parameters and bolometric technology detection method. The original development scheme, based on system level modeling, is also discussed. Finally we describe the design and present the measured performances.

6737-35, Session 5
Wavelength-selective infrared detectors
D. P. Neikirk, J. Jung, The Univ. of Texas at Austin (USA) Several approaches that produce significant wavelength selectivity in micromachined microbolometer designs will be discussed. These frequency selective surfaces can be achieved using stacks of dielectric coated resistive sheets or by replacing the normal uniform absorbing sheet used in IR microbolometers with true microbolometers (i.e., bolometers that are much smaller than the wavelength) combined with an antenna. Here we discuss designs that can substantially improve the wavelength selectivity of microbolometers. Both dielectric coated sheets and planar multimode antennas have shown that they should be able to produce enough spectral selectivity to allow a three-color system spanning the 7-14 micron band. To achieve a narrowband spectral response with the simplest possible fabrication process, we have also investigated the use of patterned resistive sheets as the frequency-selective absorber. In comparison with planar antenna-coupled microbolometers that consist of both resistive and highly conductive metal strips (acting as antennas), the absorption layer in these structures involves only a single resistive layer with patterned holes. We have developed a modified mode-

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

17

Conference 6737: Electro-Optical and Infrared Systems: Technology and Applications

6737-40, Session 5
Latest developments in MCT infrared staring arrays at Sofradir
M. Vuillermet, F. Pistone, A. Manissadjian, Sofradir (France) HgCdTe (Mercury Cadmium Telluride / MCT) staring arrays for infrared detection do show constant improvements regarding their compactness and performances. New detectors are now proposed offering system solutions in the different IR wavebands and profiting of the latest technology improvements as well as MCT performance advantages and cost reduction. Among these new detectors, one can find the family of 15 m pixel pitch detectors including a mid TV format (384 x 288), a TV format (640 x 512) and a double TV format (1280 x 1024). The latest detector development concerning the mid-TV format is performed according to very challenging specifications regarding small cost and low power consumption. These Focal Plane Arrays (FPA) are integrated in dedicated tactical Dewars, taking advantages on last development in coolers manufacturing and dewar assembly. Another development axis at CEA\LETI-LIR and Sofradir concerns the avalanche photodiodes for FPA sensitivity improvement. This very promising technology is dedicated for low flux applications as active imagery, hyperspectral applications or small aperture systems. New development results are presented and future trends are discussed.

6737-42, Session 5
Simulation of MWIR and LWIR photodiodes based on n+-p and p-n junctions formed in HgCdTe heterostructures
M. S. Nikitin, G. V. Chekanova, Alpha (Russia); A. A. Drugova, V. A. Kholodnov, Institute of Radio-engineering and Electronics (Russia) Small-pitched megapixel photovoltaic infrared focal plane arrays (IRFPA) having ultimate performance in spectral range from 3 to 5.5 m (peak wavelength p=4-4.8 m) at higher operating temperature (HOT) Top=200-240 K and in spectral range from 8 to 12 m (p=10-10.5 m) at Top=80-100 K are considered preferable for development of future thermal imaging systems. Novel MBE-grown Hg1-xCdxTe epitaxial multi-layer structures including graded-gap layers are perspective for that purpose due to low growth temperature providing precise control on individual layer thickness and composition x gradients. Objective of the present work was to examine the impact of MWIR (p ranged from 4 to 4.8 m at Top=200-240 K) Hg1xCdxTe PV device and LWIR (p ranged from 10 to 10.5 m at Top=80-100 K) performance on variation of doping level, absorption layer thickness, composition x gradient and operating temperature.

18

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6738: Technologies for Optical Countermeasures


Monday-Tuesday 17-18 September 2007 Part of Proceedings of SPIE Vol. 6738 Technologies for Optical Countermeasures IV

6738-02, Session 1
The development and application of photonic technology in infrared and electro-optic countermeasures
L. Cooke, BAE Systems plc (United Kingdom) In this paper I will discuss recent work at the Advanced Technology Centre of BAE Systems on photonic technology, in particular photonic crystal fibres, applied to infra-red and electro-optic countermeasure systems. The use of Photonic Crystal fibres or holey fibres in countermeasure systems could significantly simplify platform integration by enabling remote location of laser sources, the generation of multiple wavelengths or continuum generation from a single pump source .The paper will describe the development of these fibres, drawing examples from recent civil collaborative research projects such as PFIDEL and LAMPS. The paper will also present an overview of developments in the coherent combining of high power fibre laser technology for Laser Directed Energy systems.

output coupler and the OPO is operated so close to degeneracy that all energy is contained in a 2 nm wide region. This OPO output has been used as a pump source for a conventional ZGP OPO demonstrating efficient conversion and providing broadband tunable output in the mid-infrared.

6738-05, Session 2
High performance optically-pumped antimonide lasers operating in the 2.5-9 m wavelength range
R. Kaspi, A. P. Ongstad, G. C. Dente, M. L. Tilton, D. M. Gianardi, Jr., J. R. Chavez, Air Force Research Lab. (USA) We describe optically-pumped type-II mid-infrared lasers based on the antimonides that exhibit very high photon-to-photon conversion efficiencies (~65% \@ 80K) and low waveguide loss (< 3cm-1). The design flexibility that these type-II wells provide, when coupled with the ultra low-confinement of the transverse optical mode, produce multi-Watt lasers with emission wavelengths ranging from 2 m to 9.4 m. For example, ~11 Watts of quasi-cw power is obtained from a 3.7 m device at 85K. Despite the broad area of the devices, the lateral beam quality is surprisingly good (typically ~5 times the diffraction-limit) without any lateral mode control. This is primarily due to the suppression of filamentation as a result of low optical confinement that is afforded by the optically pumped design. However, the optically pumped devices achieve higher brightness operation as unstable resonators. Each unstable resonator is realized by polishing a diverging cylindrical mirror at one of the facets. A compact, multi-Watt mid-infrared laser transmitter that uses a novel thermal management approach to enable the cryogenic optically-pumped semiconductor laser to be packaged into a small volume has also been developed. Multi-Watt output in continuous wave and pulsed mode operation has been demonstrated.

6738-03, Session 1
High average power thulium fibre laser pumped mid-IR source
I. Elder, SELEX Sensors and Airborne Systems Ltd. (United Kingdom) Results are presented for an efficient, compact laser source of high average power in the 3-5 mm range. The laser architecture uses a high power thulium fibre laser as the pump source for a bulk Q-switched Ho:YAG laser. 66% conversion efficiency to 2.09 m with excellent beam quality (M2 <1.5) has been achieved using an L-shaped Ho:YAG resonator to allow doublepassing of the pump light through the 1% doped laser rod. The maximum Ho:YAG output power attained was 27.3 W at a Qswitch PRF of 25 kHz. The Ho:YAG output was used to pump a ZGP OPO, resulting in 12.6 W in the 3.8-4.8 m wavelength range (a conversion efficiency of 52% from pump power incident at the ZGP), with a beam quality factor M2 of 2.5. The ZGP OPO used a linear two mirror resonator with a single 15 mm long crystal, utilising a single pass of the pump beam; the maximum pump beam energy density incident at the ZGP was limited to 0.55 J.cm-2.

6738-06, Session 2
High-brightness 2-to-5 m semiconductor lasers
M. Rattunde, F. Fuchs, C. Mann, Q. Yang, N. Schulz, M. T. Kelemen, K. Koehler, J. Schmitz, G. Kaufel, W. Bronner, J. Wagner, Fraunhofer-Institut fr Angewandte Festkrperphysik (Germany) An increasing number of security-related applications require compact high-brightness 2-to-5 m laser sources. (AlGaIn)(AsSb)-based quantum-well diode lasers are well suited to cover the 2 - 2.5 m spectral range. For broad-area lasers emitting at 2 m, high power efficiencies (\>25%) and output powers of 2 W in continuous-wave (cw) mode (\>9 W in pulsed mode) have been achieved at room-temperature, while linear laser arrays (laser bars) with 19 emitters yield a cw output power of 20 W. Compared to broad-area lasers, tapered diode lasers show a significant improvement of the slow-axis beam quality. At 1.9 m wavelength these lasers produce a nearly diffraction limited output beam with M2<1.7 up to an output power of 1.5 W, resulting in a brightness of 30 MW/cm2. If the application requires a circularly symmetric output beam, the optically-pumped semiconductor disc laser (OPSDL) is a promising approach, providing an output power of 1.5 W (M2<3) at -20C and at a wavelength of 2.3 m. Optimization for highest beam quality yields a diffraction limited (M2<1.05) output beam for an only 30% reduction in output power. There is potential for further up-scaling the output power and for extension of the emission wavelength beyond 3 m. To cover the adjacent 3.5-5 m spectral range GaInAs/AlAsSb/ InP quantum cascade (QC) lasers have been demonstrated recently, yielding pulsed mode operation up to 400 K for devices emitting at 4.5-4.6 m. GaInAs/AlGaAsSb ridge-waveguide QC lasers emitting at 3.7 m are capable of emitting a maximum peak output power of 10.5 W at 77 K, (2.1 W \@ 250 K)corresponding to a power efficiency of 22% at 77 K.

6738-04, Session 1
Tandem OPO system for mid-infrared generation using quasi phase-matching and volume Bragg grating
M. Henriksson, Swedish Defence Research Agency (Sweden) and Royal Institute of Technology (Sweden); L. J. Sjqvist, Swedish Defence Research Agency (Sweden); M. Tiihonen, V. Pasiskevicius, F. Laurell, Kungliga Tekniska Hgskolan (Sweden) Efficient laser sources in the 3 - 5 m wavelength range are needed for directed infrared countermeasures, but also have applications in remote sensing, medicine and spectroscopy. We will present results and discuss the possibilities of a tandem optical parametric oscillator (OPO) scheme for converting the radiation from a 1.06 m Nd-laser to the mid-infrared. Our setup uses type I quasi phase-matched (QPM) crystals in a near degenerate OPO to generate 2.13 m radiation. The QPM crystal provides higher nonlinearity and longer interaction lengths, because walk-off is avoided, compared to conventional bulk crystals. This is an advantage especially in low pulse energy applications. To make the 2.13 m radiation usable for pumping a second OPO a volume Bragg grating is used as a cavity mirror to limit the bandwidth, which in a conventional QPM OPO at degeneracy can be several hundred nanometers. The acceptance bandwidth for efficient OPO operation of a 14 mm long ZnGeP2 (ZGP) crystal is approximately 5 nm, which makes the need for bandwidth limiting clear. Bandwidths below 0.5 nm are measured for the signal and idler from a periodically poled KTiOPO4 (PPKTP) OPO with a volume Bragg grating

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

19

Conference 6738: Technologies for Optical Countermeasures

6738-07, Session 2
Mid-infrared InAsSbP/InAsSb quantum well laser diodes
M. Yin, A. Krier, Lancaster Univ. (United Kingdom) Among the light emitting materials which can access the 3-5 micron mid-infrared spectral range the InAsSbP alloy system is a promising candidate for applications in the area of both optical countermeasures and free space optical communications. This paper describes the characteristics of a separate confinement heterostructure(SCH) laser design based on type I InAsSbP/InAsSb multiple quantum wells (MQW). An 8X8 band k.p method was used to calculate the band structure. The optical gain of active regions containing InAsSb QW with different compositions was calculated using a free carrier gain model. Other properties such as behaviour of the fundamental optical mode and refractive index profile were also determined. These were used for simulation of the resulting final device properties and to estimate the threshold modal gain and threshold current density for our InAsSb MQW laser. As a first step towards practical realisation of these lasers we report on the fabrication of the laser active region containing a single InAsSb type I strained quantum well grown using liquid phase epitaxy. We fabricated a prototype light-emitting diode structure and successfully observed electroluminescence from the confined states of the quantum well in good agreement with our calculations. Results on the growth of the InAsSbP waveguide will also be presented.

conversion efficiency and reliability of these devices continues to improve, at an ever-increasing range of wavelengths, improving performance in existing applications and enabling new developments. For the example of high power bars on microchannel coolers, output power levels of up to 1-kW per bar have been reported, with power conversion efficiency up to 85% and emission wavelength beyond 2000-nm. We present a review of latest performance in this key technology and discuss potential for further improvements.

6738-10, Session 3
A 2m-pump-laser-based DIRCM system
W. Bohn, G. Renz, Deutsches Zentrum fr Luft- und Raumfahrt e.V. (Germany) The improvement of the security of platforms (aircrafts) with countermeasure techniques in the mid-IR especially in the takeoff or landing phase is nowadays more stringent due to upcoming threats. We report on the development of a Tm:YLFfiber laser (1.908 m) pumped Ho:YAG (2.09 m) high energy laser system with pulse energies up to 90 mJ at pulse lengths close to 20 ns and repetition rates of 100 Hz. A high quality laser beam leaving a platform through a variableindex-of-refraction airflow will experience wave-front aberrations and consequently lose its ability to be perfectly focused in the far field. Two main causes of laser beam degradations are issued in this investigation. First, there is the degradation immediately around the fuselage, referred to aerooptic problems and second the atmospheric propagation influence via air turbulence. The aero-optic influence on the laser beam degradation will be investigated in a laboratory experimental approach with a mid-IR laser beam traversing a transonic free air stream relevant to a real air flow around a fuselage. The propagation characteristics of a laser beam passing turbulent air will be numerically simulated with a multiple phase-screen method and a Fourier propagation technique. Different turbulence degrees relevant to propagation directions especially behind aircrafts will be considered.

6738-08, Session 2
Monolithic high brightness diode lasers - results and developments at FBH
G. Erbert, F. Bugge, J. Fricke, K. Paschke, H. Wenzel, G. Trnkle, Ferdinand-Braun-Institut fr Hchstfrequenztechnik (Germany) Brightness is one of the most important criteria for diode laser applications beside reliability and efficiency. For power levels approaching the Watt range and above it is quite challenging to get single mode output power from monolithic diode lasers. The limitations are given by the impact of high power densities on reliability and such effects like thermal lensing and self focusing. The increase in vertical mode size by maintaining a high efficiency is one of the key issues, to overcome these limitations. FBH has developed specific designs for fundamental mode lasers, tapered devices and the implementation of gratings for spectral mode selection. The technological base consists of MOVPE, holographic pattern generation, wet and dry chemical etching processes and facet passivation techniques all included in 2" and 3" processing lines for diode lasers. Beside an overview on current FBH activities examples of monolithic high brightness devices will be given. Narrow stripe devices based on ridge waveguide structures can now deliver an output power of nearly 2W in fundamental mode at 1060nm with very low vertical divergence of 15 full width half maximum. Similar results seems possible for the wavelength range down to 800nm. With implementation of gratings leading to DFB resonators an output power of more than 0.5W in nearly true single frequency operation could be achieved. Devices with unstable, tapered resonator structures are the most promising monolithic approaches for higher output powers. Design issues and characteristics of a monolithic 10W device at 980nm will be given.

6738-11, Session 3
Development of a compact laser source for airborne countermeasures
A. Godard, M. Lefebvre, ONERA (France); S. A. Said Hassani, P. Galtier, Ctr. National de la Recherche Scientifique (France) We report on the development of a compact laser source devoted to airborne countermeasures. Several spectral lines are emitted simultaneously by the use of optical parametric oscillators (OPOs) pumped by a high repetition rate near-IR pulsed laser. The general architecture of the source is designed to locate the pump laser inside the aircraft while the conversion stages are positioned close to the orientation turret. In this purpose, the dimensions of the whole setup are minimized with a special attention on the miniaturization of the OPO cavity whose typical volume is less than 100 cm3. These small size OPOs are optimized to produce the required average power under high repetition rate pumping and the emitted output beam quality is typically 1.5 diffraction limited. The selection of the proper nonlinear material for each OPO is also a critical issue. Zinc germanium phosphide, ZnGeP2 (ZGP) is currently the material of choice to operate where conventional nonlinear materials strongly absorb. In this context, the ZGP crystal growth has been investigated. The first samples with a good optical quality have been obtained and the preliminary results are discussed. In addition to countermeasure applications, we also discuss other Defence & Security issues that can be addressed by miniature OPOs such as neurotoxic detection by use of a singlemode OPO pumped by micro laser.

6738-09, Session 3
High power semiconductor laser sources for defence and security: a review of current technology
J. Bell, nLight Corp. (USA) Semiconductor lasers are a key enabling component for advanced optical systems in security and defense. These components convert supplied electric current into light, either for use directly or as a pump source for other gain media such as YAG crystals or fiber lasers. The available power level, power

6738-12, Session 3
High power and efficient far infrared ZnGeP2based optical parametric oscillator
E. Lippert, G. Rustad, Norwegian Defense Research Establishment (Norway); K. Stenersen, Norwegian Defence Research Establishment (Norway) Optical parametric down-conversion in ZnGeP2 (ZGP)-based

20

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6738: Technologies for Optical Countermeasures


optical parametric oscillators (OPOs) pumped by 2-m-lasers has proved to be an efficient way of generating high power radiation in the infrared, and multi-watt sources in the 3-5 m spectral range have been reported. However, ZGP is also well suited for conversion to even longer wavelengths. In this work we have generated high power in the 8-10 m range by conversion of an efficient 2 m laser. The 2 m source is a fiber laser pumped Ho:YAG laser, delivering 9 W of average power, in Q switched operation, to the ZGP OPO. From this setup we obtained 915 mW of output power at 8 m in a beam with M2-value of better than 2. The conversion efficiency in the ZGP OPO from 2 m to 8 m was 10 %. We also demonstrated continuous tuning of the OPO to 9.6 m where the conversion efficiency was still 5 %. This source covers an important spectral range where few other practical high power sources exist. electro-optical ceramics which cover a broader spectral range up to the MWIR are under development. The paper outlines the concept of adaptive compensation of the wavefront error of each micro-lens telescope in a blazed grating beam steerer. The theory of combining micro-lens arrays with large format spatial light modulators for the NIR and the MWIR spectral ranges will be developed. Measurements at component level, i. e. wavefront aberrations of micro-lens arrays (256x256 lenses) and phase shifts of a commercially available liquid-crystal-on-silicon spatial light modulator (1920x1080 pixels) at 1.5 m will be presented in order to support the theory.

6738-16, Session 5
Anatomy of the MANPAD
M. A. Richardson, Cranfield Univ. (United Kingdom) This paper describes the options available for a Man-Portable Air-Defence (MANPAD) weapon. The strengths and weaknesses of the options are quickly discussed and this highlights the choice of the Infrared Surface-to-Air Missile (IRSAM) as the most likely external threat to civil aircraft. The paper then looks at the signature aspects of the aircraft which drives the choice of the operating waveband of the IR seeker. The paper then illustrates how early generation IRSAMs work and finally sets this in the context of platform survivability.

6738-13, Session 4
Phased array beam steering using fibre lasers
A. M. Scott, QinetiQ Ltd. (United Kingdom) Phased array beam steering offers a way of rapidly controlling the direction of a laser beam, but provides steering over a limited field of view. We discuss the limits of such a beam steering system and describe the performance of a phased array of fibre channels in which the phase of each fibre can be rapidly controlled.

6738-17, Session 5
The European project CASAM for the protection of commercial airliners in flight
J. Vergnolle, Sagem SA (France) Commercial aircraft, as part of mass transportation systems, are a potential target for terrorists because they represent one of the best achievements of our society. As a result, an attack would have a big psychological impact on people and economic activity. CASAM is focusing on a potential solution to reduce aircraft vulnerability against Man Portable Air Defense Systems (Manpads) during takeoff, ascent and landing. A specific on-board missile jamming system shall be developed against stringent but competitive requirements dealing with high reliability, low cost and minimal installation constraints on the aircraft. CASAM (Civil Aircraft Security Against Manpads) is a Framework Program 6 (FP6) Project** from the European Commission, DG Research-Aeronautics, putting together a group of 18 Companies, from Majors to SMEs, bringing their respective knowledge into a scientific and technical team able to address all technical, financial and legal matters dealing with this challenging topic. Project Objectives The global objective of the CASAM Project is to design and validate a closed-loop laser-based DIRCM (Directed IR Countermeasure) module for jamming the potentially Surfaceto-Air Missile(s) fired against a commercial airliner. A closedloop laser system will bring a step forward to this type of directed jamming system by providing it the new capability to actively track the approaching missile, identify its IR seeker type and rapidly return the right jamming laser beam; a way to quickly and successfully jam a threat. Time is the most critical parameter and all the sequence of events from the detection of the incoming missile till its succesfull jamming must be done very quickly in order to either be able to protect the aircraft at low altitude or address a multi-threat likely scenario. Such a DIRCM system will comply with all commercial air transportation constraints. CASAMs solution is designed accordingly. For example, consideration must be given to the following aspects: Environmental friendliness for ground objects and inhabitants close to the airport, safety for aircraft (maintenance, handling and usage) and; high efficiency against the recognized threats and upgradeability for further/future disseminated threats; Adherence to commercial operation budgets and processes. Technical Challenges CASAM explores several technological breakthroughs in lasers,

6738-14, Session 4
A real-time sub-rad laser beam tracking system
I. Buske, W. Riede, Deutsches Zentrum fr Luft- und Raumfahrt e.V. (Germany) Long-range imaging applications for identification of fast moving objects require a high precision laser beam steering and tracking system to eliminate system vibrations and to compensate for the atmospheric beam wander effect. For typical diffractionlimited f/10 objectives with a focal length of 1 - 2 m and high resolution imaging sensors with a small pixel pitch, a sub-rad accuracy of the laser beam tracking system is essential. We present a rugged and reliable real-time laser beam tracking system operating with a high speed, high resolution piezoelectric tip/tilt mirror. An industrial compactRIO programmable automation controller (PAC) is used to develop a deterministic digital PID controller. The functionality of the reconfigurable I/O system is implemented by means of the graphical development environment LabVIEW. The controller provides a 1 million field programmable gate array (FPGA) to realize a high closed-loop frequency of up to 10 kHz. Interferences and background noise can be eliminated by using a lock-in amplifier. Beam tracking with a root-mean-squared accuracy better than 50 nrad has been laboratory-confirmed. The system is intended as an addon module for established mechanical mrad tracking systems.

6738-15, Session 4
Adaptive Laser Beam Steering with MicroOptical Arrays
M. Rungenhagen, H. D. Tholl, Diehl BGT Defence GmbH & Co. KG (Germany) Micro-lens-arrays of large formats are well suited for agile laser beam steering with demonstrated high rates and moderate accuracy. Piezoelectric transducers are the appropriate choice for driving the lenslets arrangements which act like a blazed grating structure. Blazed grating micro-lens array beam steerers suffer from nonuniformity of the optical parameters across the array which leads to a reduction of the spatial coherence between the interfering beam-lets and an increase in the beam divergence. This disadvantage can be resolved by combining the blazed grating beam steerer with a phased array. If the number of pixels of the phased array is sufficiently large, several pixels cover one period of the blazed grating and the piston as well as higher wavefront errors (tip/tilt, defocus) may be corrected. In the VIS and NIR spectral range large format liquid crystal phased arrays are available for adaptive correction of the wavefront of each microlens in the array. Phased arrays based on micro-mirrors or

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

21

Conference 6738: Technologies for Optical Countermeasures


optics, electro-mechanics and processing, which will be the hard core of a future competitive equipment. A technical and architecture validation breadboard will be ground-tested against actual missile seekers. In the project framework, a specific effort will be made with respect to threat analysis and simulation as well as economic analysis, aircraft installation constraints and impact. The main technical challenges are linked to: - global requirements of airlines and airframers ( low total volume, low drag, low mass, low power consumption, high reliability, low life-cycle cost requirements and no induced risk on the ground and during takeoff and landing); - jamming efficiency against most of the IR-guided SAM fired under several constraining but realistic scenarios. The ways to comply with such challenging requirements and the associated risks are directly linked to significant technological improvements and system architecture simplification through an innovative approach. The DIRCM modules innovative work mainly focuses on the following: - Reducing optronics volume, mass and costs. The optomechanical turret will achieve outstanding performance in steering and stabilization. New focal plane array (imagery sensor) will integrate passive and active detection modes. - New, efficient laser-technology approaches, including fiber lasers and simpler frequency conversion modules (OPO), as well as directly-emitting mid-infrared semiconductor lasers. Papers Table of content would be: Summary 1. Introduction 2. Project Objectives, schedule and organization 3. Key requirements for the security of commercial jetliners 4. Innovative DIRCM closed-loop solution - Compact Architecture - Innovative and competitive sub-systems 5. Foreseen ground tests 6. Follow-up Project 7. Conclusion * Jean-Franois Vergnolle, SAGEM Dfense Scurit, 27 rue Leblanc, 75512 PARIS Cedex 15, France Tl : +33 158 11 25 37, Fax n: +33 158 11 70 84, mail: jeanfrancois.vergnolle\@sagem.com ** CASAM, FP6 contract n AST5-CT-2006-030817 show that dazzling can highly affect the performance of the used pattern recognition algorithms by generating lots of spurious edges which mimic the reference symbol. As a consequence dazzling results in detrimental effects, since it not only prevents the recognizing of well defined symbols, but it also creates many false alarms.

6738-20, Session 6
Progress and development in fibre laser technology
R. Horley, Southampton Photonics, Inc. (United Kingdom); S. Norman, M. N. Zervas, SPI Lasers plc (United Kingdom) High performance fibre lasers are now well established as an extremely robust and reliable technology enabling a growing and diverse number of demanding industrial, medical and defence applications. Compared to rival technologies, such as CO2, LPSS, DPSS and disk lasers, fibre lasers offer a number of unique characteristics that have resulted in their wide adoption in an increasing number of industrial sectors. In addition to replacing conventional lasers in existing applications, fibre lasers have been very successful in enabling new applications, both factors which explain their increasing market share. Fibre lasers have an all-fibre, all-guided architecture, which makes them robust and reliable. They display no thermal lensing effects, which results in excellent beam pointing stability compared to conventional lasers. They are unique in offering instant turn-on operation without regular tuning and realignment. The inherently large surface-to-volume ratio facilitates heat removal and minimizes the external cooling requirements. The beam quality can be engineered by proper fibre core design to match the application requirements. Fibre lasers are usually pumped by combining a number of extremely robust, telecommunication-grade, single-emitter broad-stripe multimode pump diodes. This results in laser systems with long lifetime, maintenance-free operation. In addition, due to very low loss, monolithic all-fibre geometries, they have superior optical-to-optical, electrical-to-optical and overall wall-plug efficiencies. Finally, fibre lasers have a very small and compact footprint that facilitates their system integration. Key to harnessing the benefits of an all-fibre laser configuration is the adopted pump-coupling scheme. All SPIs laser products use a proprietary cladding-pumping technology (GTWave&#61668;) that enables multi-port, distributed pump injection and facilitates output power scalability. This modular approach results in robust, easily deployable CW/CWM fibre laser/amplifier systems with output power in the 10-400W range, covering the 1060nm and 1550nm wavelength regions. All active fibres are designed to be photo-darkening free, adding into system reliability and longevity. SPIs high power fibre lasers are used in a number of industrial, medical and defence applications, including stent tube cutting, disc drive spot welding, electronic stencil cutting, automotive plastic welding, medical piece part welding, rapid prototyping, medical aesthetics and directed-energy. When in a MOPA configuration, optical fibres, being extremely high gain media (\>50dB small-signal gain per stage) can provide high performance pulsed systems. SPIs redEnergy&#61668; products are pulsed lasers with energies in the region of 0.8mJ and peak power in excess of 15kW, at 25 kHz repetition rate. A unique feature of SPIs pulsed laser is the capability of producing high peak power pulses at very high repetition rates (>400 kHz). These lasers have found widespread applications in many precision, high-throughput, marking applications.

6738-19, Session 5
Assessment of laser-dazzling effects on TVcameras by means of pattern recognition algorithms
A. Durcu, P. Bourdon, ONERA (France); H. Brsing, Forschungsfesellschaft fr Angewandte Naturwissenschaften e.V. (Germany); J. Dellinger, N. Duchateau, Institut dOptique (France); B. Eberle, Forschungsfesellschaft fr Angewandte Naturwissenschaften e.V. (Germany); O. Vasseur, ONERA (France) Imaging systems are widespread observation tools used to fulfil various functions such as detection, recognition, identification and video-tracking. These devices can be dazzled by using intensive light sources, e.g. lasers. In order to avoid such a disturbance, dazzling effects in TV-cameras must be better understood. In this paper we studied the influence of laser-dazzling on the performance of pattern recognition algorithms. The experiments were performed using a black and white TV-CCD-camera, dazzled by a nanosecond frequency doubled Nd:YAG laser. The camera observed a scene comprising different geometrical forms which had to be recognized by the algorithm. Different dazzling conditions were studied by varying the laser repetition rate, the pulse energy and the position of the geometrical forms relative to the laser spot. The algorithm is based on edge detection and locates areas with similar forms compared to a reference symbol. As a measure of correspondence it computes the degree of correlation of the different areas. The experiments

6738-21, Session 6
Laser-dazzling effects on TV-cameras: analysis of dazzling effects and experimental parameters weight assessment
A. Durcu, P. Bourdon, O. Vasseur, ONERA (France) Imaging systems are widespread observation tools used to fulfil various functions such as detection, recognition, identification and video-tracking. These devices can be dazzled by using intensive light sources, e.g. lasers. In order to avoid such a disturbance, dazzling effects in TV-cameras must be better understood.

22

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6738: Technologies for Optical Countermeasures


In this paper we studied the influence of different parameters on laser-dazzling. The experiments were performed using a black and white TV-CCD-camera, dazzled by a nanosecond frequency doubled Nd:YAG laser. Different dazzling conditions were studied by varying for instance the laser repetition rate, the pulse energy or the settings of the camera. We proceeded in two steps. First the different dazzling effects were analyzed and classified by their mainspring. Pure optical phenomena like multiple reflections, scattering and diffraction were discriminated from electronics effects related to charge transfer processes. Interactions between the laser repetition rate and the camera frequency or the camera exposure time were also observed. In a second step, experiments were carried out for different dazzling conditions. It was then possible to assess the weight of each experimental parameter on dazzling effects. The analysis of these quantitative results contributes to the better understanding of laser-dazzling, useful to design efficient means to protect imaging systems.

6738-24, Session 6
Laser dazzling of infrared focal plane arrays
R. H. M. A. Schleijpen, J. C. van den Heuvel, A. L. Mieremet, TNO (Netherlands) Laser countermeasures against Infrared focal plane array cameras aim to saturate the full image. This paper will discuss optical mechanisms causing area saturation of focal plane arrays by lasers, such as diffraction, multiple reflections and optical scatter. The role of the read-out circuit will be addressed. Experimental results of MWIR laser saturation as a function of laser power are compared with a theoretical model. An empirical method for estimating the size of the saturated area in the camera image will be presented. Based on this empirical method a technique has been developed to generate synthetically partially saturated images for various laser irradiance levels on the camera. These simulated images can be used for testing the effects of dazzling on tracking processing.

6738-22, Session 6
Helicopters on the asymmetric battlefield: challenges for photonics
J. Heikell, Helsinki Univ. of Technology (Finland) This review is part of a larger work that seeks to improve understanding of survivability of battlefield helicopters on the asymmetric battlefield. The present work aims at identifying areas of photonics that require detailed investigation. Helicopters have seen battlefield use since Korea and were used en masse in Vietnam. In the traditional Cold War scenario, the survivability discussion focuses on ballistic weapons and radiofrequency threats, while photonic threats made a modest appearance in the last stages of the Vietnam War. The shift to asymmetric threats, which intensified after the terrorist attack in September 2001, requires an unbiased reconsideration of helicopter survivability. The asymmetric battlefield has few radiofrequency threats, but there is an abundance of ballistic weapons and antitank missiles with optical and electro-optical sights together with infrared guided man-portable missiles. Mines with infrared triggers have been reported, whereas laser trigged devises have not been reported but should be expected. The frequency of hostile activities is low, but when occurring need be detected, identified and located unambiguously. Changes like these call for an evaluation of existing electronic warfare technology and of technologies under development. Both threat warning and countermeasure technologies need be assessed for their true potential. Where technology alone does not offer a solution, improved understanding of the interactions between technology, tactics, situational awareness, and other factors is needed. The first requirement, however, is to have understanding of battlefield signatures-a challenging task when the battle is fought in and above a functioning urban society.

6738-25, Session 6
Infrared smoke modelling in CounterSim
R. H. Walmsley, B. Butters, Chemring Countermeasures (United Kingdom) With the ever-growing demand for increased realism in defence modelling and simulation, smoke modelling, which is computationally expensive, must be conducted on graphics hardware to enable execution at fast rates with good fidelity. While visual smoke simulation has been successfully implemented by many authors over recent years, infrared smoke modelling adds new requirements with additional challenges. This paper discusses the introduction of a Navier-Stokes staggered grid model into CounterSim, our countermeasures simulation software, highlighting the problems and benefits of using PC commodity graphics hardware for infrared applications. The grid model is introduced first, covering the initial co-located grid and then the staggered grid, reviewing the solution techniques for the incompressible Navier-Stokes equations. The methodology used to prevent mass loss and to model thermal cooling is incorporated. Additionally, the need to use an adaptive grid is explained, with the implications for management of data storage, particularly when mixtures of smoke products with widely differing properties may be present. The impact of these heterogeneous smokes on the simulation equations is also discussed. The focus is then shifted to rendering, highlighting the requirements for greater accuracy than the standard eight bits per channel of visual applications. Problems with alpha blending on graphics hardware are also discussed, and examples are presented that illustrate how changes in pixel format give rise to sometimes startlingly different end results. The paper concludes with examples showing real and simulated results, and reviewing methods for validation.

6738-23, Session 6
Time of flight range profiling using timecorrelated single photon counting
L. J. Sjqvist, M. Henriksson, P. Jonsson, O. K. Steinvall, Swedish Defence Research Agency (Sweden) Optical range profiling with high resolution can be accomplished using single photon counting time of flight techniques. Detection of optical surfaces with high resolution is of importance for several remote sensing applications. The use of laser pulses in the picosecond regime, single photon avalanche detectors and acquisition electronics with high timing resolution provides the tools for improving the range accuracy. In this work a system based on time of flight range profiling using time correlated single photon counting techniques is described. The system is intended to be used for optical range profiling in field experiments. Experimental results from measurements on test targets will be presented. The experimental data will be compared with theoretical models. Different schemes for extraction of range information from the experimental temporal profile are tested and the influence from turbulence effects is investigated.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

23

Conference 6739A: Electro-Optical Remote Sensing, Photonic Technologies and their Applications
Wednesday-Thursday 19-20 September 2007 Part of Proceedings of SPIE Vol. 6739A Electro-Optical Remote Sensing, Photonic Technologies and their Applications

6739A-32, Poster Session


Space analysis and detection of changes for follow-up of the sand-vegetation components in the Mecheria, Algeria, area
I. Haddouche, Univ. Abou Bekr Belkaid Tlemcen (Algeria) The Algerian steppe has become for a few years the theatre of an ecological and climatic imbalance. The intense degradation for this fragile medium (stranding wind erosion, overgrazing, clearing, salinisation ....) inducing the turning into a desertification require a better comprehension in order to see how to fight against this plague and to adapt an adequate installation to him . Thus, this work lies within the scope of the follow-up of the phenomenon of the turning into a desertification on a space with the heart of the high Oranian southem steppe plains, in fact the area of Mecheria. Techniques of numerical cartography, since the satellite image processing until the geographic information systems (SIG) for the realization of the charts sets of themes, being able to highlight this calamity. The mapping way is based on two types of criteria: the criteria related to the satellite picture: geographical location; colours; geometric forms; the criteria of environment: topography; vegetation; climatic data...etc. The use of the approaches based on the exploitation of the satellite data multi dates (1998 & 2004 ) of the sensor Thematic Mapper (TM ) of Landsat 5 permitted us to obtain a gathering of an interpretative photo maps and vegetation index which , in their turn , helped us to see the changes arrived in the medium , copiously regressive that progressive .

develop defense against Improvised Explosive Devices. It is recognized that the only technique, which is potentially capable to do it is trace detection by laser-based spectroscopy. It is vital to determine what kind of traces, vapors or micro-particles, is the best target and what kind of laser-based spectroscopy is the mostly suitable for such task. We arrived to conclusion that only a few explosives have sufficient vapor pressure to permit detection of gas clouds from afar, while even the most fastidious clean-up leaves microscopic residues, which adhere to surfaces that have come into contact with explosives. Thus fingerprints, smears and so on could be exploited for standoff detection. Several laser-based spectroscopic techniques, such as timeresolved luminescence, breakdown spectroscopy, Gated Raman spectroscopy and Second Harmonic Generation, (SHG) have been studied as a candidates for explosives microparticles standoff detection and identification. It was found that military grade RDX, C4 and Semtex have very strong luminescence, several explosives, mainly TNT, are characterized by a strong SHG response under IR excitation by 1064 nm and all explosives have characteristic breakdown spectra. Nevertheless, it appears that the main technique is gated UV Raman spectroscopy, in certain cases accompanied by luminescence measurements. We developed and tested a Raman system for the field remote detection and identification of explosives on relevant surfaces at a distance of up to30 meters.

6739A-03, Session 1
Remote sensing of hazardous materials by using mid-infrared quantum cascade lasers
F. Fuchs, C. Wild, B. Kirn, C. Mann, Q. Yang, W. Bronner, B. Raynor, K. Koehler, J. Wagner, Fraunhofer-Institut fr Angewandte Festkrperphysik (Germany) Quantum cascade (QC) lasers have been demonstrated to serve as robust and versatile laser sources for a variety of sensing applications. In contrast to classical laser spectroscopy of light molecules, the spectroscopic detection of more complex molecules, such as explosives, requires a much broader spectral tuning range of the laser employed. In this paper we demonstrate a QC laser module with a significantly extended tuning range of the incorporated Fabry-Perot QC laser. This was achieved by increasing the operating temperature range through the use of a temperature-variable diamond submount. The module comprises a QC laser mounted on the diamond submount with integrated heater element and temperature sensor, enabling a controlled change of the laser temperature between 77 K and 400 K with heating rates up to 2500 K/s. This way, Fabry-Perot QC lasers emitting at around 8 m could be tuned over a frequency range of 40 1/cm, which is sufficiently large to scan across characteristic absorption bands of e.g. explosives. The QC laser module has been incorporated in a set-up for remote detection of contaminations on test surfaces by backscattering spectroscopy. The favorable tuning characteristics allow us to easily tune the QC laser wavelength on- and off-resonance with the characteristic absorption band, i.e. to switch between absorption- and reference-mode. In a first laboratory test, surface contaminations with 10 g/cm2 TNT could be detected in imaging backscattering spectroscopy.

6739A-01, Session 1
Battlefield innovation: a case study of remote sensor development
J. A. Orson, National Reconnaissance Office (USA); T. N. Hague, U.S. Air Force Academy (USA) Evolving threats encountered by coalition forces during contingency operations drive the need for new and improved airborne intelligence, surveillance, and reconnaissance capabilities. In many cases, these items are prototype, one-ofa-kind aircraft systems that have not completed a traditional course of testing. Operation Horned Owl is one such experimental quick reaction capability, in which an experimental ground-penetrating radar and an off-the-shelf electro-optical and infrared sensor suite were rapidly integrated on a C-12 airframe and deployed to support Operation IRAQI FREEDOM. Utilizing sensor fusion and change detection, 21 Improvised Explosive Devices (IEDs) and 24 weapons caches totaling over 93,000 lbs of explosives were detected and disarmed, saving the lives of countless Iraqi citizens and coalition soldiers. This paper discusses the practical application of deploying unique sensor technology in a true operational environment - forward deployed combat arenas. Due to accelerated system employment, the aircraft received only basic stateside testing to characterize sensor performance. Remaining testing was conducted concurrent with combat operations to support ground forces. This nontraditional test approach and application of sensor technology highlights several areas that challenged successful system development. These areas along with lessons learned are provided to aid decision makers of future programs when balancing system maturation with meeting operational demand.

6739A-04, Session 1
Detection of mine-like objects in high-resolution aerial images
A. Zhuravlev, S. Ivashov, V. Razevig, Bauman Moscow State Technical Univ. (Russia) A method for recognizing small-size similar-looking objects in images with minimum prior information is proposed in the paper. The method consists of three basic stages: salient points detection, local descriptor calculation for every salient point, and vector set clustering to find a small distinguished subset of closest vectors. At the stage of salient points detection the scale-space representation is used. Spatial derivatives are calculated and normalized in such manner to yield high response to blob-like structures in images. At the second stage

6739A-02, Session 1
Laser-based spectroscopy for standoff explosives detection
M. L. Gaft, L. Nagli, Laser Detect Systems, Ltd. (Israel) Real time detection and identification of minimal amounts of explosives, at a standoff distance, is a major issue in efforts to

24

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6739A: Electro-Optical Remote Sensing, Photonic Technologies and their Applications
of image analysis the SIFT-descriptor is calculated for each salient point to form a multi-dimensional vector to be used in a clustering algorithm. At the clustering stage a dendrogram is built and the selection process begins by picking a couple of closest vectors. The closest vectors associated with image regions are supposed to represent similar-looking objects, e.g. mines or other human-made objects. Next, the algorithm attempts to add another object or vector cluster by checking the link consistency in the hierarchical tree. If the link doesnt consistent with the links below the branch is cut and the cluster below this link represents the set of similar looking objects. The developed algorithm was used on an artificial mine field. Aerial images of the field were taken by a consumer camera. It was shown that the described method was capable of detecting the mines. The developed algorithms are planned to be used in humanitarian de-mining operations. produced for broad variety of process-control applications using transmission, reflection, fluorescence and Ramanspectroscopy, but their spectral range is limited to transmission of silica fibers from 0.2 to 2.4m. Nowadays the longer Mid-IR wavelength range from 2 to 6-10m may be covered by Chalcogenide IR-glass CIR-fibers, while Polycrystalline PIRfibers from Silver Halides can cover the whole finger-print part of spectrum - the most informative on absorption bands of specific molecular vibrations (Fig.1). These specific bands are 100-1000 times more intensive and more narrow than the 2nd & 3rd harmonic bands at the shorter wavelengths <2m. Remote spectroscopy of chemical reaction, monitoring of exhaust or exhalation gases or invivo tissue analysis on molecular composition could be done by PIR-fiber probes coupling with different type of spectrometers - based on diffraction grating and detector arrays, FTIRinterferometers, IRLED, optical filters or using Tunable Diode or Quantum Cascade lasers. Design of immersion probes for liquids and touch probes for tissues is based on PIR-fiber coupling with a sealed ATRelements made from Diamond, ZnSe or other IR-materials. Choice of PIR-fibers in comparison with a brittle and toxic CIRfibers transmitting in 2-6m range is done they are very flexible, non-toxic, stable from 5 to 400K and transmit in much broader range from 3 to 17m. Transmission/Reflection T/R-probes for gases have been designed with IRfibers coupling of dual-pass optical cell with QCL or TDL-spectrometer. In addition to the review on IR-fiber coupled FTIR- and QCLspectroscopy systems the first FlexISpec spectrometer test data will be presented on the innovative prototype of (Fig.2) based on use of diffraction grating with IR-pyrodetector array and PIR-fiber probes. In difference with expensive FTIR- & QCLsystems FlexISpec developed is directed to more cost effective solution for real time process-control in chemical, petrochemical and pharmaceutical industries. Fig. 1. Attenuation spectra of IR-fibers Fig. 2. Spectrum of sun-flower oil measured with FlexISpec spectrometer in 5-7mm range with ATR-PIR-probe

6739A-05, Session 1
Lidar for shallow underwater target detection
M. Tulldahl, M. Pettersson, Swedish Defence Research Agency (Sweden) Small underwater objects such as vehicles and divers can pose threats to fixed installations and ships. For ships, these threats are present both at sea and in harbors. Shallow underwater targets, including drifting mines, are difficult to detect with acoustic methods and thus complementary methods are required. If an airborne platform is available, some of those targets could be detected by passive optical means. However, for sensing from a ship or from land, optical detection can be highly improved by use of a pulsed laser system. We present simulated data of importance for the design of a lidar system with low incidence angle with respect to the water surface. We also present our first experimental data from underwater target detection with an incidence angle of five degrees.

6739A-06, Session 1
Theoretical and experimental analysis of an equivalent model for the investigation of shallow landmines with acoustic methods
L. Capineri, A. Bulletti, G. Borgioli, P. Falorni, L. F. Masotti, S. Valentini, M. Calzolai, Univ. degli Studi di Firenze (Italy); C. Windsor, United Kingdom Atomic Energy Authority (United Kingdom) Acoustic methods have been recently investigated for the detection of shallow landmines. Some plastic landmines have a flexible case which can made to vibrate by an airborne excitation like a loudspeaker. The soil-mine system shows a resonant behavior which is used as a signature to discriminate from other rigid objects. The mechanical resonance can be detected at the soil surface by a remote sensing systems like a laser interferometer. An equivalent physical model of the minesoil system has been investigated having the known physical characteristics of mine simulants. The authors have designed and built a test-object with known mechanical characteristics (mass, elasticity, damping factor). The model has been characterized in laboratory and the results compared with the classic mass-spring loss oscillator described by Voigt. The vibrations at the soil surface have been measured in various positions with a micro machined accelerometer. The results of the simulations for the acceleration of the soil-mine system agree well with the experiment. The calibrated mine model is useful to investigate the variation of the resonance frequency for various buried depths and to compare the results for different soils in different environmental conditions.

6739A-08, Session 1
Adaptive photodetectors using wide-gap photorefractive sillenite crystals for vibration monitoring
I. A. Sokolov, M. A. Bryushinin, A.F. Ioffe Physico-Technical Institute (Russia) Detection of low-amplitude acoustic vibrations of real objects, such as ultrasonic transducers, composite materials, biological objects, metal production, is an important scientific and technique problem. Homodyne laser vibrometers are suitable for practical applications in this regard for their high sensitivity which is limited in principle only by shot-noise of the laser used. These vibrometers allow detection at a distance, with high spatial localization of the measurement region and wide frequency range. However, their utilization is restrained by several problems such as slow phase drifts in the interferometer arms due to environmental reasons, necessity of fine optical adjustment and suppression of laser amplitude noise. Photorefractive materials offer a novel elegant way to solve the problem of keeping operation interferometer point constant. The crystal replaces the conventional beamsplitter and can be controlled not only electrically but also optically, i.e. based on the principles of nonlinear optics. In addition, it can be multilayered. Such a multilayered adaptive beamsplitter is nothing else but a volume dynamic hologram recorded by the reference and signal waves. In this paper we present for the first time the results of utilization of wide-gap photorefractive sillenite crystals as adaptive photodetectors (AP) for vibration measurements. The operation of the adaptive system of non-destructive testing was studied for the real operation conditions (diffusely scattering objects). The operation principle of AP is based on the effect of the nonsteady-state photoelectromotive force (photoEMF). The mechanism responsible for the effect can be described as follows. Illumination of wide-gap semiconductor by an interference pattern produces a non-uniform excitation of free carriers (photoconductivity grating). Diffusion of the photo-excited carriers towards the dark regions

6739A-07, Session 1
Mid IR-fiber spectroscopy in 2-17m range
V. G. Artiouchenko, Fibre Photonics Ltd (United Kingdom); A. Bocharnikov, G. Colquhoun, C. A. Leach, Fibre Photonics Ltd. (United Kingdom); V. Lobachov, T. Sakharova, D. Savitskij, General Physics Institute (Russia) The latest development in IR-fiber optics enables to expand spectral range of processspectroscopy from 2m up to 17m (5.000 to 600cm-1). Up to now fiber systems are already

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

25

Conference 6739A: Electro-Optical Remote Sensing, Photonic Technologies and their Applications
leads to charge redistribution between deep traps in the photoconductor. A space charge field grating arises. Small vibrations of the light pattern excite an alternating current through the crystal. The results of measurements of small vibration amplitudes and resonant frequencies of the diffusely scattering objects including micro-electromechanical systems are presented. The presented adaptive interferometric system can be used for industrial applications. challenging is accurately locating them in space. We present optical sensors that can locate fast moving objects that penetrate into virtual optical screens. Each sensor is composed of several laser sources and detectors, and the 3-dimensional mapping is based on combinations of range measurements, triangulation and geometry. These short-range laser sensors typically have detection ranges from a few centimeters to several meters. The sensors can also operate under noisy external signals, clutter and sunlight. This is obtained by using signal modulation, fast switching of sources and advanced signal processing. Also, robust mechanical design allows for operation under extreme environmental conditions. We will present several sensor configurations, which are characterized by the optical properties of the sources and detectors, as well as their placement. The effects of the sensors design parameters on their detection properties and accuracies will be discussed. Also, we will compare the properties of these sensors with other proximity sensors.

6739A-10, Session 2
Waveform analysis of lidar data for targets in cluttered environments
G. Tolt, H. Larsson, Swedish Defence Research Agency (Sweden) In this paper we study the potential of using deconvolution techniques on full-waveform laser radar data for pulse detection in cluttered environments, e.g. when a land-mine is partly occluded by vegetation. A pulse width greater than the distance between the reflecting surfaces within the footprint results in a signal that is composed by overlapping reflections that may be very difficult to analyze successfully with standard pulse detection techniques. Deconvolution improves the chance of successful decomposition of such multi-modal pulses into the components corresponding to each reflecting surface. In addition, by deconvolving with the recorded emitted pulse, the influence of non-stationary or complex pulse shape is reduced. Experiments with various targets, e.g. land-mines, behind different kinds of obscuration were carried out and analyzed in terms of improved pulse extraction capability and distance accuracy. Some common deconvolution techniques, including the Richardson-Lucy algorithm, were tested. The effect of applying different smoothing filters to reduce the influence of noise on deconvolution and pulse extraction were also studied. It was found that in many situations, surfaces separated by only a small fraction of the pulse width could be revealed but that weak reflections in noisy signals were difficult to detect.

6739A-13, Session 2
Planning of a multiple sensor system for human activities space: aspects of iso-disparity surface
J. Chen, S. Khatibi, W. Kulesza, Hgskolan I Kalmar (Sweden) This paper presents a method for planning the position of multiple stereo sensors in an indoor environment. This is a component in an Intelligent Vision Agent System, IVAS. The IVAS is an automatic target detection, identification and information processing system for human activities surveillance. It consists of multiple sensors including their deployment and autonomous servo. The crucial problem for the system is to find the optimal configuration of sensors so that the features of environment and target objects are visible under the required constraints. We propose a new approach to dynamically adjust the multiple stereo pairs position, pose and baseline length in 3D space in order to get sufficient visibility and accuracy for surveillance, tracking and 3D reconstruction. The paper proposes the visibility constraints to plan the cameras positions and poses, a depth accuracy constraint to control the baseline length, and iso-disparity surface approach for camera zooming. A human geometry modelled by a tetrahedron where its two upper triangles representing the human fore front is introduced. The dynamically update precise geometry model may optimize the visibility for human self-occlusions. The minimum number of cameras stereo pairs necessary to cover the target space is optimized by an integer linear programming. The 3D simulations of reconstruction accuracy and the human activities space coverage problem were performed in Matlab. In this paper, we implement the camera auto-disposition, vergence and zooming according to different reconstruction accuracy requirements. The experiments show how the reconstruction accuracy depends on stereo pairs baseline length and zooming.

6739A-11, Session 2
Restoration algorithm and system performance evaluation for active imaging systems
J. Gilles, Club Laser et Procds (France) In the first part of this paper, we investigate the effect of the atmosphere turbulence in the case of active imaging systems. The classical filter used in the litterature is the temporal mean filter. This filter gives good results but the edges of the objects are blurred. In order to improve the performance, we propose to use a temporal median filter. We prove that, in a statistical point of view, the median filter is well adpated than the mean filter. Its main advantage is that it doesnt create new values which never appear during the acquisition process. Differents results will be showed based on the database acquired by the NATO-TG40 group. We also compare with the results obtained by local Wiener filtering and local Laplacian filtering. In the second part of the paper, we propose a new method to evaluate the performance of this kind of systems. Our method uses some ideas taken from the german TRM3 model. We define a new metric which uses directly the values measured at different locations in the image. This method permits to build a Modulation Transfert Function (MTF) like the one for passive systems. We conduct some experiments on the different systems used in the NATO-TG40 field trials and evaluate their performances. In order to validate this method, we compare it with the results we get by a theoretic MTF model for active imaging systems. We conclude this paper by giving some new perspectives in image restoration algorithm. We currently develop a new algorithm based on warping technics and the first results seems to be very promising.

6739A-14, Session 2
Lightweight compact optical correlator for spacecraft docking
A. Bergeron, P. Bourqui, Institut National dOptique (Canada); B. Harnisch, European Space Agency (Netherlands) Spacecraft docking is a critical operation in space rendezvous operations. Docking provides the opportunity to joint two vehicles in order to change crews and deliver resources to a spacecraft. One of the main challenges in docking is to perform real-time tracking of the docking point for a precise and rapid feedback to the control system in order to achieve reliable operations. Docking operations founds multiple earth counterpart applications. Many of these earth-based applications concern the use of robotic devices to grab a specific object. In these cases various location parameters of the object are needed, such as rotation angle, scale and position. INO has developed a prototype version of a compact lightweight optical correlator. This version provides a tool for various application evaluations. INO, in collaboration with ESA, studied the use of an optical correlator for various space applications such as rendezvous and docking, landing and star tracking operations. Optical correlator provides beyond real-time image

6739A-12, Session 2
Short-range detection and ranging of fastmoving objects
R. Oron, Y. Leitner, V. Palatnik, KiloLambda Technologies, Ltd. (Israel) Detection of fast moving objects is challenging. Even more

26

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6739A: Electro-Optical Remote Sensing, Photonic Technologies and their Applications
processing capabilities and is well suited for target identification and positioning purpose. Optical correlator is also lightweight and shows low power consumption. In this paper, the latest analyses of the docking application are presented. For evaluation purpose, video sequences of Soyuz docking the International Space Station (ISS) were used. The ISS docking point ISS was used as target to track the relative position, including scale and rotation angle, of the Soyuz vehicle compared to the ISS. This presentation will review the various parameters of this analysis, will present the results obtained with the video sequence. Alternative applications such as landing and star tracking will also be briefly addressed. classified and mapped. The accuracy of the classification was greater than 90%.

6739A-17, Session 3
Optical characterization of small surface targets
P. B. W. Schwering, TNO (Netherlands) Authors: TNO: PBW Schwering, RAW Kemp CSIR: DF Bezuidenhout, FPJ le Roux, RH Sieberhagen IMT: WH Gunter Abstract: Present-day naval operations take place in coastal environments as well as narrow straits all over the world. Coastal environments around the world are exhibiting a number of threats to naval forces. In particular a large number of asymmetric threats can be present in environments with cluttered backgrounds as well as rapidly varying atmospheric conditions. In these conditions the threat contrast may be low and varying, and the amount of background clutter can be severe. These conditions require the electro-optical means of detection and classification to be optimized in order to have more time to act against threats. In particular the assessment of classification means is an important issue. Beside short-range coastal paths, also long-range horizontal paths with variable atmospheric conditions are of interest. Therefore the small differences between types of vessels, can help us determining the classification of the vessel type. But also different small payloads and people on-board can be clues to the classification of the vessel. Operations in warmer environments, limiting the atmospheric transmission due to water vapour absorption are challenging. Understanding of the impact of the warmer environments on the optical characteristics of threats is of great importance. For this purpose a trial was planned to assess the optical characteristics of different types of small surface vessels in a coastal environment. During this trial a number of small targets were used during different parts of the day and night. Furthermore positional as well as atmospheric characterisation was performed as ground truth information. From this data a first analysis was performed showing strong intensity fluctuation in target as well as background signal levels. At longer ranges and in coastal environments these target signals may be well hidden within the background clutter. This data is essential to feed sensor performance models for the assessment of sensor performance in coastal environment.

6739A-15, Session 2
Coherent ladar 3D imaging with S3
J. R. Buck, A. Malm, A. Zakel, B. Krause, B. Tiemann, Lockheed Martin Coherent Technologies (USA) The Super-resolution Sensor System (S3) program is an ambitious effort to exploit the maximum information a laserbased sensor can obtain. At Lockheed Martin Coherent Technologies (LMCT), we are developing methods of incorporating multi-function operation (3D imaging, vibrometry, polarimetry, aperture synthesis, etc.) into a single device. The waveforms will be matched to the requirements of both hardware (e.g., optical amplifiers, modulators, etc.) and the targets being imaged. The first successful demonstrations of this program have produced high-resolution, three-dimensional images at intermediate stand-off ranges. In addition, heavy camouflage penetration has been successfully demonstrated. The resolution of a ladar sensor scales with the bandwidth as dR = c/(2B), with a corresponding scaling of the range precision. Therefore, the ability to achieve large bandwidths is crucial to developing a high-resolution sensor. While there are many methods of achieving the benefit of large bandwidths while using lower bandwidth electronics (e.g., an FMCW implementation), the S3 system produces and detects the full waveform bandwidth, enabling a large set of adaptive waveforms for applications requiring large range search intervals (RSI) and short duration waveforms. This presentation will highlight the three-dimensional imaging and camo penetration.

6739A-16, Session 3
The application of hyperspectral remote sensing in classification and mapping of the marshland aggressive plant cattail, typha latifolia
G. Liu, J. Allen, Clemson Univ. (USA); G. Kleppel, SUNY/Univ. at Albany (USA); J. Parkey, C. E. Campbell, Clemson Univ. (USA); K. Lu, Towson Univ. (USA) Cattail propagates both sexually and asexually, often in a rapid and aggressive manner. Such infestation develops dense monospecies stands, reduces overall vegetative biodiversity of wetlands, and significantly impacts the wetland ecosystem. It is essential to timely and accurately monitor and map the distribution, density, extent and the aggressive intensity of Cattail in order to help effectively control it and real timely monitor wetland dynamics. Traditional ground surveys are extremely time consuming and even inoperable because of the difficulty in access to the vast marshland areas. Until recently, there have been few available methods for using remote sensing to detect and isolate single plant species like Cattail. In utilizing hyperspectral aerial remote sensing data for this study, a hyperspectral library called the Strom Thurmond Institute Hyperspectral Library (STIHL) was created for the monitoring of wetland species in the Lower Hudson River Watershed areas in New York State. The STIHL fills the gap where spectral profiles of wetland plants are lacking in the current common spectral libraries such as the USGS Vegetation Spectral Library, the Jet Propulsion Laboratory Spectral Library and the Johns Hopkins University Spectral Library. In application of the STIHL, the authors derived the spectrally pure Cattail pixels and obtained the typical spectral signals to be used as the supervised classification training signals. An enhanced supervised classification method of Spectral Angle Mapper (SAM), with minimum threshold angles was developed and performed. Using the innovative STIHL in hyperspectral signal processing, the aggressive wetland plant Cattail was successfully identified,

6739A-18, Session 3
Influence of Saharan dust on astronomical observation
E. A. Siher, Faculty of Sciences & Techniques, Beni Mellal (Morocco) The site selection for the next generation of the telescopes (ELT: Extremely Large Telescope) need to be very pointed. On the other hand, in a previous study, we showed that the astronomical extinction (AE) can be extracted from the aerosol index (AI) by using the equation of correlation between these parameters (Siher et al. 2004). To use this new tool for the qualification of astronomical sites, we will present, in this paper, our last results concerning the AE and AI correlation. Also, we will sudy the influence of the refelctivity index on tha astronomical extinction.

6739A-19, Session 4
Exploitation of EO Technologies from the EMRS DTC
I. Clark, SELEX Sensors and Airborne Systems Ltd. (United Kingdom) No abstract available

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

27

Conference 6739A: Electro-Optical Remote Sensing, Photonic Technologies and their Applications

6739A-20, Session 4
Hybrid optoelectronic vector matrix multipliers using guided-wave and micro-optic components
V. A. Handerek, L. C. Laycock, A. G. McCarthy, BAE Systems plc (United Kingdom) There is a need for faster processing hardware to provide modern radar systems with advanced capabilities such as multiple hypothesis tracking, real-time clutter removal and space-time adaptive beamforming (STAP) for jammer nulling. One approach which may help to meet this need is to use analogue methods in parts of the signal processing chain using optoelectronics. The vector-matrix multiplier is a powerful optical processing architecture that potentially offers very large gains in computation speed, but has not so far become commercially successful. This paper reports investigation of a novel arrangement for this type of processor that is aimed at improving the prospects for commercialisation, using guidedwave and micro-optic components. This approach will assist miniaturisation of the processor and improve ruggedness and scalability to large matrices.

6739A-23, Session 4
Challenges in military remote sensing
K. L. Lewis, Electromagnetic Remote Sensing Defence Technology Ctr. (United Kingdom) The military commander ideally needs a multi-dimensional (spatial, temporal, spectral) view of the battle-space, with every object of relevance accurately located and characterised. An indication is also required of the level of threat posed by enemy manoeuvre to the success of operations. When a target is hit, sensors are required to determine the extent of functional damage produced and whether any collateral damage has resulted in the area surrounding the target. Remote sensing provides the means of achieving many of these goals, at standoff ranges sufficient to minimise risk to the sensors themselves. In the UK, the Electromagnetic Remote Sensing Defence Technology Centre (EMRS DTC) is concerned with research applicable to systems that gather and process electro-magnetic signals, propagating in free-space to a collecting aperture. Sensor are required to provide day and night all-weather capability, long-range operation and rapid large-area search while enabling the detection of low signature/camouflaged/ concealed targets without compromising the ability to operate covertly as far as possible. Challenges for the detection and identification of difficult targets are driving the development of multifunctional sensor systems. These include high-power and high-resolution radars operating from unstable platforms, compact sensors for un-manned air vehicles, affordable ESM techniques for complex operational theaters and discriminative electro-optic sensors supported by novel signal processing algorithms to provide real-time operational capability. Sensors are also required for the delivery of precision effect. This presentation will explain how the EMRS DTC is meeting many of these requirements, using examples derived from its own programme.

6739A-21, Session 4
A novel wavelength division multiplexed system architecture for high resolution sampling and digitisation of broadband microwave signals
J. M. Heaton, Filtronic Compound Semiconductors Ltd. (United Kingdom); P. Jiang, Univ. of Cambridge (United Kingdom); D. Gotch, R. Fawley, Filtronic Components Ltd. (United Kingdom); Y. Zhou, S. Clements, Filtronic Compound Semiconductors Ltd. (United Kingdom); I. H. White, Univ. of Cambridge (United Kingdom); R. Lang, Filtronic Components Ltd. (United Kingdom) Military applications such as electronic surveillance (ESM), intelligence (ELINT) and countermeasures (ECM) require realtime analysis of broadband RF signals, often with very high frequency resolution. Direct sampling and digitisation, followed by Fourier analysis and signal processing, will enable such signals to be analysed with much higher resolution than can be achieved with conventional microwave techniques. Direct sampling and digitisation of signals at frequencies above a few GHz is difficult to achieve electronically because the rise and fall times of electrical sampling gates are too long. This can be overcome by using a low-jitter optical sampling pulse train to measure the voltage on an electro-optic modulator. Optical sampling pulses can be very much shorter than electrical sampling pulses because of the large optical carrier frequency. In this paper we describe a novel multiple wavelength optical sampling system architecture for the real-time digitising of microwave signals between 1 and 20 GHz with a target resolution of more than 10 effective bits (\>60 dB spurious free dynamic range). We also describe our work on making highly linear electro-optic modulators for this and other microwavephotonic applications.

6739A-29, Session 5
Glow discharge detector (GDD) for terahertz and millimeter wave radiation detection and imaging
A. Abromovich, The College of Judea and Samaria (Israel); D. Rozban, N. S. Kopeika, Ben-Gurion Univ. of the Negev (Israel) Glow discharge plasma, deriving from direct current gas breakdown, was investigated experimentally in order to realize an inexpensive room temperature detector for this unique region. Preliminary experiments results at 3 mm (100 GHz) and around 1 mm (250 GHz) wavelength with several commercial plasma indicator lamps show good response. Furthermore, the experiments show that such glow discharge indicator lamps are preferable candidates for imaging applications. The construction of a Focal Plane Array (FPA) using them as pixels is advantageous since the cost of each is around $0.2-$0.5 per lamp, and the FPA images will be diffraction limited. The detection mechanism at microwave frequencies and lower frequencies was found to be microwave enhanced ionization collision frequency. In contradiction to this, our preliminary experimental results at 100 GHz and 250 GHz show that the detection mechanism is quite different, and appears to beat enhanced diffusion current which causes decrease of GDD bias current when radiated by such radiation.

6739A-22, Session 4
Fibre laser based ultra low phase noise LO generation for future radar
M. J. Staniforth, ESL Defence Ltd. (United Kingdom) A concept architecture has been designed to use the advantages of fibre based laser technology in the technique of deriving an RF LO source from ultra stable lasers. The system aim is to provide -160 dB/Hz phase noise at 1 kHz offset referenced to 10 GHz carrier frequency. The architecture consolidates the advances of key technologies to produce a compact, useable and rugged system, commensurate with more general defence applications as compared with solid-state free-space lasers. The design concept addresses the need for ease of proliferation by simplifying the system for lower cost and by focussing on key design parameters to reduce laser power requirements. It is hoped that the ultra-high stability of the pulsed laser source can be married to photonic digital waveform generators (Photonic DDS) to provide a comprehensive solution for advanced high capability radar requirements.

6739A-54, Session 5
Sparse aperture detection and imaging of millimeter sources via optical image-plane interferometry
I. Biswas, C. A. Schuetz, R. D. Martin, D. W. Prather, Univ. of Delaware (USA); M. S. Mirotznik, The Catholic Univ. of America (USA) We attempt to perform real time detection and direct high resolution imaging of millimeter blackbody sources using sparse aperture interferometry. We reject heterodyne technology for a multitude of factors including bulky equipment, cryogenic cooling, long integration times, and indirect imaging techniques. An alternate method is to convert the incoming millimeter waves into optical and perform optical image-plane interferometry in real time. This method is suitable for snapshot-imaging of short-

28

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6739A: Electro-Optical Remote Sensing, Photonic Technologies and their Applications
lived phenomena, often encountered in defense and security situations. The approach presented in this work utilizes a millimeter wave antenna array coupled to an optical interferometer which images directly on a detector array for image read-out, processing, and storage. Based on Kogans algorithm (Kogan 1997, 1998) to reduce the minimizes the maximum sidelobes of the point spread function, we choose an antenna array composed of two concentric hexagonal rings, such that the outer ring is ~3 times the inner ring. This design ensures more or less uniform and isotropic spatial frequency coverage, eliminating difficulties associated with resolving out structures whose spatial frequencies are in between that of the single aperture diameter and those of the baselines. The Fourier coverage of this array is the sum of the Fourier coverage of the outer ring plus that of the inner ring added to that of the baselines between the inner and outer rings. The need for delay lines is done away with by mounting all the apertures on the same plane. The incoming millimeter signals are fed through electro-optical modulators for upconversion onto an optical carrier, which can be readily captured, routed, and processed using optical techniques. The optical waves are fed via a fiber optic array onto a microlens array which is a scaled down version of the antenna array configuration. Then homodyne interferometry is performed. We reject pupil-plane (Michelson) interferometry based on a multitude of factors. The main drawback is that pupil-plane interferometers dont produce images but rather gives the information about the autocorrelation of the object. We instead use a classical image-plane interferometer (Fizeau) setup and direct detection is performed on a CCD array. Imageplane interferometry has its advantages. Unlike its pupil-plane cousin, a Fizeau interferometer is a true imaging device, where each beam is used to make an image of the object and are superimposed. Because Fizeau beam combiners work in the image plane, they dont suffer from ambiguities associated with the interpretation of visibility measurements. Also since the beams traverse the same paths and superpose, unmeasured phase changes do not creep in. In the design of the Fizeau interferometer, we preserve homothetic mapping, i.e., the entrance and exit pupils are replicas of one another, scaled only by a constant factor. This ensures direct imaging over a wide bandwidth with high angular resolution, high sensitivity, and a wide field of view. Since the Fizeau setup allows access to large fields, mosaicing wide fields is possible. References 1. Kogan, L. Optimization of an Array Configuration with a Donut Constraint. 1998b MMA Memo #212 2. Kogan, L. Optimization of an Array with a Topography Constraint. 1998b MMA Memo #202 3. Kogan, L. Optimization of an Array Configuration Minimizing Side Lobes. 1997 MMA Memo #171

6739A-56, Session 5
Sparse aperture millimeter-wave imaging using optical detection and correlation techniques
C. A. Schuetz, J. Samluk, C. Chen, D. W. Prather, Univ. of Delaware (USA) In previous publications, we have described a novel technique for millimeter-wave detection based on optical upconversion, carrier suppression, and photodetection. Using these techniques, we have been able to achieve NETDs as low as 1 K /Hz in both 35 GHz and 95 GHz atmospheric transmission windows. These results were obtained without the use of millimeter-wave LNAs or cryogenic cooling, which have previously been requirements for reaching these performance levels. In this proceeding, we detail efforts to create a scanning singlepixel imager based on these results. The design of the reflective optical system is described and sample images shown. The imager performance is shown to be diffraction limited. With moderate LNA gains, we have been able to demonstrate sufficient sensitivity for indoor imaging without illumination. These images have yielded interesting phenomenological effects for indoor environments that are detailed herein.

6739A-24, Session 6
Electronic scanning for passive millimetre wave imaging
N. A. Salmon, QinetiQ Ltd. (United Kingdom) This paper reviews the use of electronic scanning technology for the passive millimetre wave imagers. Looking at the mathematics of beam-forming, the developments in technology and the end user requirement indicates a new architecture of imager might offer an attractive prospect. The potential applications are reviewed and technological and physics considerations are given in the performance prediction with respect to the frequency. A programme to build a small demonstrator of such a beam-forming system is overviewed.

6739A-26, Session 6
Stand-off detection with a heterodyne receiver at 0.8 THz
H. Hbers, A. D. Semenov, H. Richter, A. Smirnov, U. Bttger, Deutsches Zentrum fr Luft- und Raumfahrt e.V. (Germany) Suicide bombers and hidden bombs or explosives have become serious threats especially for mass transportation. Until now there exists no established system which can be used against these threats. Therefore new technologies especially for standoff detection of threats are required. Terahertz (THz) rays offer an alternative inspection method, which can cope with these new challenges. Major advantages of THz radiation as compared to other spectral regions are the possibility to penetrate clothes and that THz radiation is not harmful for human health. In this contribution the design and results of a THz stand-off detection system will be presented. The sensor is based on active illumination of the object and sensitive heterodyne detection of reflected and backscattered radiation. The system operates at 0.8 THz. A THz laser is used for illumination and a superconducting hot-electron bolometric mixer for detection. The local oscillator required for heterodyne detection is a multiplied microwave source. The optical system is designed to allow for stand-off detection up to 20 m with a spatial resolution less than 2 cm. We will present the design of the system and its subcomponents as well as the overall performance and discuss future improvements, which are necessary for a practical implementation of this technology.

6739A-55, Session 5
Enhancing millimeter-wave data by adapting visible-range image processing techniques
F. E. Ortiz, E. J. Kelmelis, P. F. Curt, EM Photonics, Inc. (USA); K. A. Krapels, Office of Naval Research (USA); D. W. Prather, EM Photonics, Inc. (USA) Millimeter wave imaging provides the ability to see through various obscurants including smoke, fog, dust, and sand. However, due to the size of the wavelength and quality of current imaging systems, low-resolution imagery is produced. In this paper, we explore two techniques for enhancing image quality: speckle imaging and super-resolution. Both have been previously used for visible and infrared applications and we discuss their extension to millimeter wave imagery. The key limitation to their use is amount of processing required to enhance images in real time. The focus of our work is on the acceleration of these algorithms for real-time processing in a low-power, light-weight device. We provide a discussion of the algorithms used, their application to millimeter wave imaging, and their implementation.

6739A-27, Session 6
Development of high power HBV multipliers for millimeter wave applications
J. Stake, Chalmers Tekniska Hgskola (Sweden) and Wasa Millimeter Wave AB (Sweden); T. Bryllert, Chalmers Tekniska Hgskola (Sweden) and Chalmers Tekniska Hgskola (Sweden); J. Vukusic, Chalmers Tekniska Hgskola (Sweden) and Wasa Millimeter Wave AB (Sweden) In order to provide compact solutions and sufficient power levels

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

29

Conference 6739A: Electro-Optical Remote Sensing, Photonic Technologies and their Applications
at terahertz frequencies for future heterodyne imaging and radar systems there is a strong need to develop broadband, highly efficient, multipliers with a high multiplication factor (\>3) and improved power handling capability. The Heterostructure Barrier Varactor (HBV) diode is a strong candidate as a compact and efficient terahertz source. An HBV quintupler (x5) with a state-of-the-art conversion efficiency of 11% has been demonstrated at 100 GHz (Bryllert et al.). HBV triplers (x3) have been shown to provide \>200 mW at 114 GHz (Vukusic et al.). At short millimetre wavelengths, corresponding to a frequency of 200-300 GHz, 10 mW and at least 10% efficiency has been demonstrated (Xiao et al.). In terms of output power, the best results have been achieved using a filter circuit on AlN instead of quartz. The AlN substrate provides a better heat-sink for the flip-chip mounted diode. According to detailed analysis, there is room for substantial improvements in terms of output power in the millimetre wavelength range with a proper electrical and thermal design. Our predictions suggests an output power of \>0.4W at 100 GHz and \>50mW around 200 GHz. Such compact high power sources will play an important role in THz-radars for standoff detection. We will present the development of an integrated high power HBV tripler to 94 GHz and a HBV quintupler to 210 GHz. Both circuits will be optimised for an input power of 1-2 Watt (at 31-42 GHz). diffraction grating. In the former, the position of a mirror is moved allowing a mode-hop-free tuning of 0.4 cm-1, comparable to the free spectral range of the Fabry-Perot resonator, and a coarse tuning of 2 cm-1 at the center frequency of 158 cm-1. In the latter, a grating in Littrow configuration is employed for achieving a broadband mode-hop-free tuning. In both cases an anti-reflection coating has been developed to reduce the reflectivity of the laser facets below 1%.

6739A-28, Session 6
Terahertz heterodyne focal plane and imaging array
S. Cherednichenko, V. Drakinskiy, Chalmers Tekniska Hgskola (Sweden); H. Hbers, Deutsches Zentrum fr Luftund Raumfahrt e.V. (Germany); J. Krieg, Observatoire de Paris (France) Terahertz waves have been recognized as an important frequency range for both imaging and spectroscopic applications in the field of homeland security. In this paper we present development of a heterodyne focal plane array based on superconducting hot-electron bolometer (HEB) mixers. Previously, HEB mixers have been investigated for application in radio astronomy, where a sensitivity of 10 times the quantum limit has been achieved. Besides high sensitivity, superconducting HEB mixers have a large dynamic range. Due to a rapid improvement of cryo coolers, currently it is possible to consider superconducting detectors for autonomous systems, which would not require consumables like liquid helium, providing short cooling time, and reliable long term operation. We have built a 16 pixel mixer camera suitable for a heterodyne system for frequencies from 300 GHz to a few THz. HEB mixers are made on a single wafer, and they can be easily replaced in the camera depending on the choice of frequency. The mixers are integrated with either a narrow band resonance antenna (for a terahertz radar application) or with a broadband antenna (for thermal imaging). Low local oscillator requirements for HEB mixers (<1000nW) make it possible to use a solid state multiplier LO source to drive the entire array. We used mini mirrors for the RF coupling to the mixers. With this approach the number of pixels in the camera can be easily extended to more than 100.

6739A-30, Session 6
Tunable THz quantum cascade lasers with external cavity
J. Xu, L. Mahler, R. Green, A. Tredicucci, F. Beltram, Scuola Normale Superiore di Pisa (Italy); D. B. Fenner, Epion Corp. (USA); J. Hensley, M. G. Allen, Physical Sciences Inc. (USA); H. E. Beere, D. A. Ritchie, Univ. of Cambridge (United Kingdom) Terahertz photonics is an emerging field, receiving considerable attention for potential applications in various areas, such as military and security scanning, sensing and imaging, trace gases detection, and communications. Terahertz quantum cascade lasers (THz QCLs) with single frequency, narrow linewidth, and wide tuning range are particularly required for spectroscopic applications. We present experimental results on single frequency tunable 4.7 THz QCLs in external cavities using two different techniques: a) tuning by varying the cavity length and b) tuning by a

30

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6739B: Optically Based Biological and Chemical Detection for Defence
Tuesday 18 September 2007 Part of Proceedings of SPIE Vol. 6739B Optically Based Biological and Chemical Detection for Defence

6739B-34, Session 6
Microfluidics-based integrated airborne pathogen detection systems
M. A. Northrup, Microfluidic Systems (USA) Microfluidic Systems is focused on building autonomouslyoperating, microfluidic platforms that interface front-end sampling devices, such as air collectors, for detection and identification of pathogens. The system uses injection-molded, low-cost, microfluidic circuitry to process large to small liquid volumes for complex reagent metering, mixing, sample processing and purification, and biochemical analysis, specifically for pathogens including bacteria, viruses, and toxins. MFSI is the prime contractor on two programs for the US Department of Homeland Security: BAND (Bioagent Autonomous Networked Detector) and IBADS (Instantaneous Bio-Aerosol Detection System). The goal of BAND is to develop an autonomous system for monitoring the air for known biological agents. This consists of air collection, sample lysis, sample purification, detection of DNA, RNA, and toxins, and a networked interface to report the results. For IBADS, MFSI is developing the confirmatory device which must verify the presence of a pathogen with 5 minutes of an air collector/trigger sounding an alarm. Updates on instrument designs, system testing, and biological assay results from both BAND and IBADS will be presented.

6739B-36, Session 6
UK small scale UVLIF lidar for stand-off airborne BW detection
K. L. Baxter, V. E. Foot, S. J. Barrington, M. J. Castle, Defence Science and Technology Lab. (United Kingdom) Detection of biological warfare agents must be achieved as far upwind of their potential target as possible to provide the time necessary to adopt an effective protective posture. A small-scale fluorescence lidar has been designed and constructed by Dstl. The active element is a solid state Nd:YAG laser the frequency of which is quadrupled to yield 266nm excitation of 9ns pulses with 40mJ energy. Fluorescence is collected from 300 to 500nm and is divided into 10 channels to enable discrimination between common fluorescent interferent aerosols and biosimulants. The system is able to scan at rates up to 2o/sec and collects spectra from clouds which can be classified into biological simulant or interferent classes by the discrimination algorithm. In 2006, the fluorescence lidar was taken to US Army Dugway Proving Ground, Utah where it operated in trials specifically for stand-off detection systems for biological weapons. The system also participated in trials held in the Standoff Ambient Breeze Tunnel (S-ABT) and the Joint Ambient Breeze tunnel (JABT) at Dugway Proving Ground. This well characterised cloud data from biological simulants and interferents has been utilised to train the discrimination algorithm and to assess the systems limit of detection and discrimination ability.

and conveyed to a dual wavelength ultra-violet laser induced fluorescence (UVLIF) optical trigger used to discriminate potential threat particles from non-threat. The UVLIF particle classification information then triggers an electrostatic deposition mechanism to deposit only those potential bio-threat particles onto a stainless steel substrate. Non-threat particles are discarded with the exiting airflow. Finally, if desired, the dry sample can be transferred into a small quantity of liquid. Having separately shown feasibility for each of the constituent technologies, the goal for the most recent development effort has been the integration and optimization of these technologies into a unit capable of producing highly enriched particulate samples from ambient air. Several key technical and engineering challenges were overcome during the course of this development including a unique solution for compensating particle velocity dispersion within the airflow, and the development of a real-time signal acquisition & detection algorithm for determining material type on a particle by particle basis at rates greater than 2,000 particles per second. Also, the introduction of a robust method for transferring deposited particulate into a 50ul liquid sample has been demonstrated. This paper will describe the overall BioACE architecture and then concentrate on the various component and system design tradeoffs required to optimize sample enrichment performance. A system performance model will be presented along with analysis of the optical components and electronic signal processing needed for achieving high sample enrichment. Experimental methods and data will be described to support the performance model.

6739B-40, Session 7
A study of aerosol particle sorting to provide enriched samples for improved bio-threat analysis
J. T. McGinn, T. A. Pletcher, D. Keller, Sarnoff Corp. (USA); V. Sivaprakasam, A. L. Huston, J. D. Eversole, Naval Research Lab. (USA) Dynamic sorting of threat particles from background clutter and dust prior to collection for analysis can provide substantially enriched samples with the advantages of greater analytical accuracy in shorter periods of time. The conceptually simple process of capturing only threat particles and rejecting background in fact requires sophisticated particle detection and classification, timing, capture and final threat identification subsystems operating in concert. The effectiveness of the process is also strongly influenced by the operational conditions including threat and background loads, as well as the time allotted for sample collection. The requirements of the final threat identification system will dictate the form factor for the collected sample. This will dictate if collection is to be done dry or into a liquid as well as the threat concentration and distribution on the final sample (e.g., microcopy based methods require a minimum number of threat particles in a field of view). This paper explores the theoretical parameters of dynamic sorting with particular attention paid to particle loading and coincidence issues as well as the affects of the capture timing and capture efficiency on the enrichment of threat particles in the final analytical sample. Enrichment factors, a common figure of merit for sorting systems, will be shown to be a poor indicator of performance unless standard operating conditions are used for comparison between different systems. Examples will be drawn form the literature and work done by the authors using UV-Laser Induced Fluorescent detection and electrostatic collection subsystems.

6739B-37, Session 6
Recent Advances in the Development of a Novel Aerosol Sorting and Deposition System for Biothreat Sensing Applications
T. A. Pletcher, Sarnoff Corp. (USA) Sarnoff Corporation and the Naval Research Laboratory, through support from the U.S. Department of Homeland Security, are developing an automated, high throughput, bioaerosol enrichment system designed for use as part of a biological-threat protection methodology. The Biological Aerosol Capture and Enrichment (BioACE) system is a bio-aerosol sampling device that combines several unique technologies to create physically enriched aerosol samples that can subsequently be interrogated by any number of bio-threat detection techniques. Briefly described, BioACE uses an airto-air concentrator to highly concentrate the ambient aerosol. This aerosol is then collimated into a narrow particle stream

6739B-41, Session 7
Performance analysis for stand-off biological warfare agent detection
O. K. Steinvall, P. Jonsson, F. Kullander, Swedish Defence Research Agency (Sweden) Stand-off biological warfare agent detection has been identified as a promising technology to detect and classify a biological

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

31

Conference 6739B: Optically Based Biological and Chemical Detection for Defence
warfare agent containing cloud before it reaches the position of the lidar or a nearby lying area under protection. Some important receiver operating characteristics are the sensor response time, the sensitivity, the probability of correct detection and the false alarm rate as a function of range. This paper is analyzing some lidar concepts with respect to the choice of laser source and receiver system and describes the dependence on weather and other atmospheric parameters. Both elastic backscattering and fluorescence are treated. enhancing the overall sensitivity of detection. There is an important distinction between the behavior of a single resonant tag sensor and the behavior of a coupled array of resonant sub-elements. Just as a crystals behavior is not just a simple linear superposition of the individual behaviors of isolated single atoms, so to can the collective response of the net can be more than simple addition would suggest. These structures can exhibit many of the properties o a crystals and waveguides and are quite similar to a variety of frequency selective surfaces and meta-materials currently under study. It is this collective behave we hope to harness to increase the chemical sensitivity of the overall surface beyond that of a single tag.

6739B-57, Session 7
LBAS: Low Cost Biological Aerosol Sensor
R. K. DeFreez, ICx Mesosystems (USA) No abstract available

6739B-44, Session 8
Fibre optic system for detection of uranyl ions in the solution phase
N. W. Hayes, C. J. Tremlett, A. M. Shaw, EvanesCo Ltd. (United Kingdom); P. J. Melfi, J. L. Sessler, The Univ. of Texas at Austin (USA) Detection of radiological hazards in the solution phase using conventional means is difficult and impractical. A new approach is required to provide a reliable, specific and low cost method of protecting sensitive national assets such as water supplies from a terrorist dirty bomb attack. Fibre optic sensors provide the required speed of response, the optical platforms are mature and of relatively low cost with proven reliability in the field. This paper describes the combination of a low cost sensor platform and smart sensor molecule (Isoamethyrin) for the selective determination of uranyl and other actinide species in water at sub ppm levels. Isoamethyrin is a synthetic porphyrin which has been demonstrated to show high selectivity for uranyl ions with an associated colour change on complexation. Fibre optic sensors are created by revealing an evanescent wave in a section of the fibre and covalently bonding the isoamethyrin to the fibre surface in this region. Colour changes occurring as a result of interaction between isoamethyrin and uranyl ions are monitored over 3 wavelength ranges covering the red, green and blue regions of the visible spectrum, Figure 1. Sensors created in this manner were found to be fast responding (<5s), sensitive (detection threshold <500ppb), specific (response restricted to certain actinides and lanthanides) and low cost.

6739B-58, Session 7
Construction of an optical particle counter with sub- and super-micron counting capability
R. J. Greaney, O. I. J. Ryan, C. D. ODowd, S. G. Jennings, National Univ. of Ireland/Galway (Ireland) No abstract available

6739B-42, Session 8
Advanced applications to security of IR smart microbolometers
C. Corsi, Consorzio C.A.R.S.O. (Italy); N. Liberatore, Consorzio C.R.E.O. (Italy); S. Mengali, Consorzio C.A.R.S.O. (Italy); R. Viola, Consorzio C.R.E.O. (Italy) Most of the sensors developed up to now for detecting the presence of toxic gases show the limit of a high rate of false alarm and the risk to be excited and alarmed by a wide and undetermined types of gases , while the main requirement for these alarm systems should be a low rate of false alarm and overall the capability of identifying the gas species to activate the necessary and urgent countermeasures. One of the most reliable technique is Spectroscopy especially in the IR field where most of the toxic gases ( natural or man made) have absorption lines: we developed a low cost, selfcontained , multi-sensors integrated systems capable of detecting and identifying species of toxic gases . Improvements will be analyzed for open path sensors systems, consisting of an array of 64 detectors of low cost smart architecture, each coupled to its own narrow-band optical filter and focusing lens, to measure radiation emitted from a thermal source at a distance of a few tens meters. Hollow-fiber-based sensors in which a fiber acts both as a miniature gas cell with an optical path length of several meters, and as a flexible optical coupler between the source and the multi-spectral detector will be analyzed. The system with a large area IR micro-emitter and with an IR FPAs large areas Smart Sensors will be presented and discussed. Possible applications and performances with future system evolution will be outlined.

6739B-45, Session 8
Chemical agent detection and identification with a hyperspectral imaging infrared Sensor
V. Farley, M. Chamberland, P. Lagueux, A. Vallires, A. J. Villemaire, J. Giroux, Telops, Inc. (Canada) Standoff detection, identification and quantification of chemicals in the gaseous state are fundamental needs in several fields of applications. Additional required sensor characteristics include high sensitivity, low false alarms and high-speed (ideally realtime) operation, all in a compact and robust package. The thermal infrared portion of the electromagnetic spectrum has been utilized to implement such chemical sensors, either with spectrometers (with none or moderate imaging capability) or with imagers (with moderate spectral capability). Only with the recent emergence of high-speed, large format infrared imaging arrays, has it been possible to design chemical sensors offering uncompromising performance in the spectral, spatial, as well as the temporal domain. Telops has developed an innovative instrument that can not only provide an early warning for chemical agents and toxic chemicals, but also one that provides a Chemical Map of the field of view. To provide to best field imaging spectroscopy instrument, Telops has developed the FIRST, Field-portable Imaging Radiometric Spectrometer Technology, instrument. This instrument is based on a modular design that includes: a high performance infrared FPA and data acquisition electronics, onboard data processing electronics, a high performance Fourier transform modulator, dual integrated radiometric calibration targets and a visible boresighted camera. These modules, assembled together in an environmentally robust structure, used in combination with Telops proven radiometric and spectral calibration algorithms make this instrument a world-class passive standoff detection system for chemical imaging. This paper presents chemical detection and identification results obtained with the FIRST sensor.

6739B-43, Session 8
Resonant chemical surveillance tags
D. P. Neikirk, Y. S. Park, The Univ. of Texas at Austin (USA) Results will be reported from efforts to develop a self-contained wireless, un-powered resonant chemical surveillance tags for sensitive detection of chemical vapors, leading to a platform for use in vapor detection without the requirement of a preconcentrator. We are extending our work on single resonant rf (radio frequency) tags that have already demonstrated the ability to sense a variety of conditions. These sensors are interrogated in a non-contact manner and rely on resonance to transmit information about the state of tag. These devices are wirelessly interrogated using a wide range of radio frequencies, depending on the specific design of the tag. We have begun testing of a resonant chemi-tag coupled to a photodiode in contact with chemiluminescent reaction to provide the transduction path. We will also discuss a net of purely passive, and further, passively coupled, resonant tags exhibiting collective phenomena forming a single sensing surface, with the goal of

32

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6739B: Optically Based Biological and Chemical Detection for Defence

6739B-46, Session 8
Printed photonics for lab-on-chip applications
R. Pieler, E. Freder, M. Sonnleitner, NANOIDENT Technologies AG (Austria) It has been long recognized that the integration of optical readout technology directly onto a chip would rapidly promote the use of lab-on-a-chip technology. However, no one has been able to create a workable solution to date. There have been multiple academic approaches that integrate optical components like light-sources, sensors, optical filters or lenses onto a single chip. However, each single component is created with a different fabrication technology and therefore the necessary integration of whole readout systems onto a chip is nearly impossible from the technological point of view and also too costly to be commercially viable. Here we present the solution to this problem. NANOIDENT can print optical readout capabilities directly onto lab-on-chip platforms. The novel concept allows for the first time to integrate complete illumination and detection capabilities onto microfluidic-based devices. Absorption and chemiluminescence readout data are presented that show the high sensitivity and dynamic range of the integrated photonics.

breakdown before the electric field reaches the level of the bulk avalanche breakdown field, which has hampered the development of III-nitride based APDs. In this study, we investigate the growth and characterization of GaN and AlGaNbased APDs on GaN substrates. Epitaxial layers of GaN and AlGaN p-i-n ultraviolet avalanche photodiodes on GaN substrates were grown by metalorganic chemical vapor deposition (MOCVD). Improved crystalline and structural quality for GaN and AlxGa1-xN epitaxial layers were achieved by employing optimum growth parameters on low-dislocationdensity bulk GaN substrates in order to minimize the defect density in epitaxially grown materials. GaN and AlGaN APDs were fabricated into 30m- and 50m-diameter circular mesas. The forward I-V characteristics and low reverse-bias voltage (up to -100V) I-V characteristics were measured. GaN APDs with UV photocurrent gains \>104 have been demonstrated. For AlGaN APDs, the avalanche gain reaches a maximum value of ~50 at a voltage of 86.75V. For AlxGa1-xN APDs with higher Al-content, the crack-free growth of thick AlxGa1-xN on a bulk GaN substrate has been investigated by employing various strain management layers. Growth of AlxGa1-xN PIN structures with higher Al-content and APD device performance of will these devices be reported.

6739B-47, Session 9
New photonic molecular immobilization technology and its use for nanolabeling, ultrasensitive microarray sensing technology and new biomarkers discovery
S. B. Petersen, M. T. C. A. Neves-Petersen, M. Duroux, E. Skovsen, L. Duroux, Aalborg Univ. (Denmark) Photonic induced immobilisation is a novel technology that results in spatially oriented and localised covalent coupling of biomolecules onto surfaces (industrially and biomedically relevant proteins), preserving their functional properties. The reaction mechanism behind the reported new technology involves light induced breakage of disulphide bridges in proteins upon UV illumination of nearby aromatic amino acids. Arrays of bio-molecules have been created with a high degree of reproducibility, where the spatial dimensions of individual spots are defined by the area of the focal spot of light on the sensor surface. The ultimate size of the immobilized spots is defined by the focal area of the UV beam, which for a diffraction limited beam can be less than 1micrometer in diameter. Biosensing with our new technology has been successfully implemented in order to create arrays of immunoglobulins aiming immuno detection. Given that suitable protein markers exists for all relevant diseases it is entirely feasible to test for a range of disease indicators in a single test. Few micrometer spotsize allows for a virtually unlimited number of protein spots. It opens the potential for the development of medically relevant multipotent arrays, all located in less than 1cm2 sensor surface. The pattern of immobilized molecules is not restricted to conventional array formats. We have written complex protein patterns using bitmaps at high resolution onto substrates. This new technology produces radically new microarray sensing technology and photonics based watermarking technology by the combination of highest level knowledge on the effect of ultraviolet (UV) light on proteins and subsequent knowledge of the reaction between light activated molecules and surfaces, and has clear potential for biomedical, bioelectronic, surface chemistry, security markers production, nanotechnology and therapeutical applications (bioactive carrier nanoparticles).

6739B-49, Session 9
Novel interference film sensor for meet degradation
M. Bauer, F. Pittner, N. Ibrisimovic, Vienna Univ. (Austria) Customer Satisfaction has globally deteriorated through multiple meat scandals. In order to restore customer satisfaction the food supply chains have been made transparent from manufacturing level to the end customer. However the documentation completely relies on printed labels, often misused and reprinted to feed back old meat into the food supply chain. Therefore we have developed an optical thin film sensor chip able to detect the decay of food through a specific colour change. The design of the sensor relates to the phenomenon of anomalous absorption, which can best be described as a thin film enhanced absorption. A metal cluster film positioned at a well defined distance to a smooth metal surface shows that the minimum of spectral reflectivity strongly depends on the thickness of the interlayer: This setup represents a special kind of reflection interference filter. In such a sensor setup we have integrated a biodegradable polymer which is degraded by the same enzymes and at the same rate as food decay will happen. The degradation of the polymer results in reduction of the film thickness and thus in a specific change of the color.

6739B-50, Session 9
Optical sensor array platform based on polymer electronic devices
M. M. Koetse, H. F. M. Schoo, P. A. Rensing, R. B. A. Sharpe, G. T. van Heck, B. Allard, TNO (Netherlands); R. De Zwart, R. J. Houben, E. Enting, TNO Science and Industry (Netherlands); S. J. F. van Veen, TNO (Netherlands) Monitoring of personal wellbeing and optimizing human performance are areas where sensors have only begun to be used. One of the reasons for this is the specific demands that these application areas put on the underlying technology and system properties. In many cases these sensors will be integrated in clothing, be worn on the skin, or may even be placed inside the body. This implies that flexibility and wearability of the systems is essential for their success. Devices based on polymer semiconductors allow for these demands to be met, since they can be fabricated with thin film technology. The use of thin film device technology allows for the fabrication of very thin sensors (e.g. integrated in food product packaging), flexible or bendable sensors in wearables, large area/distributed sensors, and intrinsically low-cost applications in disposable products. With thin film device technology a high level of integration can be achieved with parts that analyse signals, process and store data, and interact over a network. Integration of all these functions will inherently lead to better cost/ performance ratios, especially if printing and other standard polymer technology such as high precision moulding is applied for the fabrication.

6739B-48, Session 9
High-performance GaN and AlxGa1-xN ultraviolet avalanche photodiodes grown by MOCVD on bulk III-N substrates
R. D. Dupuis, D. Yoo, J. B. Limb, J. Ryou, Y. Zhang, S. Shen, P. D. Yoder, Georgia Institute of Technology (USA) Wide-bandgap GaN-based avalanche photodetectors (APDs) are important for photodetectors operating in the solar-blind UV spectral region, < 290 nm. The growth of GaN-based heteroepitaxial layers on lattice-mismatched substrates such as sapphire and SiC, a high density of defects is introduced, thereby causing device failure by premature microplasma

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

33

Conference 6739B: Optically Based Biological and Chemical Detection for Defence
In this paper we present an optical transmission sensor array based on polymer semiconductor devices made by thin film technology. The organic devices, light emitting diodes, photodiodes and selective medium chip, are integrated with classic electronic components. Together they form a versatile sensor platform that allows for the quantitative measurement of 100 channels and communicates wireless with a computer. The emphasis is on the sensor principle, the design, fabrication technology and integration of the thin film devices. Finally, we show the first results of the development of an ultra thin, flexible reflection mode sensor based on the same building blocks.

6739B-53, Poster Session


Chemical warfare and industrial agents detection and identification by LIBS and CRDS
A. Pailloux, I. Debecker, Commissariat lEnergie Atomique (France); D. Romanini, M. Triki, M. Chenevier, Univ. Joseph Fourier (France) The French Atomic Energy Commission (CEA) is developing optically based techniques to detect and identify chemical agents. Detection sensitivity and molecule discrimination are studied. In parallel, efforts are made to develop rugged and compact experimental designs that can be used for field measurements. Laser Induced breakdown spectroscopy (LIBS) is a surface analysis determining the elemental composition of the sample without any preparation and contact. This technique is now being adapted to detect and measure a surface contaminated with a chemical warfare agent, which is an organic molecule containing a few non-metallic atoms. Detection experiments of sarin-like molecules characterized by phosphorus and fluorine atoms with an elemental ratio equal to one will be presented. These molecules were deposited on samples coming from the Parisian subway. Cavity ringdown spectroscopy (CRDS) or cavity enhanced absorption spectroscopy (CEAS) are optical absorption based technique, sensitive, with rapid response time, and well adapted to detect small gaseous molecules. These may include industrial toxic compounds, such as hydrogen fluoride, which may also be used by terrorist groups. Hydrogen fluoride (HF) is measured by CRDS in the ppb range with a very compact device developed by the University of Grenoble. Measurements will be presented, both in a laboratory experiment, using an HF permeation oven, and in a field experiment, near an HF outgassing plant.

6739B-51, Session 9
Combination of a Biological sensor using light emitting bacteria and a UV Spectrometer probe for Homeland Security and Drinking Water Safety
J. W. Appels, microLAN B.V. (Netherlands) Analytical Monitoring systems can only cover a part of the range of potential contamination in water. Therefore the interest in the last years in on-line Biomonitors has increased significantly. This has resulted in the introduction of new Biomonitors which can provide a (near) real-time information on water quality. They can also be used for drinking water protection against Intential Contamination. But still no universal monitor is available which protects against all kinds of threats and a combination of systems and techniques is required if all criteria are to be met. This can not be realised with the use of a single monitor. However, the combination of complementary systems into a single integrated monitoring platform will greatly enhance the applicability of real time monitoring devices for water quality monitoring. For this the TOXcontrol, a biological toxicity monitor (using luminescence bacteria) with the S::CAN spectro::lyserTM, an analytical sensor (a submersible UV-VIS spectrophotometer probe) will be combined and a software tool will help the user to make better judgements and give more insight information to make decisions. This combination allows verification of alarm signals from one instrument with the signal of others reducing false alarm rates. After a study it was concluded that both the UV-VIS spectrophotometer and the Toxicity Biomonitor are valuable sensors in an Water Security Monitoring System, capable of detection a broad range of contaminants at low concentrations. Results of both lab and field tests for Homeland Security and Drinking Water Safety monitoring, will be presented

6739B-52, Poster Session


New optical cylindrical microresonators
Y. K. Gunko, The Univ. of Dublin, Trinity College (Ireland) In this work we describe a novel technique for the production of aluminosilicate microtubes which are shown to act as optical cylindrical microresonators. The free-standing microtubes were fabricated by using vacuum-assisted wetting and filtration of silica gel through a microchannel glass matrix. The microtubes were studied using Scanning Electron Microscopy, microphotoluminescence spectroscopy and fluorescence lifetime imaging confocal microscopy. In the emission spectra of the micro-resonators we find very narrow periodic peaks corresponding to the whispering gallery modes of two orthogonal polarizations with quality factors up to 3200. A strong enhancement in photoluminescence decay rates at high excitation power demonstrates the occurrence of amplified spontaneous emission from a single microtube. These microtubes show a large evanescent field extending many microns beyond the tube radius. Applications for these novel microresonators will be in the area of optical microsensors for a single molecule detection of biological and chemical species, including anti-terrorism and defense sectors.

34

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6740: Optical Materials in Defence Systems Technology


Monday-Tuesday 17-18 September 2007 Part of Proceedings of SPIE Vol. 6740 Optical Materials in Defence Systems Technology IV

6740-01, Session 1
Tailored molecular systems for efficient thin films second harmonic generation
R. Zamboni, S. Caria, M. Murgia, Istituto per lo Studio dei Materiali Nanostrutturati (Italy) No abstract available

6740-02, Session 1
New electrooptic polymer configurations for high frequency modulators and digital signal processing applications
H. R. Fetterman, B. Seo, B. J. Bortnik, S. K. Kim, Univ. of California/Los Angeles (USA) No abstract available

6740-03, Session 1
Engineering and performances of pi conjugated polymers for all optical switching applications
I. Ru, POLITEHNICA University of Bucharest (Romania); F. Kajzar, Univ. dAngers (France); P. Armatys, J. Niziol, P. Chollet, Commissariat lEnergie Atomique (France) Quasi 1-D conjugated pi electron polymers emerged as an interesting class of nonlinear optical materials with potential applications in all optical signal processing. They are characterized by a large and fast, electronic in origin, cubic chi(3) susceptibility. This is arising from the strong pi electron delocalization, leading a large molecular hyperpolarizabilty gamma xxxx in the electron delocalization direction xxxx and depends strongly on the conjugation length. However when passing from the molecular to macroscopic level, this dimensionality character and its benefit is partly lost because of usually observed randomization of polymer chains orientation. Therefore, in order to increase the material performances, the ordering of polymer chains is very important, as the improvement factor may reach value of 5 when passing from a completely disordered (isotropic) system to a monooriented one. In this talk we will discuss the techniques which lead to obtaining thin films with oriented polymer chains. It is applied for several soluble conjugated polymers such as: poly(3-alkyl thiophenes), polyaniline and a soluble form of poly(phenyl vinylene). These materials were processed into thin films by solution drawing technique, which, depending on deposition conditions, allows to obtain partly oriented thin films. The degree of orientation was measured by the optical dichroism and their NLO properties were studied by optical third harmonic generation and by the z-scan techniques. For some of them the figures of merit for all optical switching applications were derived.

their TPA properties will be interpreted using theoretical calculations. We will present the effect of coordination for twophoton absorption and optical limiting properties in new Ru and Zn (II) complexes based on oligofluorenes ; interestingly, the two-photon excitation of the 3MLCT opens the way to a wide range of applications in the near infrared (NIR). The rapid development of frequency-tunable pulsed lasers up to telecommunication wavelengths (1400-1600 nm) led to the design of new materials for nonlinear absorption in the NIR region. In this spectral range, such materials could address various applications for optical limiting (OL), such as gain equalization of Erbium amplifier for telecommunication systems, protection of monostatic LIDAR (light detection and ranging) detectors against intense retro-reflection / backscattering (in addition to or replacement of detector gating) and protection of active imagery systems against laser threat. In this wavelengths range, dipolar and symmetrical dyes with a strong linear absorption in the NIR (700-900 nm) will be described. These chromophores present nonlinear absorption properties in the range 1400-1600 nm, which will be interpreted on the basis of TPA and TPA induced excited state absorption.

6740-05, Session 1
Rotational contribution to cubic susceptibility of rotaxanes and catenanes in solution
F. Kajzar, Univ. dAngers (France); I. Ru, POLITEHNICA University of Bucharest (Romania); R. Czaplicki, A. Humeau, J. Luc, O. Krupka, B. Sahraoui, Univ. dAngers (France) No abstract available

6740-06, Session 1
Model kinetics of surface relief gratings formation in organic thin films: a Monte Carlo study
A. C. Mitus, G. Pawlik, Politechnika Wroclawska (Poland); B. Sahraoui, F. Kajzar, Univ. dAngers (France) The semi-macroscopic mechanisms responsible for the surface relief gratings (SRG) formation on azobenzene-containing films are far from deep understanding [1]. Recently [2], we have proposed a simple kinetic Monte Carlo model which mimics some of experimentally known features of mass transport accompanying multiple photoisomerization trans-cis cycles under spatially inhomogeneous light illumination. This model relates the mass transport with a specific reaction of polymeric chains to mechanical forces originated by photoisomerization cycles. Preliminary simulations [2] show that mass transport takes place from illuminated to dark places, similarly as in the process of SRG formation in organic systems. An analysis of non-gaussian motion of polymeric chains is presented. We extend the model to account for the effects of light polarization. Restricted geometry effects are briefly discussed in light of recent experiments on organometallic Rhutenium complexes. [1] Z. Sekkat and W. Knoll (Eds.), Photoreactive Organic Thin Films, Academic Press (2002). [2] G. Pawlik, A. C. Mitus, A. Miniewicz, F. Kajzar, Nonlinear Opt. Quantum Opt, vol. 35, 21 (2006).

6740-04, Session 1
Nonlinear absorption by two-photon absorption in the visible and at telecommunication wavelengths
C. Girardot, C. Andraud, P. Bouit, C. Barsu, Y. Bretonnire, G. Lemercier, O. Maury, Ecole normale suprieure de Lyon (France); P. L. Baldeck, Univ. Joseph Fourier (France); K. Kamada, National Institute of Advanced Industrial Science and Technology (Japan); B. Cao, D. Riehl, Delegation Generale Pour LArmament (France); G. Wetzel, P. Feneyrou, Thales Research & Technology (France) The two-photon absorption (TPA) process, involved in a lot of applications (three-dimensional (3D) fluorescence imaging, 3D optical data storage, 3D lithographic microfabrication, photodynamic cancer therapy and optical limiting), has been the subject of growing interest in recent years. Oligomers of fluorene are of great interest for the design of molecules with enhanced TPA cross-sections in the visible. Their high TPA efficiency may be interpreted in terms of coherent coupling of transition dipole moments in oligomers. 2D and 3D dendrimeric polyfluorene derivatives will be also presented and

6740-08, Session 1
Broadband optical limiting of covalently fullerene-functionalized carbon nanotubes
B. Cao, D. Riehl, A. Piret, Delegation Generale Pour LArmament (France); J. A. Delaire, Ecole normale suprieure de Cachan (France); T. Bark, E. Doris, Commissariat lEnergie Atomique (France) The need for passive optical limiters to protect human eyes and sensors from intense laser beams has given rise to the development of optical limiting (OL) devices. OL has been investigated in a number of materials with distinct nonlinear mechanisms. The unique properties of carbon nanotubes (CNTs) make them very attractive for use in these devices because of

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

35

Conference 6740: Optical Materials in Defence Systems Technology


their broadband performance in the visible and NIR region. However, the manipulation and processing of CNTs have been limited by their insolubility in most common solvents. During the past few years, several functionalization strategies have been proposed in order to overcome the poor solubility without disrupting the physical properties of the nanotubes themselves. Recently, it was shown that the performances of CNTs can be significantly improved by the attachment of organic and organometallic molecules which rely on reverse saturable absorption (RSA). The objective of this study was to investigate the OL properties of some covalently fullerene-functionalized single-walled CNTs suspended in chloroform. Nonlinear transmittance measurements in the visible-NIR region were carried out using a nanosecond optical parametric oscillator. The OL performance of functionalized CNTs was found better than that of their parent nanotubes alone, while the modified fullerenes show poorer OL responses. We demonstrate that the functionalization of the CNTs significantly improve their OL properties of the carbon nonotubes, possibly due to an electron transfer effect. Further studies for better understanding of the CNT-C60 interactions in this new OL material by using other optical spectroscopy techniques such as Raman and pump probe are still under investigation. investigation.of such measurements. The first one is based on the recording of the dynamic holographic gratings in photorefractive crystals. The second one uses the effect of nonsteady-state photoelectromotive force (photo-EMF). The mechanism responsible for the effect can be described as follows. Illumination of wide-gap semiconductor by an interference pattern produces a non-uniform excitation of free carriers (photoconductivity grating). Diffusion of the photo-excited carriers towards the dark regions leads to charge redistribution between deep traps in the photoconductor. A space charge field grating arises. Small vibrations of the light pattern excite an alternating current through the crystal. High-resistance wide-gap materials, such as photorefractive sillenites, molecular crystal SnS2, boron nitride (BN), GaN nanoparticles in porous glasses, polypyrrole nanostructures in chrysotile asbestos matrix are used in the experiments.

6740-13, Session 1
Anisotropy of photoconductivity and nonlinear effect in GaSe
M. Karimi, A. A. Moghaddam Saray, Islamic Azad Univ. (Iran) In work are investigated particularities of photoelectric properties of GaSe monocrystals in the region of absorption edge at various configurations of current contacts at low and high optical excitation levels. It is established, that the form of a spectrum of photoconductivity in the region of absorption edge is determined by two factors: by localization electronic and exciton states along c-axis, due to presence defects of packing and nonlinear light absorption at high excitation levels in the exciton absorption region. It is shown, that the localization of electronic state and excitons in the field of one_dimensional fluctuation potential along c_axis results to an anisotropy of spectrum of photoconductivity at various configurations of current contacts. At E&#61534;c the photoconductivity is observed as in the region hv<Eg , and in the region hv>Eg . And at hv<Eg the photoconductivity as in impurity absorption region with energy of a maximum 1.960/1.975 eV, and in exciton absorption region with energy of a maximum 2.0 eV are observed. With rise of level excitation is observed suppressions of photoconductivity in the exciton absorption region and increasing in impurity absorption region. At E || c the considerable photoconductivity is observed only in the region impurity absorption, which also is increased with rise of level excitation. It is supposed, that suppression of photoconductivity in the exiton absorption at high levels of excitation is connected with exiton_exciton interaction, which results in a nonlinear light absorption.

6740-09, Session 1
Walk-off correction in biaxial crystals
H. Lee, H. E. Meissner, Onyx Optics Inc. (USA) A comprehensive model has been developed to construct phase matching conditions, including Poynting vector directions, parametrically interacting beam wavelengths, relative walk-off angles and polarization states in a biaxial nonlinear single crystal with known refractive index dispersion. The model optimizes phase matching conditions by providing a strategy for walk-off compensation that determines the optimum periodicity of twisttwin Adhesive-Free Bond (AFB(r)) composite pair designs. The model is validated experimentally by measuring the calculated walk-off angles and the crystal orientation of KTP that correlates to the given Poynting vector. The method is useful in reducing uncertainties of OPO designs, in providing walk-off compensation design data, and in confirming the final walk-off corrected design configuration. It is generally applicable to biaxial and uniaxial nonlinear crystals.

6740-11, Session 1
Cubic effects in doped DNA systems
F. Kajzar, Univ. dAngers (France); B. J. Derkowska, Univ. Mikolaja Kopernika (Poland); A. El-Ghayoury, O. Krupka, Univ. dAngers (France); I. Ru, POLITEHNICA University of Bucharest (Romania); B. Sahraoui, Univ. dAngers (France); J. G. Grote, Air Force Research Lab. (USA) No abstract available

6740-14, Session 2
Learning from Mother Nature: DNA and other natural materials for biophotonics
A. J. Steckl, Univ. of Cincinnati (USA) No abstract available

6740-12, Session 1
Space-and-time current spectroscopy of widegap semiconductors
I. A. Sokolov, M. A. Bryushinin, A.F. Ioffe Physico-Technical Institute (Russia) Investigation of electronic kinetic processes in solid state material and especially in semiconductors has great importance because it provides information about the main fundamental characteristics of the material such as energy level structure, relaxation processes, Fermi surface shape, volume charge distribution, specific properties of charge carriers, etc. These parameters can be measured both using electrical and optical methods. The latter are preferable because they allow nondestructive testing of materials to be carried out. The dynamics of spacecharge fields in wide-gap high-resistive materials is very slow (the Maxwell relaxation time is inversely proportional to the conductivity of the material). Utilization of conventional methods for the characterization of material parameters, for example, photoconductivity of the material, is difficult due to the influence of the internal space-charge field on the results of the measurement. There are two modern techniques for wide-gap materials

6740-15, Session 2
Biopolymer photonics
F. Omenetto, Tufts Univ. (USA) No abstract available

6740-16, Session 2
Bio-organic materials for electronic and photonic applications
J. G. Grote, Air Force Research Lab. (USA) No abstract available

36

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6740: Optical Materials in Defence Systems Technology

6740-18, Session 3
Tunable wavelength VCSEL quantum cascade lasers for chemical sensors in the 3-6 micron spectral region
A. Christou, C. Zhang, Univ. of Maryland/College Park (USA) Abstract. The authors present the design and performance of a low threshold selectively oxidized Vertical Cavity Surface Emitting Laser (VCSEL) fabricated for operation at a wavelength of 3.0 microns. The device is based on III-V quaternary semiconductor alloys and is grown by Molecular Beam Epitaxy technique. The theoretical investigation of the optical properties of the compound semiconductor alloys allows it to select the optimum materials for highly reflective Bragg mirrors. The simulation of the designed VCSEL performance has been carried out by evaluation of the important laser characteristics such as threshold gain, threshold current density and external quantum efficiency. We present a design which integrates a thermoelectric controller and multi-section lasers to provide continuous tenability. Compound semiconductor based long wavelength surface emitting lasers are becoming important light sources for large capacity optical communications and optical interconnection systems. Those devices exhibit such advantages as low threshold currents, single mode operation, high coupling efficiencies into optical fiber [1], and high speed modulation [2]. The development of vertical cavity lasers at 3?m has been centered on the investigation of structures and materials for highly reflective Bragg mirrors grown on InP. The most commonly used material systems for the Distibuted Bragg Reflectors (DBR) are GaInAs/InP which requires the growth of 50 layer pairs to achieve a reflectivity of 99.9% [3], and AlGaInAs/AlInAs which requires the growth of 41 pairs for R=99.9% [8], thus showing the need for very thick structures to obtain high reflectivity Bragg mirrors. Higher refractive index contract DBRs reported with fewer mirror pairs are based on the AlGaAsSb/AlAsSb material system [9,10]. These Bragg mirrors use only 20 pairs of alternating AlGaAsSb/AlAsSb layers to obtain reflectivity of 99%. However, these structures are easily degraded and are usually difficult to grow [4]. These structures suffer from interface roughness near the top layers and from thickness variations of greater than ten percent. Many diatomic molecules (CO, HCl, NO etc) as well as triatomic molecules have absorption bands in the 3-20 micron range. High sensitive chemical sensors rely on the resonant absorption of infrared radiation (3-20 microns) of the gas molecules within a contained optical ring. The differential absorption lidar technique allows for the detection of low concentration of molecules in the atmosphere. In this paper, we report the design, fabrication and performance of a tunable VCSEL which can emit within the desired wavelength region of interest for both differential lidar and other detection methods. The tunability is achieved with an integrated thermoelectric device and multiple lasers on a single backplane. Comprehensive investigation of the dielectric constants and energy band gaps of quaternary semiconductor alloys have been utilized to optimize the structure of the vertical cavity surface emitting laser, and to achieve improved performance. The study also includes threshold current, gain and external quantum efficiency calculations for the VCSEL structure designed, and comparison with experimental results. References [1] Tadokoro T, Okamoto H, Kohama Y, Kawakami T, and Kurokawa T, 1992 IEEE Photon. Techn. Lett. 4(5) 409-411 [2] Choa F, Lee Y, Koch T, Burrus C, Tell B, Jewell J, and Leibenguth, 1991 IEEE Photon. Techn. Lett. 3 697-699 [3] Blum O, Fritz I, Dawtson L, and Drummond T, 1995 Electron. Lett. 31(15) 1247-1248 [4] Harmand J, Kohl A, Juhel M, and Le Roux G, 1997 J. Cryst. Growth 175 372-376

6740-20, Session 3
Fabrication of midwave infrared InAs photodiodes with reduced surface leakage current
C. H. Tan, The Univ. of Sheffield (United Kingdom) We report studies of wet etching using a range of etchants on InAs photodiodes used for detection of midwave infrared wavelengths up to 3.6m. Current-voltage measurements and analysis using scanning electron microscope were carried out to assess the electrical characteristics and surface profiles of the photodiodes. Phosphoric acid, hydrobromic acid and acetic acid based etchants were found to yield high surface leakage currents while a sulphuric acid based etchant provide the lowest surface leakage current. The surface leakage current is highest in devices etched with phosphoric acid based etchant. However this surface leakage current was drastically reduced (by ~2 orders of magnitude) when a two steps etching, starting with a phosphoric based etchant and finishing off with a sulphuric based etchant, was used. Surface profile analysis on all samples showed that higher etching rates were obtained in the direction parallel to the <100\> direction. However this preferential etching rate can be removed by using a 45o wafer rotation with respect to the <100\> direction. Etch rates and surface profiles obtained in this work can be used for fabrication of InAs midwave infrared focal plane array.

6740-22, Session 3
Thick orientation-patterned gallium arsenide (OP-GaAs) for mid-infrared laser sources
D. Faye, E. Lallier, A. Grisard, B. P. Gerard, Thales Research & Technology (France) The improvement of tunable laser sources in the mid-infrared region (MIR) appears nowadays as one of the ways to address several defence optronics applications. Since few direct lasers exist in the 3-12 m part of the infrared spectrum, nonlinear optical materials play a key role as they permit the frequency down-conversion of mature near-infrared solid-state lasers into the desired longer wavelengths. In the past few years, we have been developing a new technology of Quasi-Phase-Matched (QPM) nonlinear devices based on Gallium Arsenide (GaAs) materials. Beside its favorable infrared transparency (2 m-12 m), GaAs has also a combination of properties, such as high nonlinear coefficient and thermal conductivity, which potentially makes it a good candidate for high average power MIR applications. Furthermore, GaAs is also a mature material produced in high volume for a wide range of commercial applications. The main obstacle to develop GaAs for nonlinear devices is still to find a practical method of periodically patterning the material in order to fulfill QPM requirements. We proposed an approach which consists in growing orientation-patterned thick layers by Hydride Vapour Phase Epitaxy (HVPE) compatible with freespace beam propagation, on top of an Orientation Patterned template (OP-GaAs). This technique exhibits numerous advantages. First, the template pattern dimensions can be set using photolithography with a large versatility. Furthermore, the device lengths can be extended up to wafer size (2 or 3 inches diameter) without difficulty and many devices with different gratings can easily be fabricated in parallel. Finally, benefiting from the GaAs mature processing technologies, growth rates up to at least 30 m/h are achievable by our HVPE process with film properties close to those of bulk material. This paper will review the recent progresses we achieved with thick OP-GaAs structures. We will present results obtained in growing thick-layer (500 m) on 3 cm long structures with very low optical losses (0.01 cm-1). This loss coefficient is low enough to allow the realization of a high power OPO in the MIR band.

6740-23, Session 3
The effects of monolayer thickness and sheet doping density on dark current and noise current in quantum dot infrared photodetectors
C. H. Tan, The Univ. of Sheffield (United Kingdom); C. L. T. M. Souye, University of Sheffield (United Kingdom); P. Vines, J.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

37

Conference 6740: Optical Materials in Defence Systems Technology


P. David, M. Hopkinson, Univ. of Sheffield (United Kingdom); L. R. Wilson, P. Aivaliotis, The Univ. of Sheffield (United Kingdom) We report measurements on a series of quantum dot infrared photodetectors grown with different combinations of monolayer thicknesses (2.2. 2.55 and 2.9ML) and quantum dot layer sheet doping densities (6&#61620;1010cm-2 and 12&#61620;1010cm-2). The dark current and noise current were higher in devices grown with sheet doping density of 12&#61620;1010cm-2. At a given bias voltage the dark current and the noise current was found to be lowest in devices having 2.55ML and sheet doping density of 6&#61620;1010cm-2. This combination gives a sheet doping density to dot density ratio of approximately unity. Gain extracted from noise current showed that it varies reasonably linearly with electric field suggesting that the gain mechanism is photoconductive. Highest gain was achieved in devices with 2.55ML and sheet doping density of 6&#61620;1010cm-2.

6740-25, Poster Session


Photo-physical properties and OPL of novel thiophenyl containing arylalkynyl Pt(II) compounds
E. Glimsdal, Norges Teknisk-Naturvitenskapelige Univ. (Norway); M. Carlsson, B. Eliasson, Ume Univ. (Sweden); M. Lindgren, Norges Teknisk-Naturvitenskapelige Univ. (Norway) The multi-photon absorption and optical power limiting (OPL) properties of several new thiophenyl-containing arylalkynyl Pt(II) complexes with longer arylalkynyl groups, named ATP3, ATP4, ATP6 and ATP7 were studied. Thiophene units were introduced into the structure as an attempt to modify photo-physical properties and OPL capability. The compounds have their thiophene rings either close to the Pt-atom (ATP7), in the middle of the chain (ATP3), or at the terminal position (ATP4). ATP6 is similar to ATP4, but with methoxy groups on the arylalkynyl ligands. The measurement results were compared with those of the earlier studied PE3 compound. Just as PE3, all thiophenyl derivatives showed large intersystem crossing capabilities and triplet phosphorescence, thus having the potential of large nonlinear absorption and good OPL performance. All compounds are characterized with absorption and emission spectra, fluorescence quantum yield, excited state emission lifetimes (fluorescence and phosphorescence), two-photon absorption capabilities, and OPL performance. All investigated compounds show OPL clamping levels of approximately 3 uJ pulse energy in 30 mM concentration samples. All compounds have high transmission in the visible region (above 450 nm) and fluorescence quantum yields in the order of 0.01.

6740-24, Session 3
Simulation of a small Si plate oxidation in a cwCO2 lase light
S. Balint, A. M. Balint, R. Szabo, West Univ. of Timisoara (Romania) In the spot of cwCO2 laser of power P the temperature of a small bare Si plate increases from T=300K to an equilibrium temperature T0(P). The moment of time t1(P) at which T0(P) is achieved, t0(P) at which the first oxide layer of 1 nm is accomplished, the sample temperature at t0(P) , the oxide layer thickness at t1(P), the oxide layer growth speed at t0(P) and t1(P) are computed, for dry and wet oxidation and for <100\> and <111\> oriented samples. Computation shows that for P bigger than a critical value, layers of thickness below 10 nm grow during the transition period.

6740-26, Poster Session


Al2-X MeX (WO4)3 single crystal growth, Me =Sc, Y, Ga and In, as a new tunable laser media
D. Ivanova, V. Nikiolov, P. Peshev, Institute of General and Inorganic Chemistry (USA) The rapid detection of chemical and biological agents in the air is an important task for the todays security and defence. Solid state tunable lasers are one of the main instruments for such kind of detection. Besides the synthesis of new laser media with broad band laser emission is an extremely important task. Al2-X MeX (WO4)3 compounds in which Al3+ could be relatively easy replaced by Me3+ ions of 3d-elements (Cr,Ti) are suitable for matrix of solid state tunable laser media for ecology monitoring. The main purpose of the presented investigation has been to clarify the conditions of growth of single crystals by the mentioned solid solutions, as a base for further detailed study of their properties and possible applications. In the first stage the 27,5 Na2O-72,5WO3 solvent was established as the most suitable for single crystals growing by flux method. Its main advantages are: wide concentration range of crystallization; wide temperature range; low viscosity; negligible weight losses due to evaporation; quick homogenization. In the second stage the conditions for growth of single crystals from solid solutions in different levels of Al3+ replaced by Sc, Y, Ga and In have been studied. The structure of the single crystals by the obtained solid solutions has been proved to be as Sc2(WO4)3-type structure and the exact chemical composition has been determined by EPMA analyses. Some preliminary optical characteristics were measured and discussed.

6740-19, Poster Session


Advances in mid-infrared GaSb-based lasers
G. L. Belenky, L. Shterengas, D. V. Donetsky, S. Suchalkin, M. V. Kisin, Stony Brook Univ. (USA) G. Belenky , L. Shterengas, D. Donetsky, S Suchalkin and M. Kisin 1) State University of New York at Stony Brook, Stony Brook, NY 11974-2350 1) Power Photonic Corporation, Stony Brook, NY 11790 The latest achievements in design and performance of GaSb based interband lasers will be discussed. The maximum continuous-wave (CW) room temperature output powers for 2.5, 2.7 and 2.8m devices were 1W, 500mW and 160mW, respectively. 2.5- m diode at pulsed-current mode demonstrates an output power of 5 W at 20oC, a peak power of 2.5 W was recorded in the pulsed mode operation for 2.7m and 2 W at 2.8-m lasers [1]. The characteristics of the double-quantum-well GaSb-based diode laser operating at 2.4 m with a room-temperature CW output power of 1050 mW and a maximum power-conversion efficiency of 17.5 % will be presented and discussed [2,]. We will discuss a novel ultra-wide, voltage-tunable Type-II midIR interband cascade laser based on InAs/Ga0.8In0.2Sb/GaSb heterostructures [4]. 1. L. Shterengas, G.L. Belenky, J.G. Kim, R.U. Martinelli Design of high-power room-temperature continuous-wave GaSbbased type-I quantum-well lasers with \>2.5m, Semicond. Sci. Tech. 19, 655 (2004). 2. L. Shterengas, G.L. Belenky, D. Donetsky, M. Kisin High power 2.4m heavily strained Type-I quantum well GaSb-based diode lasers with more than 1W of continuous wave output power and a maximum power conversion efficiency of 17.5%. Appl. Phys. Lett. 90, 011119 (2007). 3. S. Suchalkin, M.V. Kisin, S. Luryi, G. Belenky, F.J. Towner, J.D. Bruno, and R.L. Tober , High-speed wavelength tuning of Mid-IR cascade lasers, IEEE Photon. Technol. Lett. 19, 360 (2007).

38

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6741: Optics and Photonics for Counter-Terrorism and Crime-Fighting


Wednesday 19 September 2007 Part of Proceedings of SPIE Vol. 6741 Optics and Photonics for Counterterrorism and Crime Fighting III

6741-37, Poster Session


Secure OFDM communications based on hashing algorithms
D. Blasi, P. Campisi, A. Neri, Univ. degli Studi di Roma Tre (Italy) In recent years wireless technology has become widespread in the design of telecommunications systems. The main issues related to wireless communication can be identified in: dataintegrity, user authentication, and privacy of the communications. An innovative modulation and encoding scheme based both on PH-OFDM (Phase Hopping - Orthogonal Frequency Division Multiplexing) modulation and on encrypted hash algorithm (such as MD2, MD5, SHA-1, SHA-256, SHA-384 and SHA-512) is presented. M of the N OFDM carriers are spent for transmitting the data using the Phase Hopping, while the remaining M - N are used to carry the hashed version of the information availing of a standard OFDM modulation. We refer to an L-PSK (L-ary Phase Shift Keying) constellation for every sub-band. The proposed systems can be adopted either for a point-topoint data exchange, providing performance improvement, mutual authentication and communication privacy, or in the access to reserved areas, networks or databases. In the last situation the information bits could contain biometric data (like a digital signature, as an instance), and the hash of this data should also be stored in a unit dedicated to the authentication stage. In the first scenario, signal alteration introduced by the communication channel, is taken into account by protecting both data and the hash bits by FEC (Forward Error Correction) techniques. The other scenario is even more complex, since apart from the effect of the channel, also the variability of the biometric features must be considered. To compensate for this drawback, in addition to the use of FEC, only the bits relative to the characteristics resulting statistically enough stable have to be checked out.

6741-01, Session 1
Threat image projection in CCTV
N. Thomas, B. Baker, Home Office Scientific Development Branch (United Kingdom) Key components in a CCTV system are the operators. They are the link between the system technology and its effective use. The operators performance will largely determine the level of service provided by the system. It is unsatisfactory to consider the performance of the CCTV technology in isolation from that of the operators. However, this is frequently the case. There have been few studies to give evidence about operator performance, while much work has been done to classify the performance of the technology. Previous work on CCTV operator performance has been carried out by the Home Office Scientific Development Branch (HOSDB). However, such studies used filmed video and subjects who knew they were undergoing testing, meaning subjects are likely to be concentrating harder on performing well. There is a need for a test that would be able to be routinely used in a CCTV control room throughout the course of a normal shift to provide management with operational performance data. Threat Image Projection (TIP) is routinely used in X-Ray baggage scanners to keep operators alert to possible threats. At random intervals, a threat target image is superimposed over the image of the baggage being screened. The operator then responds to this threat. A similar system could be used for CCTV operators. A threat image would be randomly superimposed over the live CCTV feed and the operator would be expected to respond to this. This paper details the requirements for such a system, which have been used to procure a prototype.

6741-39, Session 1
Holistic video detection
S. Gong, Queen Mary Univ. of London (United Kingdom) No abstract available

6741-38, Poster Session


Imaging of space with impulse illumination of scenes
M. Piszczek, K. Rutyna, M. Szustakowski, Wojskowa Akademia Techniczna (Poland) In the systems used for protection of large objects such as airports, boundaries, industrial objects (atomic power stations, chemical plants, and the like), various remote detection devices and intruder detection devices are used. These are bolometric (uncooled) thermal cameras of middle range of observation and detection (1-2 km) or far-range cameras (6-8 km), radars of millimetre wavelengths of mid-range and far-range microwave radars. Recognition of a person image using the above mentioned detection devices is difficult and his identification almost impossible. At night and under difficult metrological conditions (fog, rain, snow) image recognition and, what is more, identification can be done with the higher uncertainty. To solve the above mentioned tasks, it is proposed to record the suspected object, e.g. detected with radar or thermal camera, by means of multifunctional, highly sensitive optoelectronic system of image registration for its further recognition and identification. The proposed observation system give possibility good quality images, independently of changing the conditions of object illumination (day, night) and presence of disturbing elements in the atmosphere (fog, rain, snow) appearing between an object and observation system. The paper will be presents idea of active observational system, methods of scene visualization its application possibilities in save and security systems and results of computer modeling of image acquisition.

6741-40, Session 1
TBA
No abstract available

6741-04, Session 2
Multi-camera tracking: UK government requirements
P. D. Hosmer, Home Office Scientific Development Branch (United Kingdom) Multi-Camera Tracking - UK government requirements: Home Office Scientific Development Branch, UK Abstract The Imagery Library for Intelligent Detection Systems (i-LIDS) is the Governments new standard for Video Based Detection Systems (VBDS). The standard was launched in November 2006 and evaluations will commence in 2007. With the first four i-LIDS scenarios completed, we are looking toward the future of intelligent vision in the security market by adding a fifth scenario to the standard. The fifth scenario will concentrate on the development, testing and evaluation of systems for tracking of people across multiple cameras. The HOSDB and the Security Service identified a need to be able to track targets site-wide using both live and post event imagery. The Detection and Vision Systems group were asked to determine the current state of the market and develop an indepth Operational Requirement (OR) based on government end user requirements. Using this OR the i-LIDS team will develop a full i-LIDS dataset to aid the machine vision community in its development of multiple camera tracking systems. By defining a requirement for multiple camera tracking and

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

39

Conference 6741: Optics and Photonics for Counter-Terrorism and Crime-Fighting


building this into the i-LIDS standard the UK Government will build a bridge which system developers can use to aid them turn theory into front line application. This paper will briefly describe the i-LIDS project and then detail the work conducted in building the new tracking aspect of the standard. Authors personal data: Mr Paul Hosmer Detection and Vision Systems Home Office Scientific Development Branch Langhurst House Langhurstwood Road Horsham West Sussex RH12 4WX UK Phone +44 (0)1403 213816 Fax +44 (0)1403 213827 Email paul.hosmer\@homeoffice.gsi.gov.uk site. We will present success and challenges yet to be addressed by the proposed approach.

6741-07, Session 2
Video scene assessment with an unattended sensor network
S. Guler, intuVision, Inc. (USA) This paper describes a prototype video imagery based analysis system to monitor remote areas which may be in unfriendly zones to be aware and alert of unusual activity in these territories. This application involves analysis and metadata extraction from collected multiple video streams, assessment and prioritization of the scene understanding information and representation of the extracted knowledge. We present a highly efficient video content extraction and scene assessment paradigm with algorithms designed after human visual cognition system. Our 2-tier video tracking approach uses a fast spacebased peripheral vision component and tunnel vision components for object-based detail analysis, when combined produces an efficient overall scene analysis framework. The representation for this extracted information has to be in a form that is efficient and readily available for applications upstream including bandwidth limited communications and inference of higher level understanding tasks involving other sensor data such as other cameras, aerial imagery, radar imagery etc. The current prototype sensor system has a wireless mesh network backbone which can incorporate a range of sensor types including video, acoustic and chemical. The data from these sensors can be correlated and combed for specific targets (vehicles, people etc). The application requirements impose competing richness of context and bandwidth constraints hence pushing for an efficient, flexible and rich representation of the extracted content.

6741-05, Session 2
Smart cruise control: UAV sensor operator intent estimation and its application
H. Cheng, D. Butler, Sarnoff Corp. (USA) Due to their long endurance, superior mobility and the low risk posed to the pilot and sensor operator, UAVs have become the preferred platform for persistent ISR missions. However, currently most UAV based ISR missions are conducted through manual operation. Event the simplest tasks, such as vehicle tracking, route reconnaissance and site monitoring, need the sensor operators undivided attention and constant adjustment of the sensor control. The lack of autonomous behaviour greatly limits of the effectiveness and the capability of UAV-based ISR, especially the use of a large number of UAVs simultaneously. Although fully autonomous UAV based ISR system is desirable, it is still a distant dream due to the complexity and diversity of combat and ISR missions. In this paper, we propose a Smart Cruise Control system that can learn UAV sensor operators intent and use it to complete tasks automatically, such as route reconnaissance and site monitoring. Using an operator attention model, the proposed system can estimate the operators intent from how they control the sensor (e.g. camera) and the content of the imagery that is acquired. Therefore, for example, from initially manually controlling the UAV sensor to follow a road, the system can learn not only the preferred operation, tracking, but also the road appearance, what to track in real-time. Then, the learnt models of both road and the desired operation can be used to complete the task automatically. We have demonstrated the Smart Cruise Control system using real UAV videos where roads need to be tracked and buildings need to be monitored.

6741-09, Session 2
Detection and tracking of humans in urban environments for visual surveillance
R. Nevatia, Univ. of Southern California (USA) A necessary component for systems that assist in counterterrorism and crime fighting is an ability to detect and track humans in the environment as humans are the principal agents in committing criminal or terrorist actions. Video cameras have become quite inexpensive and quality has improved tremendously; thus, it is feasible to deploy a large network of such cameras in urban environments of interest. The cameras may be placed in a variety of locations such as on light or traffic poles, rooftops and walls of buildings. The bottleneck is in monitoring the outputs of these sensors for suspicious activities. Automatic detection and tracking of humans is an important step in providing computer assistant to the security personnel. Detection of humans is complex due to changes in image appearance caused by viewpoint changes, illumination changes and clothing variations. Furthermore, in urban environments, humans are in close proximity and often occlude each other or are occluded by various objects in the scene. Thus, simple methods, such as detection and tracking of motion blobs are not adequate as multiple humans may merge into a single blob or one human split into multiple blobs. We propose that use of human shape models can allow us to separate individual humans and track them under these conditions. We use a part-based model so that humans can be detected even if certain parts are occluded. Our methods have been evaluated on a number of publicly available datasets. They show good performance compared to previously published approaches.

6741-06, Session 2
Tracking moving objects across nonoverlapping cameras
I. Cohen, Y. Ma, B. Miller, Honeywell Technology (USA) In this paper we present an approach for tracking people across non overlapping cameras. The approach proposed is based a multi-dimensional feature vector and its covariance, defining an appearance model of every detected blob in the network of cameras. The model integrates relative position, color and texture descriptors of each detected object. Association of objects across non-overlapping cameras is performed by matching detected objects appearance with past observations. Availability of tracking within every camera can further improve the accuracy of such association by matching several targets appearance models with detected regions. For this purpose we present an automatic clustering technique allowing to build a multi-valued appearance model from a collection of covariance matrices. The proposed approach does not require geometric or colorimetric calibration of the cameras. We will illustrate the method for tracking people in relatively crowded scenes in a collection of indoors cameras taken in a mass transportation

6741-10, Session 2
Real-time object recognition by the photoanisotropic copies
B. N. Kilosanidze, G. A. Kakauridze, Institute of Cybernetics (Georgia)

40

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6741: Optics and Photonics for Counter-Terrorism and Crime-Fighting


A new method of optical image real time recognition by means of photoanisotropic copy is suggested. The method is to obtain a photoanisotropic copy of object images on the polarizationsensitive material. It was shown that when an object image is recorded on polarization-sensitive material by linearly polarized light which is actinic for this material then photoanisotropy is induced in material. In this case the distribution of intensity of polarized light is transformed into distribution of photoinduced anisotropy on the surface of material. As a result a photoanisotropic copy of an initial object is received. At consequent transillumination of the photoanisotropic copy received in such way with a parallel circularly polarized beam of nonactinic light the transmitted light becomes elliptically polarized. The objects of amplitude transparent type are considered as an example. It is shown that the characteristics of the summary polarization ellipse (ellipticity, direction of rotation and an azimuth) in the Fraunhofer diffraction region uniquely identify the given object. The real time definition of the characteristics of the summary polarization ellipse is made by means of diffraction gratings of anisotropic profile and comparison of these characteristics with etalon from database. In comparison with the methods of recognition on the basis of scalar complex-conjugate filtration, the suggested method is easily realized, since it is invariant to object scaling and displacement. The use of dynamic polarization-sensitive materials for obtaining a photoanisotropic copy will allow the recognition of fast variable fields and dynamic objects to be realized in real-time. Such a method has not been studied up to now and the suggested approach to the task of objects recognition is rather promising. The method can be used for the creation of automatic recognizer of the most different application, information retrieval systems. Especially promising is the use in the tasks of identification and recognition of dynamic objects in real time, and remote sensing. domains - broadcast news and meetings. In this paper, we present the results of a face detection and tracking algorithm on broadcast news videos with the objective of establishing a baseline performance for this task-domain pair. The detection algorithm uses a statistical approach that was originally developed by Viola and Jones and later extended by Lienhart. The algorithm uses a feature set that is Haar-like and a cascade of boosted decision tree classifiers as a statistical model. In this work, we used the Intel Open Source Computer Vision Library (OpenCV) implementation of the Haar face detection. The optimal values for the tunable parameters of this implementation were found through an experimental design strategy commonly used in statistical analyses of industrial processes. Tracking was accomplished as continuous detection with the detected objects in two frames mapped using a greedy algorithm based on the distances between the centroids of bounding boxes. Results on the evaluation set containing 50 sequences (2.5 mins. approx.) show good performance of the algorithm reflecting the state-of-the-art which makes it an appropriate choice as the baseline algorithm for the problem.

6741-13, Session 4
Recent advances in the physical, optical, and chemical visualization of latent prints
A. A. Cantu, Independent Scientific Consultant (USA) Latent prints on substrates of varying porosity are visualized by a host of physical, optical, and chemical techniques that are often used sequentially. Optical techniques include viewing the scattering of UV radiation by latent print residue on smooth surfaces and inducing natural (inherent) fluorescence of such residue; physical techniques include visualizing the latent print using diverse powders and vapors that selectively polymerize on the residue; and chemical techniques include reacting components of latent print residue with reagent that render color and/or fluorescence to the residue. Recent developments in visualizing techniques include the use of nanoparticles, hyperspectral imaging, and the development of more sensitive reagents for visualizing the water-soluble components of latent print residue (e.g., amino acids) and the water-insoluble components (e.g., lipids, dried proteins, etc.). The chemistry of latent print residue and current research in the methods to visualize its components will be discussed.

6741-11, Session 2
VideoQuest: an advanced aerial video exploitation and management system
H. Cheng, D. Butler, Sarnoff Corp. (USA) Aerial platforms, especially unmanned aerial vehicles (UAVs), with their high maneuverability, long endurance and low-risk to pilots and payload operators have emerged as a preferred means for intelligence, surveillance and reconnaissance. However, as the amount of aerial videos captured increases, there is a need for systematic exploitation and effective management of the large amount of aerial videos. In this paper, we will introduce VideoQuest, an advanced aerial video exploitation and management system that provides realtime aerial video enhancement, archiving, indexing and analysis capabilities such as sensor metadata enhancement, moving target detection and tracking and event detection. To effectively and efficiently utilize archived aerial videos, VideoQuest also provides spatial, temporal and content based indexing. To quickly retrieve videos in a large-scale video database, the system summarizes aerial vidoe hierarchical and based content, such as object, tracks and events extracted from videos. Additionally, VideoQuest allows user to interactively search and browse large aerial video database through a virtual UAV GUI that dynamically assembles visual information according to users needs. Using the VideoQeust system, a user can search and retrieve mission-relevant information several magnitudes faster than without using our system.

6741-15, Session 4
Scene segmentation from multi-spectral imagery to aid automatic human gait recognition
D. A. C. Pearce, J. M. Nothard, C. Harvey, K. Smart, J. Hargreaves, QinetiQ Ltd. (United Kingdom) Primarily focused at military and security environments where there is a need to identify humans covertly and remotely; this paper outlines how recovering human gait biometrics from a multi-spectral imaging system can overcome the failings of traditional biometrics to fulfil those needs. With the intention of aiding single camera human gait recognition, an algorithm was developed to accurately segment a walking human from multispectral imagery. 16 band imagery from the IRIS camera system is used to overcome some of the common problems associated with standard change detection techniques. Fusing the concepts of scene segmentation by spectral characterisation and background subtraction by image differencing gives a uniquely robust approach. This paper presents the results of real trials with human subjects and a prototype IRIS camera system, and compares performance to typical broadband camera systems.

6741-12, Session 2
A baseline algorithm for face detection and tracking in video
R. Kasturi, D. B. Goldgof, P. Soundararajan, V. Manohar, V. Korzhova, M. Boonstra, Univ. of South Florida (USA); R. J. Bowers, J. Garofolo, National Institute of Standards and Technology (USA) Establishing benchmark datasets, performance metrics and baseline algorithms have considerable research significance in gauging the progress in any application domain. These primarily allow both users and developers to compare the performance of various algorithms on a common platform. In our earlier works, we focused on developing performance metrics and establishing a substantial dataset with ground truth for object detection and tracking tasks (text and face) in two video

6741-16, Session 5
Video face recognition against a watch list
J. Abbas, T. S. Huang, Univ. of Illinois at Urbana-Champaign (USA) Due to a large increase in the video surveillance data recently in an effort to maintain high security at public places, we need more robust systems to analyze this data and make tasks like face recognition a realistic possibility in challenging environments. In this paper we explore a watch-list scenario

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

41

Conference 6741: Optics and Photonics for Counter-Terrorism and Crime-Fighting


where we use an appearance based model to classify query faces from low resolution videos into either a watch-list or a non-watch-list face. We then use our simple yet a powerful face recognition system to recognize the faces classified as watch-list faces. Where the watch-list includes those people that we are interested in recognizing. Our system uses simple feature machine algorithms from our previous work to match video faces with still images. To test our approach, we match video faces against a large database of still images obtained from a previous work in the field from Yahoo News over a period of time. We do this matching in an efficient manner to come up with a faster and nearly real-time system. This system can be incorporated into a larger surveillance system equipped with advanced algorithms involving anomalous event detection, activity recognition and then face or identify recognition. This is a step towards more secure and robust surveillance systems and efficient video data analysis.

6741-18, Session 5
Iris-based authentication system with template protection and renewability
C. Ercole, P. Campisi, A. Neri, Univ. degli Studi di Roma Tre (Italy) The most emerging technology for people authentication and identification is biometrics. In contrast with traditional approaches based on what a person knows (e.g. a password) or what a person possesses (e.g. a card), biometric authentication approaches are based on who a person is (physiological biometrics) or what a person does (behavioral biometrics). Therefore biometric traits are much more difficult to be forgotten, lost, stolen, copied, or forged than a password or a card. However, whereas a new credit card can be reissued and a new password can be set, biometric features cannot be replaced. Therefore, security is an essential parameter in the design of a biometric based authentication system. This paper focuses on two vulnerable points of the whole biometric system: the Enrollment Database for user template storage and the Communication Channel between the Enrollment Database and the Matcher. In particular, in this paper, a method for protecting and securing the stored biometric templates is proposed, exploiting error correction codes tailored on the intra-class and inter-class discriminability. The aforementioned method allows template renewability applied to iris based authentication and guarantees high security being the match performed in the encrypted domain. The proposed algorithm is extensively tested on the entire CASIA v.1 Iris Database, showing the effectiveness of the proposed approach in terms of false acceptance rate (FAR) and false rejection rate (FRR), when compared to existing traditional non-secure correlation based classifiers.

6741-17, Session 5
Biometric identification of non-compliant subjects at range
I. M. Firth, LogicaCMG (United Kingdom) In November 2005, LogicaCMG were invited to conduct a proof of concept of a facial biometric identification system at an airport by integrating a leading algorithm. The purpose of the trial was to ascertain if biometric identification algorithms were mature enough to detect non-compliant subjects*at range. This trial concluded in October 2006. The aim of the proof of concept was to ascertain if a figure of 75% detection could be achieved. The system consisted of 12 analogue cameras split into two groups of four and two groups of two; depending on the width of each thoroughfare that they were required to monitor. The height of these cameras ranged from 2.5 meters to 4.5 meters monitoring a range of between 1-7 meters. The system was designed as a client-server architecture whose main components are a single Watchlist Station with multiple Tracking Stations all connected to the Watchlist Station. The algorithm automatically searches CCTV streams for faces. Once a face is located, a crop of that image is taken and sent to the Watchlist Station to ascertain if the face matches any of the enrolled suspects on the database. At the airport location, the 12 cameras were connected to a number of 19" rack mounted HP servers, with integrated frame grabbers, at a ratio of two cameras per server. These servers acted as the Tracking Stations performing an analysis of video streams from the connected cameras and sending biometric identification requests to the Watchlist Station of faces that it has detected in the CCTV stream. The Watchlist Station acts as the main identification station trying to identify the faces using a gallery of enrolled suspects. For testing purposes a number of volunteers were enrolled using a digital SLR camera. The enrolment images were standard mug shot images. Each subject was photographed 5 times with subtle changes in the head position. These images were then enrolled into the Watchlist Station with unique identifiers. A further batch of images was added to ascertain how many false positives would be seen with a database of up to 1000 enrolled images. Each subject was asked to walk through each location and the algorithm applied a score to each identification ranging from 0.0 through to 1.0. By studying the scores we were able to adjust the threshold to a level where the software would still show alerts but eliminate many of the false positives that scored a low score. Testing involved the volunteers mixing with passengers and walking through each location. As a successful identification was made the timeline on the Watchlist Station would show the frame from the CCTV image and the image from the Watchlist database. During the testing, LogicaCMG learned significant lessons around lighting, depth of field, field of view, enrolment images and database size. By adjusting these and working with the algorithm vendor, LogicaCMG achieved 75% detection. *at range for this purpose was deemed to be several meters

6741-19, Session 5
Signature-based authentication system using watermarking in the ridgelet domain
E. Maiorana, P. Campisi, A. Neri, Univ. degli Studi Roma Tre (Italy) In the last few years we have witnessed a constant increase of the utilization of biometrics-based authentication systems, due to several inherent advantages they offer over classic methods. Recently, a lot of efforts have been devoted to the development of secure and multi-level biometric authentication systems. In real life biometric applications, the issue of data security and integrity is extremely critical. A biometric multi-level authentication system is able to offer different levels of security and acceptability, depending on the specific application. There exist several security techniques to secure biometric data. Template encryption can be used to make the data useless without the knowledge of the key that must be kept secret. However, once the key is broken or the data are decrypted they are not protected anymore. To overcome this limitation, watermarking techniques can be used. In our proposal a signature-based biometric system is implemented, where watermarking is used in order to hide and keep secret some signature features into a signature image. Being a behavioural biometric, signatures are intrinsically different from other commonly used biometric data, possessing dynamic and statistic properties, which can not be extracted from a single signature image. The marked images can be used for user authentication, letting their static characteristics being analyzed by automatic algorithms or security attendants. When a higher security is needed, the embedded features can be extracted and used, thus leading to a multi-level decision procedure. The proposed watermarking technique is designed in the Ridgelet domain which is suited to represent images with sharpen edges, like a signature picture. In order to obtain a method both robust and capable of hiding relevant data while keeping intact the original structure of the host, the mark is embedded in the high energy regions of the signature image. An extensive set of experimental results, concerning the system performances will show the effectiveness of our approach.

42

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6741: Optics and Photonics for Counter-Terrorism and Crime-Fighting

6741-20, Session 5
Remote physiological assessment using laser doppler vibrometry
J. W. Rohrbaugh, J. A. Stern, E. J. Sirevaag, W. D. Richard, S. Kristjansson, Washington Univ. (USA); J. A. OSullivan, Washington Univ. in St. Louis (USA) We describe a novel application of laser Doppler vibrometry (LDV) to the remote assessment of physiological activity. Many physiological activities have mechanical components (in the form of vibrations, sounds and pulses) that can be detected at the skin surface using the LDV method. The exceptional sensitivity of the LDV method supports sensing of cardiovascular, respiratory, muscle and tremor activity, thus providing a basis for comprehensive and differential assessment. We have validated the measures within context of conventional methods, and have demonstrated their sensitivity to a variety of laboratory maneuvers as well as clinical status. No preparation, contact or specific posture is required, and data can be obtained at extended offsets (\> 25 meters). These capabilities have been integrated into a portable system (ca 10 lbs, battery powered and with a notebook computer) that includes video tracking of unconstrained examinees. We have confirmed that a system using an infrared laser (invisible, eye safe) is technically feasible. Our principal concern has been with cardiorespiratory measures. Gross ballistic cardiorespiratory activity can be assessed from clothing or other body wear, but direct imaging from the skin-particularly over the carotid artery-provides an extremely detailed analog to the internal pressure pulse. The pulse consists of a highly textured waveform that includes multiple incident and reflected pressure fronts. Collectively, these signals yield measures of not only heart rate, but also advanced measures of contractility, systolic time intervals, pressure pulse amplitude and arterial dynamics including signs of cerebral perfusion. The signals are of sufficient quality to support beat-to-beat assessment, are selectively responsive to social, physical and cognitive stressors, and provide advanced clinical information relating to cardiac and vascular health including cardiorespiratory interactions. Several aspects of respiration can be sensed, including respiratory effort as well as lung and tracheal sounds. The LDV method provides a basis for comprehensive, sensitive and differential assessment of a number of key physiological signs of stress, emotion, intent, and health.

6741-23, Session 6
Optical digital chaos cryptography
A. Arenas-Pingarrn, A. P. Gonzalez-Marcos, J. A. MartinPereda, Univ. Politcnica de Madrid (Spain) Several schemes have been proposed for secure communications with different types of chaos generators. Specific approaches have to be considered when optical signals are involved. The more usual configurations are based on Pecora and Carroll papers, where the information signal is added in the emitter to a chaotic signal and the resulting one is processed at the receiver with other chaotic signal. Both signals era generated by synchronized generators. But in any case a time-frequency analyzes shows how to recovery the information if it is not uses in CDMA configuration. We present a new way to codify the information signal in two steps. In the first one, we add the phase information to the initial signal with a lineal photonic device. This signal is later modulated, in the second step, with a chaotic signal. The chaotic signal is obtained from a photonic generator based on an Optically Programmable Logic Cell, previously reported by us. This cell is able to provide a digital signal without any additional operation, as it usual in other optical chaos generators. The resulting signal is decode in the receiver with a new chaotic signal generated by synchronizing both generators, at emitter and receiver. This situation has been reported previously by us. Involved signals are analyzed by timefrequency distributions that show it is not possible to obtain data in a first approach.

6741-24, Session 6
Optical encryption and encrypted holographic storage using phase-only data pages
P. Koppa, T. Sarkadi, F. Ujhelyi, J. Remnyi, G. Erdei, E. Lricz, Budapest Univ. of Technology and Economics (Hungary) Using phase only input images has several advantages in optical encryption and holographic storage systems: It provides a smooth Fourier plane that allows making an efficient use of the material dynamic range, it makes possible object plane phase encoding and enhances the encryption strength and noise tolerances of double random coding. However usual phase-to amplitude conversion methods require bulky and very sensitive interferometric setups. In this paper we propose a new simple and compact phase to amplitude data page conversion method for efficient recovery of the data encoded in phase modulated data pages used in optical encryption and holographic storage. The method is based on the interference between the data page and its copy shifted by an integral number of pixels. Key properties like Fourier plane homogeneity, bit error rate and positioning tolerances of the proposed approach are investigated by computer modeling and a comparison is provided with amplitude modulated data pages. The feasibility and the basic properties of the proposed method are experimentally demonstrated. Experimental results of double random phase encryption of phase modulated data prove the long high key-length and very high tampering resistance of the method. We also propose an optical implementation with a single phase SLM, showing reduced system complexity and small size compared to amplitude modulated systems.

6741-21, Session 6
Photo-luminescent quantum-dots used for security identification
S. Chang, K. Yu, J. Liu, National Research Council Canada (Canada) Information retrieval is critical in security technologies such as for status identification and documentation authentication. Ideally, coding materials should be difficult to locate, impossible to counterfeit, and easy to process. This presentation addresses a novel information retrieval technology with these ideal features of its coding materials: the photo-luminescent (PL) quantumdots (QD) synthesized via wet-chemistry approaches. As compared to traditional PL materials, they exhibit emission with narrower full width at half maximum, greater brightness, and higher photo-stability; also, their PL wavelength can be easily and accurately tuned via their size, structure, and composition. Due to such a feasible tune-ability, mainly, QDs have demonstrated enormous potential applications in security and defense. When QDs are excited, they can provide coded information with their PL wavelength and intensity. If the coding wavelengths from the QD PL are designed as the Fraunhoffer lines, i.e. black lines in solar spectrum, the retrieval system can extract the useful information even under sunshine covering areas. Multi-photon excitation (MPE) technologies can further extend applications of QDs to multi-layer information extraction. For an info-label of 2-millimeter in depth, a MPE system with the depth resolution less than one micro-meter can thus achieve 2 GB resolutions, when a coding material exhibiting 6 PL wavelengths with 10 intensity levels. In general, transparent thin-film coating of QDs can be applied to various substrates, such as documents, fingernails, and military helmets and vehicles. Moreover, QD based security information can be easily destroyed by preset expiration in the presence of timing agents.

6741-25, Session 6
Theory of dispersion-managed solitons
M. F. S. Ferreira, Univ. de Aveiro (Portugal) New ways in optimization of existing telecommunication systems based on dispersion management technology attracted recently wide research interest from soliton-based groups. The main idea was to combine a high local groupvelocity dispersion with low path-average dispersion. The former feature results in the reduction of the four-wave mixing while the latter one reduces the Gordon-Haus timing jitter effect. Due to their characteristics, dispersion-managed (DM) solitons offer tremendous advantages that make them a preferred option for upgrading the embedded fiber plant and for use in new ultrahigh-speed multiplexed systems operating at 40-Gbit/s per channel. While numerical simulations of the governing equations

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

43

Conference 6741: Optics and Photonics for Counter-Terrorism and Crime-Fighting


accurately capture the pulse dynamics of dispersion-managed systems, serious design optimization requires intensive computation. This process is generally much too time consuming and there is no guarantee of obtaining a stationary solution that corresponds to the desired pulse characteristics. In this paper we use a variational approach to reduce the governing equation in the case of a periodic dispersionmanaged fiber system to a coupled set of nonlinear ordinary differential equations (ODEs). Numerical simulations show that the ODE model accurately captures the pulse dynamics as governed by the dispersion-managed nonlinear Schrdinger equation (DMNLSE). Moreover, the adiabatic evolution of the dispersion-managed pulse parameters under perturbations is obtained. These dynamical systems of the pulse parameters are used to estimate the effects of several perturbations on the soliton propagation, such as the amplifier noise, higher-order dispersion, nonlinear interaction with other solitons, etc. The obtained results are valid for an arbitrary pulse profile. However, we will consider particularly the Gaussian and the superGaussian type pulses. applications by showing initial results on explosive compound detection. This paper will present the performance and results from a QC laser based people screening portal developed and tested over the past 12 months and aimed at the detection of precursors used in the make up of improvised explosive devices (IED). The tests carried out on a large number of potential interferers together with the target materials reinforce original assumptions that compound fingerprinting can be carried with the technique, achieving extremely high degree of specificity within sub second response time. Furthermore test results of the portal prototype show that not only precursor compounds on their own or mixed with interferers can be detected but that their signature can also be seen in IED such as TATP or other intermediate stages in the make up of IED. The initial portal sensitivity configuration was set up for few parts per billion (ppb) detection level thresholds. It is currently being upgraded to sub-ppb level. Overall, the early results obtained from the QC laser based portal seem to indicate that it is possible to develop low cost detection systems with enhanced features such low false positive and high throughput screening of individuals or items. The tests were carried out with support from the UK government.

6741-26, Session 6
A new type protection system
B. N. Kilosanidze, G. A. Kakauridze, Institute of Cybernetics (Georgia) With the purpose to increase a level of protection of important documents, securities, industrial goods, etc. from counterfeiting a new type of protection system based on polarizationholographic methods is suggested. The polarization hologram is able to record all characteristics of the polarization ellipse of the field that is being recorded. Such redundancy of the initial data recorded on the material carrier makes it possible to record physically ultimate complete information respectively. The suggested polarization-holographic system of protection is based on polarization properties of light and the use of specially synthesized polarization-sensitive materials for protective elements recording. In polarization-sensitive materials the anisotropy of optical characteristics is induced under the action of polarized light. The functions of scalar (isotropic) and two vector (anisotropic and gyrotropic) reactions on the action of actinic polarized light were introduced to describe the photoresponse of such media. The values of reactions uniquely characterize polarization-sensitive material. Using various technologies of making polarization-sensitive materials, it is possible to change their characteristics and the response to acting polarized light. The recording of a protective element is made by two writing beams with the use any predetermined combination of polarization state from infinite number of possible combinations. For each concrete system of protection the material with certain characteristics and its own combination of polarization state can be chosen to which diffracted beams which are concrete according to polarization state will unambiguously correspond. Visually the protective elements look absolutely homogeneous. It is impossible to determine information that is written on the given protective element by any usually used methods of authenticity determination. Only while using the special identifying device developed by us it is easy to determine the authenticity of a protective element by comparing the polarization state of light beams diffracted on the protective element with the etalon from database of the identifying device. Unlike holographic systems of protection existing now, an essential advantage of the suggested system is the impossibility of copying such protective elements by optical methods. This system is simple enough and noiseprotected and gives the possibility to essentially increase a level of protection.

6741-28, Session 7
Raman detection of illicit materials using portable equipment
W. E. Smith, Univ. of Strathclyde (United Kingdom) Raman spectroscopy provides a very effective method of identifying an illicit substance in situ without separation or contact other than with a laser beam. The equipment required is steadily improving and is now reliable and simple to operate. Costs are also coming down and hand held portable spectrometers are proving very effective. Examples of the use of hand held equipment to detect powdered and liquid samples will be given. The main limitations on the use of the technique are that it is insensitive in terms of the number of incident photons converted into Raman scattered photons and fluorescence produced in the sample by the incident radiation interferes. Simple stratagems to overcome these problems will be discussed. Newer methods still largely in the development phase will increase the potential for selected applications. The use of picosecond pulsed lasers can discriminate between fluorescence and Raman scattering and this has been used to examine street samples of illicit drugs. Surface enhanced Raman scattering in which the analyte requires to be adsorbed onto a roughened metal surface creates a sensitivity to compete with fluorescence and quenches fluorescence for molecules on a surface. This provides the ability to detect a sample coded with a dye mixture at a distance with hand held equipment and the ability to detect very low concentrations of some explosives after derivatisation. The improving optics and detection capability and the reliability of the new methods indicate that the potential for the use of Raman spectroscopy for security purposes will increase with time.

6741-29, Session 7
Diffuse reflection imaging at terahertz frequencies for security applications
P. Dean, S. P. Khanna, S. Chakraborty, M. Lachab, E. H. Linfield, A. G. Davies, Univ. of Leeds (United Kingdom) We report diffuse reflection imaging of concealed powdered samples in atmospheric air using a quantum cascade laser operating at 2.83 THz. The imaging system uses a heliumcooled silicon bolometer for mapping radiation diffusely reflected and scattered from samples, and a room temperature pyroelectric sensor for simultaneously acquiring a specular image. A range of powders concealed within plastic packaging and standard FedEx envelopes were imaged with a resolution of better than 0.5 mm, and it was possible to detect powdered samples concealed within packaging from which there was a strong component of surface reflection. The feasibility of performing dual wavelength diffuse reflection imaging for identification of illicit drugs and explosives is discussed.

6741-27, Session 7
Quantum cascade laser-based screening portal for the detection of explosive precursors
E. L. Normand, Cascade Technologies Ltd. (United Kingdom) Over the past few years, quantum cascade (QC) lasers based gas sensors have been proven in simple high-performance gas analyzers aimed at the continuous emission monitoring (CEM) market and such able to operate in harsh environments. Nearly a year to date Cascade Technologies reported on progress towards adapting its technology for homeland security

44

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6741: Optics and Photonics for Counter-Terrorism and Crime-Fighting

6741-30, Session 8
Fenestration obscuration techniques
M. Smalley, Security Services Group (United Kingdom) There are situations where it is advantageous to visually obscure through glass, to an external observer, the movement of people within a well lit room. It may be that the building use has changed or existing measures which had provided obscuration such as Bomb-blast curtains have been discontinued. Recognising that implemented solutions must create the minimum disruption to outward visibility and involve the least procedural effort (be simple to use), the Centre for Protection of National Infrastructure, CPNI, commissioned this study, defining key requirements including: (a) Automatic or simple manual operation (b) Obscuration of movement within the building (c) Varying levels of obscuration depending on the difference in internal and external light levels. (d) Minimum disruption to outward visibility (e) Acceptable for use on heritage and iconic sites (f) Easy to retrofit (g) Low cost This report reviews earlier work carried out into the protection of Guardrooms by the use of lighting techniques coupled with the use of reflective and screen printed films. Other innovative solutions including Electrochromatic controllable glazing which may prove more appropriate to office and commercial buildings are also considered. It is seen that some measures, (window films or blinds), are cost effective and unsophisticated while more complex automatic systems using reactive glazing can offer critical design advantages. It must be noted however, some of the key requirements are mutually exclusive and any solution chosen will always be a compromise based on client needs and circumstances. Practical evaluation of different solutions is ongoing including examination of resulting physical properties of glazing to ensure that solutions to this problem do not conflict with other measures such as bomb-blast protection. Results will, where appropriate, be included in the final paper presented.

6741-33, Session 8
Integrated multi-sensor perimeter detection systems
P. Fretwell, P. J. Kent, D. Barrett, QinetiQ Ltd. (United Kingdom) The presentation describes the results of a multi-year programme of research aimed at the development of an integrated multi-sensor perimeter detection system capable of being deployed at an operational site. At the beginning of the programme a requirements survey identified areas of protective security, particularly in threat detection and assessment, where effective capability was either not available or prohibitively expensive. The development programme was designed to prove the feasibility and utility of cost-effective systems, based on grouped technologies, in addressing these capability gaps. Some of these techniques include novel video based biometrics to improve performance against human subjects and use multiple technologies to give better indication of a threat than any single technology. This has culminated in a modular integrated architecture based on commercially available components. Special video analytics have been designed to provide robust detection of pedestrians in clutter while new radar detection and tracking algorithms provide wide area day/night surveillance. The fusion of video, radar and other sensor data provides the basis of a threat detection capability for real life conditions. A graphical user interface has been developed to visualise and interact with the sensors. The system was designed to be modular and extendable in order to accommodate future and legacy surveillance sensors. The current sensor mix includes stereoscopic video cameras, mmWave ground movement radar, CCTV and a commercially available perimeter detection cable The presentation will outline the development of the system and describe the lessons learnt after deploying the system in a pilot trial.

6741-34, Session 8
Implementing advanced image processing technology in sensor systems for security and surveillance
D. L. Hickman, M. I. Smith, Waterfall Solutions Ltd. (United Kingdom); P. K. Kimber, SELEX Sensors and Airborne Systems Ltd. (United Kingdom) The range and scope of EO/IR sensor systems within security and surveillance applications is growing, and this places a corresponding demand on the image processing functionality required to meet the end-users needs and requirements. Within this paper, the application of different image processing architectures and techniques is reviewed in terms of situational awareness criteria and is illustrated through specific system applications. The concepts and benefits of multi-modal and distributed sensor systems are also considered together with the attendant data registration and fusion techniques. Finally, the exploitation of a priori information within the integrated security and surveillance picture is considered from both a processing technology and image display perspective.

6741-31, Session 8
Covert optically-scanning enhanced zoom pinhole lens technology
H. S. Rana, Defence Science and Technology Lab. (United Kingdom) There is a desire to carry out covert surveillance in buildings using pinhole lenses that offer a wide field of view for coverage but also offer a zoom capability that allows targets to be identified. It is further desirable to pan and tilt onto the target if it is at the edge of the field. Lens technology has been developed that incorporates an internal pan and tilt function in a pinhole zoom. Furthermore the lens technology offers a stand-off distance that allows easy installation.

6741-32, Session 8
Spectral Vision Research (SVIR) for homeland security applications
P. W. Yuen, Cranfield Univ. (United Kingdom) Low reflectance materials in the visible spectral band such as dark coloured vehicles and clothing are found difficult to be detected and classified accurately by using spectral detection techniques such as hyperspectral imaging (HSI). The relatively featureless spectral characteristics of these dark coloured targets have made them hard to be distinguished from the background such as tarmac, as far as spectral measurement methods like spectral angle mapper (SAM) and matched filter (MF) detection as concern. In this paper it is reported that by extending the spectral range into short wave infrared (SWIR) and couple with a band selection scheme, these dark coloured targets can be readily detected and classified accurately. Example data taken from a car park will be utilised to illustrate the effectiveness of monitoring dark coloured targets using HSI/ multispectral techniques under a wide range of weather conditions.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

45

Conference 6742: Remote Sensing for Agriculture, Ecosystems, and Hydrology


Wednesday-Thursday 19-20 September 2007 Part of Proceedings of SPIE Vol. 6742 Remote Sensing for Agriculture, Ecosystems, and Hydrology IX

6742-12, Poster Session


Climate changes impact the surface albedo of a forest ecosystem based on MODIS satellite data
M. A. Zoran, National Institute of Research & Development for Optoelectronics (Romania) Surface albedo is one of the most important biophysical parameter responsible for energy balance control and the surface temperature and boundary-layer structure of the atmosphere. Forest land surface albedo is also highly variable temporally showing both diurnal as well as seasonal variations. In forest systems, albedo controls the microclimate conditions which affects ecosystem physical, physiological, and biogeochemical processes such as energy balance, evapotranspiration, photosynthesis. Due to anthropogenic and natural factors, land cover and land use changes result is the land surfaces albedo change. The main aim of this paper is to investigate the albedo patterns due to the impact of atmospheric pollution and climate variations of a forest ecosystem placed to the North-East of Bucharest city, Romania based on satellite remote sensing Moderate Imaging Spectroradiometer (MODIS) data and climate station observations. Our study focuses on 3 years of data (2004-2006), each of which had a different climatic regime. MODIS Terra and climate station observations show that surface albedo changes of a forested zone placed close to a large urban area highly respond to atmospheric pollution influence and climate variations.Since surface albedo dynamics in forest ecosystem are close related with ecosystem dynamics, the impacts of atmospheric pollution and climatic changes and variations on ecosystem processes could possibly affect surface albedo characteristics. As the physical climate system is very sensitive to surface albedo, forest ecosystems could significantly feedback to the projected climate change modeling scenarios through albedo changes. The results of this research have a number of applications in weather forecasting, climate change, and forest ecosystem studies.

Estimation of AE on fields ranged from 0.72 for cotton to 0.90 for sugar beet. Moreover, irrigation consumption was estimated by METRIC with high accuracy considering global results (real average water use was 699 mm while the irrigation consumption estimated by METRIC was 677 mm, equivalent to an estimation error of 3%).

6742-44, Poster Session


Using MODIS/NDVI imagery for the validation and calibration of a live vegetation moisture content model
P. M. Fiorucci, F. F. Rosso, F. M. Gaetani, R. M. Minciardi, Ctr. di Ricerca Interuniv. in Monitoraggio Ambientale (Italy) Dead and live fine vegetation moisture dynamics plays a key role in wildland fire risk assessment. While dead fuel moisture dynamics can be considered only dependent on the dynamics of the meteorological variables, live fuel dynamics models must take into account also the current phenological state of the considered species. The short-time variations on moisture of live vegetation are less significant than the dead ones. However, the great amount of fuel load potentially interested by fire represents a amplifying factor in wildfire risk assessment. This work presents a model to predict the moisture content of live fine fuel based on the phenological principles of leaf growth cycle. This approach assumes that the phenology state of a given vegetation group at a given time instant can be modelled as a function of the local meteorological conditions, soil and vegetational parameters (leaf area index). The Vegetation Index products of MODIS sensor have been used to parameterize and calibrate the model. To this end, 40 different experimental stations, equipped with complete meteorological stations, and fully representative of the Mediterranean vegetation cover of Liguria Region (Italy), have been selected and used as test points. These 40 areas are periodically subject to sampling campaign aiming at characterizing the phenological state and the moisture contents. In addition, for each area the observations of MODIS-NDVI from 2001 to 2006 have been collected and geocoded. The paper provides a validation and calibration procedure, carried on using the whole data sets, and formalized through a mathematical programming approach.

6742-43, Poster Session


Integration of satellite-based energy balance with simulation models applied to irrigation management at an irrigation scheme of southern Spain
C. Santos, I. Lorite, Instituto de Investigacin y Formacin Agraria y Pesquera (Spain); M. Tasumi, Univ. of Miyazaki (Japan); R. G. Allen, Univ. of Idaho (USA); P. Gavilan, Instituto de Investigacin y Formacin Agraria y Pesquera (Spain); E. Fereres, Univ. of Crdoba (Spain) Integration of a satellite-based energy balance as METRIC (Mapping EvapoTranspiration with high Resolution and Internalized Calibration) model with a simulation model based on cascade approach provided accurate irrigation scheduling guidelines for individual fields. This procedure was applied during 2004/05 irrigation season to Genil-Cabra Irrigation Scheme in Southern Spain. The relatively high frequency of images collected in this study (eleven Landsat 5 TM images) produced a high level of confidence in ET estimates, especially during the period of high irrigation requirements. The satellitederived ET was used to correct the daily soil water balance model, improving the accuracy of field-by-field ET demands and subsequent field-scale irrigation schedules. Thus, a reduction in water use of 23% was obtained for cotton by following the improved irrigation schedules. Seasonal ET variability among fields and within fields determined the irrigation management by crops. Variation coefficient of ET from sunflower fields was 0.28, while for cotton fields was 0.12. Irrigation application efficiency (AE), calculated using satellite-based ET and actual irrigated water measurements, helped to identify specific agricultural fields experiencing problems in water management, as well as to estimate general AE of the scheme by irrigation and crop type.

6742-45, Poster Session


Analysis of surface biophysical parameters of urban system derived from satellite data
M. A. Zoran, National Institute of Research & Development for Optoelectronics (Romania); C. H. Weber, Univ. Louis Pasteur (France) Bucharest town, the biggest industrial, commercial center in Romania has experiencing rapid an urban expansion during the last decades. A large amount of forest and agricultural land has been converted into housing, infrastructure and industrial estates. The resultant impervious urban surface alters the surface energy balance and surface runoff, which in turn could pose serious environmental problems for its inhabitants (e.g., urban waterlogged and thermal pollution). Multi-spectral and multi-temporal satellite imagery provide the most reliable technique of monitoring of different urban structures regarding the net radiation and heat fluxes associated with urbanization at the regional scale. Investigation of radiation properties, energy balance and heat fluxes is based on satellite data from various satellite sensors and in-situ monitoring data , linked to numerical models and quantitative biophysical information extracted from spatially distributed NDVI-data and net radiation. Based on LANDSAT TM and ETM , IKONOS , MODIS and SAR satellite images over Bucharest urban area, Romania were analyzed surface biophysical parameters for 1989 - 2005 period. Spatio-temporal changes of surface biophysical parameters were examined in association with

46

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6742: Remote Sensing for Agriculture, Ecosystems, and Hydrology


landuse changes to illustrate how these parameters respond to rapid urban expansion in Bucharest and surrounding region. This study attempts to provide environmental awareness to urban planners in future urban development. The land cover information, properly classified, can provide a spatially and temporally explicit view of societal and environmental attributes and can be an important complement to in-situ measurements. ecological relationships between wildland fire and landscape structure. Remote sensing data provide valuable information for the characterization and mapping of fuel types and vegetation properties at different temporal and spatial scales including the global, regional and landscape levels. This study aims to ascertain how well remote sensing data can characterize fuel type at different spatial scales in fragmented ecosystems. For this purpose, satellite ASTER and TM data were processed using neural nets for a test area of Southern Italy characterized by mixed vegetation covers and complex topography. Fieldwork fuel type recognition, performed at the same time as remote sensing data acquisitions, were used to assess the results obtained for the considered test areas. Results from our preliminary analysis showed that the use of neural nets provides detailed mapping of fuel types with accuracy levels higher than 75%.

6742-46, Poster Session


MODIS and Landsat TM data image fusion based on improved resolution method: assessing the quality of resulting NDVI images
J. Park, Chungbuk National Univ. (South Korea); S. La, SUNDOSOFT, Inc. (South Korea) The monitoring of vegetation in nearby urban regions is made difficult by the low spatial and temporal resolution image captures. Image fusion is one of the important techniques for spatial image resolution enhancing. In order to utilize respective information from different remote sensing images, we propose an image fusion method based on improved resolution method. Recent studies have successfully estimated NDVI using improved resolution method such as from the MODIS onboard EOS Terra satellite. Enhancement of MODIS NDVI image using Landsat TM image is performed on various images. The results demonstrate accurate spectral preservation on vegetated regions where MODIS image enhances the fusion product, which can be usefully applied for both visual analysis and classification purposes. Subjective visual effect and objective statistical results indicate that the performance of the improved resolution method is better than original MODIS images. It not only preserves spectral information of the original multi-spectral image well, but also enhances spatial detail information greatly. To provide a continuous monitoring capability for NDVI, in situ measurements of NDVI from paddy field was carried out in 2006 for comparison with remotely sensed MODIS data. We compare and discuss NDVI estimates from MODIS sensors and in-situ spectroradiometer data over Ochang plain region. These results indicate that the MODIS NDVI is underestimated by approximately 50%.

6742-49, Poster Session


Correlation analysis of simulated MODIS vegetation indices and the red edge and rice agricultural parameter
Q. Cheng, X. Wu, Zhejiang Gongshang Univ. (China) In this study, Hyperspectral data of two variety of rice (common rice and hybrid rice) in whole growing stage during 2002 and 2003 was measured using the ASD FieldSpec UV/VNIR Spectroradiometer with resolution of 3 nm, and the LAI and leaf chlorophyll content of rice agricultural parameter were obtained. Analyses of the correlation between rice agricultural parameter, and hyperspectal data, normalized difference vegetation index (NDVI), enhanced vegetation index (EVI) and the red-edge position (REP) were studied. Results showed that a strong non-linear correlation was found between the rice LAI of two varieties and REP. The REP, EVI and NDVI were well related with LAI for the common rice, but the REP and EVI were more sensitive than the NDVI to rice LAI for the hybrid rice because of different body for two variety rice.

6742-50, Poster Session 6742-47, Poster Session


Estimating land surface evapotranspiration from ASTER data using a novel approach
Y. Liu, Y. Yamaguchi, T. Hiyama, Nagoya Univ. (Japan) The major difficulty in estimating land surface evapotranspiration is partitioning of available energy into sensible and latent heat fluxes. Based on the principles of energy balance, equilibrium evaporation, and non-equilibrium thermodynamics, we proposed a novel approach for retrieval of evapotranspiration from satellite data. We applied this approach to a semiarid area located at the Loess Plateau of China (35.2N, 107.7E). Considering the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) can provide land surface temperature data with a high accuracy of 1.5K, we used ASTER product datasets to estimate surface evapotranspiration for the area on May 10, 2005. Alternatively, taking advantages of the 30-m tower observation, we calculated the footprint area using a flux footprint model. We found that the aeral average of the satellite-retrieved evapotranspiration was in good agreement with that of the tower observation. Because the proposed approach is very simple, it is highly valuable to remote sensing applications. With the development of the increased accuracy of satellite retrieval techniques for the necessary input variables, this approach offers a practical way with a high accuracy for long-term monitoring of surface evaporation on either regional or global scales.

Application of reflectance spectra of two-layer leaf systems to chlorophyll estimation in crops


T. Kazantsev, S. M. Kochubey, Institute of Plant Physiology and Genetics (Ukraine); V. Donets, Public Enterprise Plant Arsenal (Ukraine) Hardware-software complex for chlorophyll estimation under field conditions have been designed. It is based on measuring and processing of crops reflectance spectra using the original derivative vegetation indices. As the indices have been developed on a base of reflectance spectra of separate leaves, it is necessary to adjust the procedure of index calculation for data obtained under field measurements. There are at least two reasons causing distinctions between spectra of separate leaf and crop: superposition of reflectance from several leaf layers and contribution of soil reflectance. Influence of these factors has been studied in this work. Reflectance spectra of two-layer leaf systems were measured. The systems included two superposed vine leaves with various chlorophyll contents on different background. Reflectance and transmittance spectra of each separate component of the systems were also measured. Two derivative indices were calculated using green and red regions of reflectance spectrum. Presence of lower leaf caused an increase of reflectance in red and much less in green regions. Chlorophyll in lower leaf influenced only on the red index but not on the green one. Both indices were found to be low sensitive to contribution of soil reflectance even at low extent of projective covering. Features of two-layer system reflectance spectra were explained with their simulation using reflectance and transmittance of separate components of the systems. Results obtained gave approach for chlorophyll estimation in upper and lower leaves separately. The approach is tested under field measurements using our device.

6742-48, Poster Session


Satellite multispectral fuel type mapping by using neural nets
R. Coluzzi, I. di Donna, Univ. degli Studi della Basilicata (Italy) In the context of fire management, fuel maps are essential information requested at many spatial and temporal scales for managing wildland fire hazard and risk and for understanding

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

47

Conference 6742: Remote Sensing for Agriculture, Ecosystems, and Hydrology

6742-51, Poster Session


A new approach to evaluate the performance of agricultural ecosystem using MODIS remote sensing data in China
Z. Qin, M. Gao, J. Qiu, H. Tang, Chinese Academy of Agricultural Sciences (China) Agricultural ecosystem is a complicated system with various factors determining its functioning and performance. Evaluation of agricultural ecosystem performance has been an important academic area in recent decades. Currently this evaluation is mainly done through the models proposed on the basis of socioeconomic statistics data. This limits the evaluation to a region as a unit without considering the spatial variation and temporal dynamics of the region. With the development of space technology, it is possible to monitor the ecosystem performance spatially and dynamically for the entire China using MODIS remote sensing data. In this study we proposed a new approach to evaluate the performance of agricultural ecosystem in the whole China. The approach is an index method with data inputs from remote sensing and Geographical Information System (GIS). Considering key factors determining the functioning of agricultural ecosystem in a large scale, we choose the following five indicators to develop a model for the evaluation: vegetation index, land surface temperature, precipitation index and disaster index. These factors indicate the performance and functioning of agricultural system and have been considered in the study as the key factors in evaluation of the system in a large scale. Using the approach, we integrated the five indicators into a model with vegetation index and land surface temperature retrieved from MODIS data and precipitation index and disaster index from GIS database of agricultural statistics. Performance of agricultural system of the entire China from March to October for every 10-days in 2006 has been evaluated in the study. The results indicate that the performance of the agricultural ecosystem in general is very encouraging in the whole China. Only a few regions within specific period of time, for example, Guangdong and Guangxi in spring and Chongqing in august, have low level of performance due to severe disasters. Taking the first 10-days of May as a case, we make a thorough analysis of the performance for various regions and present the detailed results in the study. Our study confirms that the applicability of the proposed approach for a rapid evaluation and monitoring of agricultural ecosystem performance in the entire China.

6742-53, Poster Session


Snow mapping for water resource management using MODIS satellite data in Northern Xinjiang Basin, China
H. Pei, Nanjing Univ. (China); Z. Qin, Chinese Academy of Agricultural Sciences (China) Snow is the most important freshwater resource in Northern Xinjiang Basin, which is a typical inland arid ecosystem in western China. Snow mapping can provide useful information for water resource management in the arid ecosystem. In this paper we intend to develop an applicable approach for snow mapping in Northern Xinjiang Basin using MODIS. We conducted field campaigns to measure snow depth of 80 sampling sites around 35 hydrology stations in the study region. Geographical coordinates of the sampling sites were also recorded using a GPS device. Geometric correction was done to the selected MODIS images of the study region for georeference registration to match the geographical coordinates of the sampling sites. Reflectivity of these images was computed from their DN values for each band. After retrieval, we analyzed correlation between the image reflectivity and the measured snow depth for the 80 sampling sites. An algorithm had been developed on the basis of the correlation for snow mapping in the region. Validation for another dataset with 24 sampling sites resulted a RMS error of 1.63, indicating that the algorithm was able to provide an estimate of snow depth at an accuracy of 1.63cm in the region. Finally we applied the algorithm to the study region for snow mapping. Our results indicated that average snow depth in north Xinjiang was 13.8 cm in the winter of 2005. Spatial variation of the snow depth was very obvious, depending on such ground conditions as vegetation cover, soil type, and slope directions. Generally speaking, the depth had a trend of gradually decreasing from north to south, from east to west and from the surroundings to the center. Temporally, the depth reached maximum in January, the coldest month in the region. Our study showed that the spatial and temporal variation of snow cover was very critical for water resource management in the arid inland region and MODIS satellite data provide an alternative for snow mapping through dedicate development of mapping algorithm suitable for local application.

6742-54, Poster Session


Validating the MODIS LAI product by scaling up LAI measurements at a VALERI alpine meadow site, China
M. Ma, Cold and Arid Regions Environmental and Engineering Research Institute (China); F. Veroustraete, Flemish Institute for Technological Research (Belgium); L. Lu, X. Li, Cold and Arid Regions Environmental and Engineering Research Institute (China); R. Ceulemans, Univ. Antwerpen (Belgium); J. Bogaert, Univ. Libre de Bruxelles (Belgium); C. Huang, T. Che, Cold and Arid Regions Environmental and Engineering Research Institute (China); Q. Dong, Flemish Institute for Technological Research (Belgium) The sampling protocol adopted during a field campaign at an Alpine meadow site (Shandan site), during July 2002 is based on the so-called Valeri protocol (VALERI). The field campaign LAI measurements in Shandan are scaled up to 30~30 m(c) raster maps based on Landsat ETM+ imagery. Regression analysis is applied to construct empirical transfer functions for the determination of Leaf Area Index (LAI) raster imagery from ETM+ Normalized Difference Vegetation Index (NDVI) and Simple Ratio (SR) data. Subsequently, the scaling up of the LAI raster maps is performed by the aggregation of the 30x30 m(c) data into 1~1 km(c) pixels by calculating the average LAI values for the low resolution pixels. The up-scaled data are used to validate the MODIS LAI product at the Shandan site. A power regression model (LAI=2.3758*NDVI3.5216, R2=0.66, P<0.01), established between field measured LAI and ETM+ NDVI, elicits a high statistical significance. A linear regression model (LAI=0.1798*SR-0.3574, R2=0.55, P<0.01) is established between field measured LAI and ETM+ SR. The MODIS LAI product correlates best with the ETM+ LAI transfer function obtained with NDVI data. Its R2 reaches 0.46, its slope 0.97, but the intercept is 0.7, which suggests that MODIS LAI is systematically underestimated. The results illustrate that LAI measured with a LAI-2000 instrument at the VALERI Shandan

6742-52, Poster Session


Land surface temperature retrieval from MODIS data for agricultural drought monitoring in China
Z. Qin, M. Gao, J. Qiu, H. Tang, Chinese Academy of Agricultural Sciences (China) Water and heat are two important factors governing dynamic changes of soil moisture available for crop growing in farmland, hence become the basic indicators for agricultural drought monitoring. Due to its directly relating to soil moisture and heat flux characteristics, thermal remote senisng has been extensively applied to agricultural drought monitoring through soil water content estimation. In this paper, we take the MODIS data extensively employed in agricultural drought monitoring as an example to present an applicable approach for land surface temperature retrieval, with fucos on algorithm selection, parameter determination, and computation processes. Though MODIS data contains 8 thermal bands for earth observation, bands 31 and 32 are especially suitable for land surface temperature (LST) retrieval required by agricultural drought monitoring. After comparing their accuracy, we select two-factor split window algorithm for the retrieval. Then we discuss estimation of the algorithms essential parameters for LST retrieval, especially how to fastly estimate the ground emissivity and atmospheric transmittance from other bands of the same scene MODIS data. Finally we conduct an experiment of applying the approach to China for LST retrieval in 2006. The results from our preliminary applications confirms the applicability of the proposed methodology in retrieving surface temperature parameters from MODIS data required for agricultural drought monitoring in China.

48

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6742: Remote Sensing for Agriculture, Ecosystems, and Hydrology


site leads to an underestimation of the MODIS LAI product. A plausible cause for the systematic underestimation related with the LAI field measurements is discussed.

6742-57, Poster Session


The spatiotemporal analysis of snow depth on Qinghai-Tibet Plateau based on passive microwave remotely-sensed data
T. Che, X. Li, R. Jin, Cold and Arid Regions Environmental and Engineering Research Institute (China); R. L. Armstrong, T. Zhang, Univ. of Colorado/Boulder (USA) It is well known that snow can play an important role in the climate change and water resources in the cold land process. For obtaining the time sires data of snow on the Qinghai-Tibet Plateau, we validated the global snow retrieval algorithm from passive microwave brightness temperature data according to the snow depth observations of meteorological stations. It had shown that the global snow depth algorithm had a visible overestimation on the Qinghai-Tibet Plateau. On the other hand, we considered other scattering signals similar to the snow cover, such as precipitation, cold desert, frozen ground, which can make the snow retrieval complicated. Therefore, we modified the algorithm of global snow retrieval, and established a snow retrieval flow chart by using a snow classification tree. The snow flow chart only needs the passive microwave brightness temperature. We used the flow chart to retrieve the time sires snow datasets from 1978 to 2006, which covered the all of SMMR and SSM/I periods. For the accuracy assessment of the new method, the comparisons to the MODIS and AMSR-E snow products also showed the good agreement with this study. Furthermore, the spatiotemporal variations of snow were analyzed in detail. In the temporal, the snow storage on the Qinghai-Tibet Plateau had shown a likely increasing trend, though the snow cover on the north hemisphere has a week decreasing in past several decades. In the spatial, the snow storage mainly distributed the South-eastern Qinghai-Tibet Plateau, which dominates the snow mass in a whole. According to the digital elevation model (DEM) data, the correlations between snow and elevation and topographical indexes also were discussed in this study.

6742-55, Poster Session


Managing grain protein content by remote sensing in winter wheat
W. Huang, National Engineering Research Ctr. for Information Technology (China) Advanced site-specific determination of grain protein content by remote sensing can provide opportunities to optimize the strategies for purchasing and pricing grain, and to maximize the grain output by adjusting field inputs. Field experiments were performed to study the relationship between grain quality indicators and foliar nitrogen concentration. Foliar nitrogen concentration at the anthesis stage is suggested to be significantly correlated with grain protein content, while spectral vegetation index is significantly correlated to foliar nitrogen concentration around the anthesis stage. Based on the relationships among nitrogen reflectance index (NRI), foliar nitrogen concentration, and grain protein content, a statistical evaluation model of grain protein content was developed. NRI proved to be able to evaluate foliar nitrogen concentration with a coefficient of determination of R2= 0.7302 in year 2002. The relationship between measured and remote sensing derived foliar nitrogen concentration had a coefficient of determination of R2=0.7279 in year 2003. The results mentioned above indicate that the inversion of foliar nitrogen concentration and the evaluation of grain protein content by NRI are surprisingly good.

6742-56, Poster Session


A based-on support vector machine (SVM) downscaling approach for evapotranspiration retrieved from MODIS, Landset TM and ASTER remote sensed image
D. Zhao, Nanjing Univ. (China) and Institute of Atmospheric Physics (China); W. Zhang, Institute of Atmospheric Physics (China) Evapotranspiration is an important role for water and energy exchange in land-atmosphere interactions, which is heterogeneous in spatial scale. Optical remote sensing imagery became a promising data source to acquire spatial information of evapotranspiration with the estimation algorithms becoming maturity over the last few decades. But scale influences on the spatial expression of evapotranspiration strongly because of the grid nature of remote sensing image. A based-on support vector machine (SVM) approach was proposed for statistic downscaling of evapotranspiration from MODIS (1000m), Landset TM (30m) and ASTER (15) satellite image at daily time scale in this paper. An integrated and easy-to-handle algorithm was developed to estimate daily evapotranspiration from multiple satellite sensors, in which the trapezoidal diagrams between surface temperature and fractional vegetation cover was to calculate the surface temperature-vegetation cover index and land surface moisture index. Daily evapotranspiration over the Heihe River Basin located in northwest, China were mapping and downscaling using the based-on SVM downscaling approach and the daily evapotranspiration estimation algorithm in this research respectively. A comprehensive scale intercomparison of remote sensed evapotranspiration retrieved from satellite image with different pixel resolutions was carried out and a higher degree of consistency between evapotranspiration retrieved from the higher-resolution satellite platforms (Landset TM and ASTER) was observed compare to one from MODIS image. By validating estimated evapotranspiration using observed data from several meteorological stations over the Heihe River Basin, northwest China, the results show that the based-on SVM approach provide a promising alternative for statistic downscaling of daily evapotranspiration over the watershed scale.

6742-58, Poster Session


Precipitation controlled the greening trend in northwest China from 1982-2003
M. Ma, Cold and Arid Regions Environmental and Engineering Research Institute (China); X. Wang, Cold and Arid Regions Environmental and Engineering Research Institute (USA); Y. Song, Cold and Arid Regions Environmental and Engineering Research Institute (China) The arid and semi-arid regions in northwest China are the first or second-degree sensitivity zones for global change. Vegetation is a mediator in climate and climate change as well. A long time series (22 years) of Advanced Very High Resolution Radiometer (AVHRR) NDVI data with 8x8 km(c) spatial resolution were used to monitor the vegetation cover in the northwest China. The monitoring results indicate a persistent greening trend exists. The precipitation is the primary drivers for inter-annual vegetation changes and is also a strong predictor of regional spatial patterns of NDVI. The NDVI is sensitive to precipitation and its sensitivity decreases with increasing precipitation. Precipitation has a higher effect on the mountainous regions vegetation cover than on the oasis regions vegetation cover. Precipitation also elicits a lag effect on the MNDVI with a phase of one year at a yearly scale mainly in the mountainous regions.

6742-60, Poster Session


Becoming a near-infrared-sensitive aerial archaeologist
G. J. Verhoeven, Univ. Gent (Belgium) For reasons of cost, resolving power and lack of digital equipment, aerial archaeologists (i.e. a specimen which takes oblique imagery from a low flying aircraft of various anomalies that could indicate an archaeological entity) use(d) film when studying their objects in the near infrared (NIR) range (some examples are - amongst others - the work of Bradford, Strandberg, Solecki, Edienne and Martin). However, even the use of colour infrared emulsions remained severely restricted till today due to some ignorance or a severe

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

49

Conference 6742: Remote Sensing for Agriculture, Ecosystems, and Hydrology


lack of knowledge about the subject and - not at least - the critical imaging process. Such a film-based workflow is after all very error-prone, as the emulsions need to be stored cooled and developed by specialised labs directly after exposing them. Moreover, determining the right exposure is not as straightforward as with conventional (i.e. visible light) photography. However, this changed completely with the advent of digital cameras. As their sensors are very sensitive to infrared radiation, the whole process of taking NIR photographs is much less of a cumber stone. The poster under consideration wants to show how NIR imagery can be taken easily with a converted digital camera, what the images look like, compare the advantages (and disadvantages) to normal archaeological aerial imaging (e.g. haze penetration, enhanced clarity of detail and visualization of stressed vegetation) as well as outline a basic approach of NIR image processing. Finally, the author will also present a specifically designed and build system, which allows to remotely acquire very detailed, large scale NIR imagery. Instead of using a kite, a blimp or ordinary balloon, this system makes use of an Allsopp helikite, a device which joins a helium balloon with a kite and is both smaller and more versatile than comparable lighter-than-air constructions. Moreover, it operates in stronger winds than traditional blimps and - by the additional lift through its wings - supports more payload for its size when compared to ordinary aerostats. Using the helikite, a very stable and easily maintainable platform is created for a sturdy cradle, the latter attached to a tether below the helikite and allowing to remotely control the NIR sensitive digital SLR.

6742-01, Session 1
Integrating earth observation data in hydrological runoff models
R. A. M. de Jeu, H. Beck, T. R. H. Holmes, H. de Moel, J. Aerts, Vrije Univ. Amsterdam (Netherlands); J. van de Vegte, Koninklijk Nederlands Meteorologisch Instituut (Netherlands); M. Owe, NASA Goddard Space Flight Ctr. (USA) The objective of this study is to improve runoff models through the integration of satellite products. We use a hydrological model which is based upon a multi-compartment methodology where the hydrological cycle of a drainage basin is described as a series of storage compartments and flows. It calculates per grid cell a water balance, which describes the transformation of input (precipitation) and output (runoff and evapotranspiration) through this cycle. Satellite observations are assimilated into this hydrological model over a historical simulation time window and applied to several different large river basins, including the river Rhine, Nile and the Mississippi. This study clearly demonstrates the impact on the performance of stream flow simulations through the exploitation of additional information about the system. The satellite data flow is facilitated by the Atmospheric Data Access to the Geospatial User Community (ADAGUC) web servers, and comprises of precipitation, temperature, land cover characteristics and soil moisture from passive microwave observations. In addition, this study will show how the ADAGUC data platform can be used by the hydrological community.

6742-61, Poster Session


The use of satellite remote sensing and GIS for assisting flood risk assessment: a case study of Agriokalamin Catchment area in Paphos-Cyprus
D. G. Hadjimitsis, Cyprus University of Technology (Cyprus) The main aim of this study is to evaluate the feasibility of utilizing satellite remote sensing and GIS tools to assist assessment of the catchment area of Agriakalamin River located in Kissonerga Village in Paphos-Cyprus. The remote sensing technique has been used to quantify the actual increase in urban area over the past twenty years. The Agriokalamin River area in Kissonerga Village in Paphos- Cyprus has been chosen as the study area. By analyzing the temporal satellite imageries and the observed meteorological and hydrologic data of the catchment, the author is able to successfully establish the concern for investigating and assessing all the catchment areas due to urbanization. The results of this study will encourage decision makers or the local authorities to consider land use planning as an effective nonstructural measure for flood risk mitigation. Remotely sensed data such as Landsat-5 TM and Quickbird image data have been used to track the urbanization near the catchment area. Several change detection techniques have been used to determine the urbanization factors. GPS has been used to locate in-situ the boundaries of the catchment area and finally GIS has been used to provide information regarding the geological and town planning characteristics of the area under investigation.

6742-02, Session 1
Integration of ground and satellite data to simulate forest carbon budget on regional scale
F. Maselli, M. Chiesi, Consiglio Nazionale delle Ricerche (Italy); M. Moriondo, Univ. degli Studi di Firenze (Italy); L. Fibbi, Consiglio Nazionale delle Ricerche (Italy); M. Bindi, Univ. degli Studi di Firenze (Italy); S. W. Running, The Univ. of Montana (USA) Simulating the main terms of forest carbon budget (GPP, NPP, NEE) is important for both scientific and practical reasons. Such simulation was achieved for a region of Central Italy (Tuscany) by the integrated processing of ground and satellite data. The operation required the preliminary collection of several data layers to characterize the eco-climatic and forest features of the region. Ancillary information layers (meteorology, forest type, volume, etc.) were derived from existing sources. Monthly FAPAR estimates with 1 km resolution were obtained by processing AVHRR, VGT and MODIS NDVI data. Relying on these data sets, monthly estimates of forest GPP were produced by means of a simplified, NDVI-based parametric model, CFix. These GPP estimates were used to calibrate a well known bio-geochemical model, BIOME-BGC, in order to find its best configurations to simulate all main functions (photosynthesis, respirations, allocations, etc.) of the most widespread forest species in Tuscany. The calibrated versions of BIOME-BGC were then applied to produce respiration estimates for all regional forest surfaces during the study period. To perform this operation, a simplified approach was applied which relies on the ratio of actual over potential tree volume as an indicator of forest distance from climax. The C-Fix photosynthesis estimates were finally integrated with BIOME-BGC simulated respirations in order to assess net forest carbon fluxes. The NPP and NEE values obtained were finally validated against ground measurements of accumulated woody biomass and net carbon exchange. The results of this validation indicated that the methodology proposed is capable of producing accurate estimates of net forest carbon exchanges at regional scale.

6742-62, Poster Session


Recognition of high risk regions of Malaria incidence using 7ETM+ data
A. Ahmadian, M. J. Valadan Zoej, M. R. Mobasheri, K.N.Toosi Univ. of Technology (Iran); Y. Rezai, K.N. Toosi Univ. of Technology (Iran) Satellite Technology is increasingly being used in monitoring environmental parameters. Some of these parameters are indicative of disease outbreak such as malaria. By boosting the level of accuracy in determination of environmental parameters using satellite imageries, we may be able to assess the potential of disease outbreak more precisely and consequently decrease the expenses of mitigation efforts. In our work we introduce the most suitable techniques for the environmental parameters using 7ETM+ images applicable in the south of Iran around cities of Kahnooj and Minab in Kerman and Hormozgan provinces respectively. In this regards some ground truth data has been collected and the work has so far shown promising progress.

6742-03, Session 1
Evaluation of the surface of objects by use of Minnaert constants
H. Okayama, C. Li, Chiba Univ. (Japan) Minnaert constants are calculated with the intensity of the light scattered from the surface of objects. Focused on remote sensing of vegetation on the earth, the light reflected from leaves is measured. A remote sensing simulator is used in an experimental room with a halogen lamp as an optical source and a Wratten gelatin filter No. 25 as a filter. Bidirectional

50

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6742: Remote Sensing for Agriculture, Ecosystems, and Hydrology


reflectance from two kinds of leaves with different degrees of roughness is measured to obtain their Minnaert constants. The Minnaert constants for leaves with a smoother surface are larger than 1 (Lambertian surface) and those with a rougher surface are smaller than 1. wavebands ranging from the visible through the thermal infrared wavelengths. Seven channels at 500 m ground resolution have been designed for land applications in the reflective domain. A previous investigation demonstrated that MODIS reflectance data can be successfully used to retrieve both EWT and FMC at leaf level. However, scaling the retrieval problem at canopy level can be problematic, since more sources of variability may hinder the signal due to water in the leaf tissues. In our research we tried to understand to which extend the structural properties of the canopy affect the retrieval of vegetation moisture parameters under the MODIS acquisition geometry. The PROSPECT radiative transfer code was used to simulate leaf reflectance and transmittance as a function of leaf properties, and the SAILH model was adopted to simulate the top of the canopy reflectance. Two sets of simulation parameters were used, for both PROSPECT and SAILH: the site specific parameters were chosen to represent the possible variations of vegetation properties in Southern Italy; the global parameters were chosen in such a manner to represent a larger variety of both vegetation properties and acquisition geometries. Results showed the effectiveness of the higher number of infrared reflective channels of the MODIS system, compared with the that of the SPOT-VEGETATION, in successfully estimating FMC, both at local and global scale. The parameter that mostly affects the estimation is leaf area index (LAI). Higher results (R2\>0.78) are achieved for LAI\>2, while acceptable results (R2\>0.60) can still be achieved for lower vegetation cover density (1<LAI<2).

6742-04, Session 2
Satellite-based fire danger estimation for winter fire occurrence
A. Lanorte, Istituto di Metodologie per lAnalisi Ambientale (Italy); I. di Donna, Univ. degli Studi della Basilicata (Italy) Winter fires represent a remarkable damage for a few European countries characterized by continental climate and complex orography. They usually strike mature conifer woods or mixed woods placed in alpine and subalpine areas far from big water surfaces and hard accessible from fireproof vehicles. The possible water freezing of fire putting out means and the limited day light time make arduous the possibilities of facing the winter fires. The studys main objective is to support the responsible organisms in taking decisions in the preventive and operating phases of management fire during the winter season providing a dynamic fire risk representation. The fire risk estimation has been, therefore, based on dynamic models able to represent the influence in the space and the time that outside variables and the vegetable physiology have on the fuel features and so on the potential fire behaviour. The study has been directed towards the obtainment of the following products: a) fuel types map applicable to alpine and sub-alpine ecosystems: the map is obtained through the application of supervised classification techniques and spectral analysis methodologies at subpixel level on multispectral images; b) fire danger map applicable to alpine and sub-alpine ecosystems: the map is obtained through the application of indices of fire risk on multispectral images with high temporal resolution.

6742-06, Session 2
ESA activities related to fire: ATSR World Fire Atlas, GlobCarbon, RISK-EOS
O. Arino, European Space Agency/ESRIN (Italy); M. Paganini, Consultant (Italy) During the past few years, the European Space Agency has launched several projects related to forest fires from global to local scales. The ATSR World Fire Atlas (WFA) project started in late 1995 and is still running today. It provides the longest time series ever produced on the global distribution of active fires. The WFA consists of 12 years of coherent and consistent data sets. The night time fire occurrences are derived from the 3.7 micron channel on-board the ATSR instrument series hosted by the ERS-2 and ENVISAT satellites. The ATSR WFA products were validated first in 1998 with the support of IGBP and more recently by extensive comparisons with existing data sets on forest fires events. A smooth transition from the ERS-2 ATSR2 to the ENVISAT AATSR has been performed in January 2003 and the quality of the WFA products continuity verified. The ATSR WFA products are available in near real time since May 2006. The distribution of the ATSR WFA products will be thoroughly analysed in this paper and a synthesis of the work performed by more than 900 registered users will be presented. The GlobCarbon project started in early 2003 with the objective to develop a service for the production of multi-year / multisensors global level 3 Land products to be used as input to carbon assimilation models. Understanding the spatial and temporal variation in carbon fluxes is essential to constrain models that predict climate change. However our current knowledge of these spatial and temporal patterns is uncertain, particularly over land. One of the bio-geophysical parameters that the GlobCarbon project aims to measure is the fully calibrated estimate of the burned areas quasi-independent of the original satellite sensor. These burned areas estimated are used in dynamic global vegetation models, a central component of the IGBP-IHDP-WCRP Global Carbon Cycle Joint Project. The service will feature global estimates of burned areas amongst other variables from 1998 to 2007, derived from Earth Observation sensors (ERS-2 ATSR-2, ENVISAT AATSR and SPOT VEGETATION). Finally the RISK-EOS project started in 2003 under the framework of the Global Monitoring for Environment and Security (GMES) initiative, with the objective to establish a network of European service providers for the provision of geo-information services in support to the risk management of meteorological hazards (floods and fires). The Fire component of RISK-EOS feature two main services: the Burn Scar Mapping (BSM) service that provides some seasonal mapping of forests and semi-natural burned areas at high spatial resolution (minimum mapping unit of 3 to 5 ha); and the Regional Fire Monitoring (RFM) service that provides near real time observation of active fires, based on middle resolution satellite

6742-05, Session 2
Retrieval of canopy moisture content for dynamic fire risk assessment using simulated MODIS bands
C. Maffei, Mediterranean Agency for Remote Sensing (Italy); A. P. Leone, Istituto per i Sistemi Agricoli e Forestali del Mediterraneo (Italy); G. Meoli, Mediterranean Agency for Remote Sensing (Italy); G. Calabr, Istituto per i Sistemi Agricoli e Forestali del Mediterraneo (Italy); M. Menenti, Istituto per i Sistemi Agricoli e Forestali del Mediterraneo (Italy) and Univ. Louis Pasteur (France) Forest fires are one of the major environmental issues in large areas of Central and Southern Italy, and more generally in Mediterranean Europe. Biomass burning reduces carbon fixation in terrestrial vegetation, while risk of soil erosion increases in burned areas. The premier action against fires is prevention, and in this context fire risk mapping is an invaluable tool. Various factors, either static or dynamic, contribute to the definition of fire risk. Among them, vegetation moisture plays a key role, since forests susceptibility to fire increases with increasing plant water stress and biomass dryness. A tool is needed to allow a timely detection of such forest conditions, and space-borne and airborne remote sensing can be very effective to this end. Many authors have demonstrated the role of remote sensing in the assessment of vegetation equivalent water thickness (EWT), which is defined as the weight of liquid water per unit of leaf surface. Various multi-spectral systems have been reported to be useful, such as Landsat TM, SPOT-VEGETATION, and NOAAAVHRR. However, forest fire danger models use the fuel moisture content (FMC) as a measure of vegetation moisture. FMC is defined as the ratio of the weight of the liquid water in a leaf over the weight of dry matter, and its retrieval from remote sensing measurements might be problematic, given its dependence on two independent biophysical properties. The aim of this research is to evaluate the potential of the Moderate Resolution Imaging Spectrometer (MODIS) in retrieving both EWT and FMC. The MODIS system is carried on board of EOS Terra and Aqua satellites, and has 36 spectral

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

51

Conference 6742: Remote Sensing for Agriculture, Ecosystems, and Hydrology


data (AQUA/TERRA MODIS and MSG SEVIRI). The RISK-EOS BSM service builds on the achievements of ITALSCAR, a demonstration project for the yearly mapping of burned areas in Italy, using the LANDSAT Thematic Mappers. The paper will provide a synthesis of the RISK-EOS products validation and utility reports collected during the 2006 summer season.

6742-10, Session 3
A novel approach for satellite retrieval of surface evapotranspiration
Y. Liu, Nagoya Univ. (Japan) Reliable retrieval of land surface evapotranspiration from satellite data still remains as a challenging task to date. The retrieval approach that based on the contemporary theory of evaporation involves the complicated parameterization, and this makes it hard to use for long-term change and global-scale studies. The triangular approach takes advantages of the spatial texture relationship between surface temperature and vegetation index, but determination of the triangular function is often subjective. Based on energy balance principle, equilibrium evaporation, and non-equilibrium thermodynamics, we proposed a novel approach to estimating surface evapotranspiration. The necessary inputs are net radiation, soil heat flux, surface emissivity, land surface temperature, and surface air temperature. To verify the approach, we used available field data obtained on a flat, semiarid tableland in the mid-south of the Loess Plateau, China (35.2N, 107.7E). A flux and radiation observation system (FROS) was used to accurately measure radiation components and turbulent fluxes in the atmospheric surface layer. Using the field data at a height of 2-m, we used the proposed approach to estimate latent heat and sensible heat fluxes. The produced results agreed surprisingly well with that of field observation. Furthermore, uncertainty analysis revealed that the overall uncertainty in the estimates would be within 30W/m2, given an accuracy of land surface temperature within 2 degrees. In general, the simplicity and clarity of the proposed approach make it highly valuable to remote sensing.

6742-07, Session 2
Detection of small fire-affected areas by satellite data in Tuscany region
C. Conese, Institute of Biometerology (Italy); R. Carla, National Research Council of Italy (Italy); L. Bonara, Istituto di Biometeorologia (Italy) In Italy, a large percentage of the total forested area burned every year is affected by few fires of large size but, at the same time, a great number of small fires affect areas of few hectares each. Thus, an accurate operational mapping of fire affected areas by satellite needs the employement of methods able to recognize the presence of both the large burnt areas and the small or very small ones. Indeed, up to now very little attention has been paid to the recognition and analysis, over large territories, of areas of very few hectares affected by small fires. Nevertheless, the possibility to derive by satellite data a map of burnt areas at regional or national scale which includes also small burnt areas is of great importance for the choice of suitable environmental management policies. When the task of locating burnt areas must be carried out periodically on large areas of varying characteristics as the Tuscany Region (Italy), the crucial problem is that of automatically locating the presence of fire scar on the territory, reducing as much as possible the number of false alarms, that is, of pixels that are erroneously flagged as burnt area. In this work, pre- and post-fire Landsat-ETM images have been used to characterize the ability of some spectral indices used as binary classifiers to detect small fire affected areas, in particular to discriminate between pixels corresponding to burnt and non-burnt areas in the Tuscany Region. Their efficiency was evaluated with regard to commission and omission errors as a function of the threshold value.

6742-11, Session 3
Potential errors in the application of thermalbased energy balance models with coarse resolution data
W. P. Kustas, N. Agam, M. Anderson, F. Li, U.S. Dept. of Agriculture (USA) A thermal infrared (TIR)-based two-source (soil + vegetation) energy balance (TSEB) model validated with remotely sensed imagery over a wide variety of landscapes is applied to an agricultural region with significant variability in vegetation cover and soil moisture conditions. The TSEB model uses sub-field TIR pixel data (60 m) from Landsat ETM aggregated to 120 m, 240 m and ~1 km resolution to assess potential model errors and inability to discriminate fluxes from individual fields with the coarser resolution data. The motivation for this study stems from the fact that there are no definitive plans for supporting any future satellite-based high resolution (~100 m) TIR sensor, specifically on the next proposed Landsat platform. Therefore, in the near future only km-scale TIR data will be available from satellite sensors (e.g., MODIS). A methodology for sharpening TIR data from coarser resolutions as a means of obtaining higher resolution thermal imagery will also be described and compared to the native resolution data. The resulting errors in fluxes between native versus sharpened TIR used by the TSEB model will be discussed as well as differences in aggregated fluxes applying the TSEB model with coarse versus high resolution TIR data. The implications of having no Earth observing TIR satellite system with resolutions similar to Landsat for modeling surface fluxes, water use and crop/vegetation stress for distinct land cover types, as well as errors in regional-scale estimates and on validation using flux tower observations will be discussed.

6742-08, Session 2
Estimating fire-induced variability in vegetation covers by using SPOT-VGT time series
R. Lasaponara, A. Lanorte, L. Telesca, Istituto di Metodologie per lAnalisi Ambientale (Italy) Fires are considered one of the most important causes of land degradation since they induce significant alterations in short as well as long-term vegetation dynamics. The estimations of fire induced variability can be effectively approached by using satellite data. In this study, SPOT VEGETATION temporal series (1998 to 2005) of NDVI [obtained using the NDVI= (NIR- Red)/(NIR+ Red)], and NDWI [obtained using the NDWI= (SWIR- Red)/(SWIR+ Red)] were analyzed to perform a dynamical characterization of burned and unburned vegetation covers. To this aim, VGT-NDVI data acquired for both fire-affected and fire-unaffected vegetational covers were analysed using the Detrended Fluctuation Analysis (DFA) proposed by Peng et al. (1995). The DFA is a well-known methodology, which allows the detectin of long-range power-law correlations in signals possibly characterized by nonstationarity, which features most of the observational and experimental signals. Results from our analysis point out that the persistence of vegetation dynamics is significantly increased by the occurrence of fires. In particular, a scaling behavior of two classes of vegetation (burned and unburned) has been revealed. The estimated scaling exponents of both classes suggest a persistent character of the vegetation dynamics. But, the burned sites show much larger exponents than those calculated for the unburned sites. This result points out to the role played by fires in driving a more unstable vegetation patterns for burned areas, which indicates an efficient fire-induced vegetation recovery processes The methodology approached in the present study could be fruitfully applied to investigate other types of vegetation stresses.

6742-13, Session 3
A hybrid approach for estimating spatial evapotranspiration from satellite imagery
C. M. U. Neale, Utah State Univ. (USA); M. P. Gonzlez Dugo, Instituto de Investigacin y Formacin Agraria y Pesquera (Spain); L. Mateos, Instituto de Agricultura Sostenible (Spain); W. P. Kustas, U.S. Dept. of Agriculture (USA) Two common approaches for estimating crop evapotranspiration (ET) using satellite imagery are the reflectance-based crop coefficient method and the energy balance method. The reflectance-based crop coefficient method relates a reflectance-based vegetation index such as

52

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6742: Remote Sensing for Agriculture, Ecosystems, and Hydrology


the soil adjusted vegetation index (SAVI) to ET basal crop coefficients such as those described by Wright (1982) and the FAO 56 manual (Allen et al, 1998). A time-series of remotely sensed inputs is then used to build the crop coefficient curve in each field being monitored. In order to obtain actual ET, a water balance must be maintained in the root zone of the crop in order to make the appropriate adjustments due to soil moisture deficits and wet soil surface from irrigation and/or rain. Ground meteorological data must be provided by a weather station located in the modeled area for the estimation of reference ET. In the energy balance approach, surface temperatures are used in the estimation of sensible heat fluxes and depending on the complexity of the model, different methods are used to either handle the aerodynamic temperature term or deal with sparse canopies (empirical approaches, twosource model, SEBAL model). Remotely sensed inputs are also used for the estimation of net radiation and soil heat flux, with latent heat flux (ET) obtained as a residual from the energy balance equation. The energy balance approach results in the actual ET being estimated directly. Instantaneous values of ET must be extrapolated to the entire day and over time in between satellite overpass inputs. This paper describes a hybrid approach that uses both methods in combination to monitor actual ET over a growing season for irrigated and non-irrigated crops. The model has been coded in an ArcGIS environment, using visual basic for the calculations. Testing of model results against measured energy balance fluxes using eddy covariance energy balance systems will be presented.

6742-17, Session 4
Parameterization and sensitivity analyses of an analytical irradiance radiative transfer model for plant canopies: implications for ecosystem monitoring needs
C. R. Hall, Dynamac Corp. (USA); C. R. Bostater, Jr., Florida Institute of Technology (USA) An objective of remote sensing is determination of biochemical and biophysical characteristics of plant canopies utilizing high spectral resolution sensors. Leaf and canopy properties shape reflectance signatures through wavelength dependent absorption, scattering, conversion, and attenuation processes. The canopy reflectance is influenced by projected foliage cover (PFC), canopy depth (h), leaf area index (LAI), leaf thickness (Lh), zenith leaf angle distribution (LAD), leaf moisture and pigments. An analytical radiative transfer model was parameterized for use with plant canopies using a hierarchical approach using three models with increasing numbers of variables. A canopy scale model parameterization incorporated diffuse canopy absorption, a(), backscatter b(), solar beam to diffuse conversion c(), and attenuation () coefficients. A leaf scale model parameterization incorporated leaf absorption a (), backscatter b(), beam to diffuse conversion c(), and attenuation () coefficients. A leaf volume correction index (LVCI) was developed and demonstrated to translate the leaf scale data to the canopy level. In a third model parameterization, the LVCI was decomposed into LAI, leaf thickness adjusted of zenith angle distribution, and canopy depth h (m). Leaf moisture and pigment influences on above canopy reflectance Ra () were incorporated through the diffuse leaf absorption and backscatter coefficients using estimates of chemical specific absorption a*()(m2g-1) and backscatter b*()(m2g-1). Sensitivity analyses indicated the conversion coefficient was most important in the canopy derived parameterization. In the leaf derived parameterization, LVCI, leaf absorption and backscatter and canopy depth were most important. LAI and bottom reflectance did not rank high on the normalized sensitivity scale results obtained from model simulations of canopy reflectance. The above results are placed within the context for ecosystem monitoring needs.

6742-15, Session 4
Soil moisture content of composite rough surfaces based on a variational technique to distinguish between the larger and smaller scale surface spectral density functions
E. Bahar, Univ. of Nebraska/Lincoln (USA) The determination of soil moisture content is often based on the measurements of the ratio of the vertically and horizontally polarized cross sections for large angles of incidence, where the cross sections could be significantly different. Using the high frequency, physical optics model of the earths surface, this ratio depends primarily on the Fresnel reflection coefficients for the two polarizations while the impact of surface roughness factors out of the cross section ratio. Thus for highly conducting moisture saturated soils, this ratio approaches one. Using the low frequency, small height/slope perturbation model of the earths surface, the vertically and horizontally polarized cross sections are critically dependent on polarization for large angles of incidence, even for perfectly conducting rough surfaces. However using the standard perturbation model, the ratio of the cross sections are also independent of the surface roughness. Applying the small perturbation approach to highly conducting rough surfaces, the ratio of the horizontally to vertically polarized cross sections approaches zero for grazing angles of incidence. There is ample experimental evidence that neither the physical optics nor the small perturbation models are adequate. The standard hybrid two scale physical optics-perturbation approach depends critically upon the decomposition of the composite surface into smaller and larger scale surfaces and the smaller scale surface is restricted to small Releigh roughness parameters, proportional to the mean square height. Using a two scale full wave approach, the cross section are expressed as a weighted sum of a physical optics cross section for the larger scale surface and a cross section for the smaller scale surface that is modulated by the slopes of the larger scale surface. A variational technique is used to decompose the surface height spectral density function into surface height spectral density functions for the larger and smaller scale surfaces. It is shown that the corresponding polarization dependent rough surface cross sections are stationary over a wide range of the variational parameters. The ratio of the cross sections are to a degree dependent on the surface roughness.

6742-19, Session 4
Analysis of the attenuation in soils and water content in remote sensing in-situ surveying
V. Prez-Gracia, L. G. Pujades, R. Gonzlez-Drigo, D. Di Capua, Univ. Politecnica de Catalua (Spain) Ground-penetrating radar (GPR) is a high resolution remote sensing in-situ surveying method applied to civil engineering, surface geology, archaeology and other disciplines. Mainly it is used solving the direct problem and obtaining a model of the studied medium. Otherwise, the study of the inverse problem could provide other valuable information: the electromagnetic parameters which characterize the medium. These parameters are obtained from the changes of the velocity, attenuation and frequency of the recorded wave. The physical properties related to those wave parameters are, mainly, the water content and the porosity. Several lab experiences were carried out in order to obtain these parameters from different soil samples. In these soil samples, porosity and water content were measured and controlled, together with the mineral composition and the conductivity. Composition was obtained with a DRX analysis and conductivity was determined with a Wenner dispositive. Velocity was obtained by measuring the two-way travel time of the reflected wave and comparing wave reflected amplitudes on the surface of the samples. Attenuation coefficients were determined from the analysis of the amplitude of the wave travelling in different thickness samples. Frequencies, velocities and wave attenuation were analysed in the different cases in order to characterize those different media and to relate its water content and its porosity with these measured parameters. The experimental results were also compared with a theoretical CRIM model of soil. Interesting results were obtained correlating the conductivity with the attenuation, the velocity and the water content.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

53

Conference 6742: Remote Sensing for Agriculture, Ecosystems, and Hydrology

6742-20, Session 4
Vegetation canopy structural and physical variability based on radiometric and laser analysis
J. R. Dim, K. Kajiwara, Y. Honda, Chiba Univ. (Japan) A comprehensive vegetation monitoring and/or management involve detailed understanding of the distribution of energy among plant components of this vegetation. And, the main energy source here is solar radiation. Combining radiometric and laser scanner data, does not only provide reflection properties associated with the distribution of the energy received by the plants, but also, structural and geometric information of the vegetation canopy. In this study, airborne multispectral radiation data and, laser pulse returns are used to derive canopy structural and physical variables. Laser scanner technology provides an efficient tool for the generation of physical models, so often used for the interpreation of radiation characteristics. Advanced processing laser techniques give information about the vertical structure of canopies (shape, height, density and roughness of trees) and their horizontal extension (e.g. crown width). The aim of this study is to examine the impact of the vegetation 2-dimensional canopy structure on the distribution of solar radiation among plants of a forest area. The results obtained show that laser retrieved canopy height and other structural data can be used for the selection of the most appropriate configuration of radiation measurements in future airborne missions and optimization of canopy characteristics.

resources. Crop yield simulation models have shown great success in predicting crop yields at field and small scales; however, there is still a gap in using field-scale models for predicting yields at regional scales. Imagery from the MODIS sensor onboard the Terra satellite offers an excellent opportunity for daily coverage at 250 m resolution. Integrating MODIS derived parameters with crop yield simulation models were successful for small-scale. However, this method may not be practical for operational applications and at large scales. Timely and accurate assessment of crop yields is important for USDAs operational program. This study focused on the major corn and soybean production states of Iowa and Illinois in the U.S. Corn Belt. The study objectives were; a) develop a decision tree classification algorithm to separate the crops during the crop season and b) evaluate a crop yield algorithm based on NDVI and surface temperature data from MODIS Terra. Classification accuracy for corn and soybean crop were found to be within 80-85 % of the Landsat-based classification. The crop yields for state and county levels were assessed for four years (2003-06). The state yield estimates were closely correlated with the USDAs assessment. County level estimates were not as well correlated, however were within 20% of the USDA estimates. The remote sensing method is a bottom-up approach and the USDAs estimate for county level is a topdown approach and may attribute to the differences in yield estimates.

6742-23, Session 5
Evaluation of different physical based methods for retrieving crop characteristics by using high resolution airborne data
K. Richter, Univ. fr Bodenkultur Wien (Austria); F. Vuolo, Ariespace s.r.l. (Italy); G. DUrso, Univ. degli Studi di Napoli Federico II (Italy) In the context of sustainable agricultural production, Earth Observation (E.O.) technologies provide a valuable data base for the monitoring of crop and soil characteristics on a large scale, in a rapid, accurate and cost-effective way. The study aimed at evaluating promising methods for the estimation of canopy parameters to provide future algorithms for the upcoming E.O. satellite Sentinel developed by ESA in the framework of GMES (Global Monitoring for Environment and Security). The work was based on hyperspectral data acquired by the optical airborne instrument CASI in the framework of ESA AgriSAR 2006 campaign in the agricultural area Demmin, in Mecklenburg-Western Pomerania, Germany. The performance of airborne imagery for the determination of within-field variability of canopy parameters, such as leaf area index, leaf chlorophyll content and soil water levels, was exploited by comparing different physical algorithms for parameter estimation. The combined radiative transfer model PROSAILH, including PROSPECT (at leaf level) and SAILH (at canopy level) was inverted by using three different algorithms: a traditional iterative optimization technique, a simple and fast look-up-table (LUT) approach and a global search routine similar to genetic algorithms. The quality of the canopy parameter estimation was evaluated using ground measurements, acquired within the campaign on wheat, corn and sugar beet crops. The implementation of the approaches into an operative routine using E.O. data for agricultural land monitoring is discussed.

6742-21, Session 4
Investigating structural-spectral interactions in managed, even-aged eucalyptus plantations using lidar and multi-spectral high resolution data
J. A. van Aardt, W. Roberts, Council for Scientific and Industrial Research (South Africa) The main goal of this research was to investigate the structuralspectral interactions that exist in managed, homogeneous, even-aged Eucalyptus plantations in Kwazulu-Natal, South Africa. Eucalypts plantations used in this study range between four and ten years old. Small-footprint light detection and ranging (lidar; ALTM 3033 two-return laser system; 0.2 mrad footprint, 33 kHz pulse rate) and ASTER and Ikonos multispectral data were used in this study. Lidar and multi-spectral data were collected during the spring season of 2006. Segmentation (multi-resolution/hierarchical algorithm) of a lidarderived canopy height model was performed in order to derive areas with homogeneous structural properties. Structural characterisation of segments were performed by comparing first and second return distributions, which mimic the integrated response found in waveform lidar sensors. The reflectanceprocessed multi-spectral data were used to (i) characterise the structural differences among different sets of homogeneous segments and (ii) determine the contribution of structural variability to spectral responses. The degree with which indicators derived from multi-spectral data, e.g., indices, could be used to describe the structural nature of and differences among various segment classes, was important in the first instance. This addresses the question of complimentarity between spectral and structural data sets in such forest environments. The second question pertains to the spectral variability found in structurally homogenous even-aged, monoculture forest stands. Specific implications relate to the use and interpretation of widely accepted forest state indicators, and how structure affects indicator outcomes. Results will be presented at the conference.

6742-24, Session 5
Crop yield monitoring based on a photosynthetic sterility model using NDVI and daily meteorological data
D. Kaneko, Matsue National College of Technology (Japan) This research is intended to develop a model to monitor rice yields using the photosynthetic yield index, which integrates solar radiation and air temperature effects on photosynthesis and grain-filling from heading to ripening. Monitoring crop production using remotely sensed and daily meteorological data can provide an important early warning of poor crop production to Asian countries, with their still-growing populations, and also to Japan, which produces insufficient grain for its population. Grain production monitoring would support orderly crisis

6742-22, Session 5
Potential use of MODIS imagery for operational crop yield assessment
P. C. Doraiswamy, U.S. Dept. of Agriculture (USA); B. Akhmedov, Science Systems and Applications, Inc. (USA); A. Stern, U.S. Dept. of Agriculture (USA) Monitoring crop condition and yields at regional scales remains a challenge. Ground-based sampling for assessment of crop yields at regional and national scales require enormous

54

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6742: Remote Sensing for Agriculture, Ecosystems, and Hydrology


management to maintain food security in Asia. The author improved a photosynthesis-and-sterility based crop production CPI index to crop yield index CYI, which estimates rice yields, in place of the crop situation index CSI, which gives a percentage of rice yields compared to normal annual production. The model calculates photosynthesis rates including biomass effects, low-temperature sterility, and hightemperature injury by incorporating: solar radiation, effective air temperature, normalized difference vegetation index NDVI, and the effect of temperature on photosynthesis by grain plant leaves. The method is based on routine observation data, enabling automated monitoring of crop production at arbitrary regions without special observations. The method oversees the quantity of grain in production at an early stage to raise the alarm in Asian countries, which are facing climate fluctuation that can be expected to continue through this century of global warming. of data pertaining to the size of trees (in terms of canopy surface area) obtained by means of remote sensing methods. Geostatistical methods are suitable because the locations of the trees are at the nodes of an unstructured grid. More specifically, we present a variogram analysis to detect correlations in the tree size spatial distribution, and we apply a novel method of anisotropy analysis to search for possible anisotropy in the size distribution. We use a combination of aerial photographs and satellite images in four snapshots covering 37 years to investigate the temporal behavior in addition to the spatial distribution at a single time. The aerial photographs were taken in 1964, 1984, 1993, and the IKONOS satellite image in 2001. We follow five plots covering over 1,000 ha and over 13,000 tree individuals. Our plots are located in the Southern Kalahari savanna near the city of Kimberley, South Africa.

6742-25, Session 5
Spectral radiance characteristics and vegetation indices of crops in relation to growth stages and leaf area index
H. N. Das, India Meteorological Dept. (India) ABSTRACT The spectral reflectance of crop and other vegetation mainly depends upon (i) leaf optical properties (ii) canopy geometry (iii) back ground soil (iv) solar illumination and zenith angle, and (v) atmospheric transmittance. Optical reflectance of crops is determined by the interaction of solar radiation with crop canopy. The amount and characteristics of reflectance from leaf surfaces are affected by structure and orientation of plant leaves. In the vegetative growth stage the leaves are primarily scattering elements. The low reflectance (and transmittance) in the visible region is due to absorption by leaf pigments. The higher reflectance as well as transmittance in near infrared plateau is explained by multiple reflections in the internal mesophyll structure. The interaction processes between solar radiation and crop canopy have been extensively studied from late sixties and it has been shown that canopy reflectance is highly related to crop variables such as Leaf Area Index (LAI) as a fraction of ground cover. LAI is a very useful estimate for various purposes, viz. water stress monitoring, crop condition monitoring and crop yield modeling. Several scientists studied the relation between LAI and different vegetation indices like Normalised Difference Vegetation Index (NDVI), Ratio Vegetation Index (RVI), Weighted Difference Vegetation Index (WDVI), Perpendicular Vegetation Index (PVI) etc. Keeping these in view, a number of experiments have been conducted by Agricultural Meteorology Division, India Meteorological Department, Pune using hand-held multiband Ground Truth Radiometer (GTR) for studying reflectance characteristics of different crops like groundnut, sunflower, wheat, maize etc. under contrasting soil moisture conditions and for establishing relationship between vegetation indices and LAI. Crop acreage estimation for sorghum has also been carried out for same districts of Maharashtra during the recent past. Experimental results revealed that there exists a good relationship between reflectance in red and near infrared bands and crop growth stages. Spectral reflectance characteristics have been found suitable for studying crop growth status. LAI could also be estimated with reasonable accuracy for irrigated and water-stressed crops using relationship between LAI and different vegetation indices. These results may help to explore the possibility of using satellite data for identification of state and stage of crops, estimation of leaf area index and crop acreage estimation. It has also been found that RVI is a better predictor of LAI during the early stage of growth while NDVI is a better predictor during the latter part of groth as compared to other vegetation indices.

6742-27, Session 5
Spatial vegetation variation patterns in southern Italy as detected by AVHRR and MODIS observations
G. Quarta, D. Conte, G. Marra, F. F. Parmiggiani, Istituto di Scienze dellAtmosfera e del Clima (Italy) Spatial variations of vegetation cover in Southern Italy were analyzed for the years 1995-2005. The analysis was carried out using NDVI (Normalized Difference Vegetation Index) time series from NOAA AVHRR images retrieved from DLR/EOWEB archives [1] and corrected using an ad-hoc method which exploits more accurate MODIS NDVI maps available from NASA archives [2]. DLR NDVI map time series, which start in 1995, present significant artificial anomalies as no atmospheric and anisotropic reflection corrections are applied to the data [3]. On the contrary, MODIS NDVI are radiometrically, atmospherically and geometrically corrected but are only available since 2002. Aim of this study is to emphasize changes occurred in biomass production in the last decade by analyzing vegetation behaviour during the growing seasons. This approach, which removes winter months from the analysis, improves the quality of the results as during winter vegetation exhibits a stationary behaviour and NDVI data are less significant because of a larger presence of clouds and snow. The correlation between vegetation change pattern and terrain elevation derived from SRTM topographic data [4] was also investigated, making use of a 2-D correlation analysis. The results show the presence of different vegetation change patterns such as: i) a degradation in several flat areas with variations up to -1%; ii) a systematic increase over the Apennines up to 6%. These findings, which match with temperature trends in the winter months in the last ten years [5], highlights climate change processes going on in the Mediterranean: temperature mitigation facilitates robust vegetation as is found in uplands and mountains (conifer stands, deciduous stands and, in general, Mediterranean maquis), while, especially in flat areas, plant foliation in the autumn-winter period is strongly affected by sudden drops in temperature and by human activities. REFERENCES [1] Earth Observation on the Web, German Aerospace Center (DLR). German. URL: http://taurus.caf.dlr.de:8080/index.html [2] Earth Observing System Data Gateway, NASA, USA. URL: http://delenn.gsfc.nasa.gov/ imswww/pub/imswelcome/ [3] Dech, S.W.; Meisner, R.E.; Tungalagsaikhan, P.; Gesell, G.; Popp, T.; Thomas, W., 1997, Operational generation of AVHRRbased level-3 products at the German Remote Sensing Data Centre: status and perspectives. Proc. IGARSS 97, Vol. 4, 3-8 Aug. 1997, 1505-1508. [4] Farr, T.G., Kobrick, M., 2000, Shuttle Radar Topography Mission produces a wealth of data, Amer. Geophys. Union Eos, v. 81, p. 583-585. [5] Brunetti, M., Maugeri, M., Monti, F., and Nanni, T., Temperature and Precipitation Variability in Italy in the last two centuries from homogenised instrumental time series. Int. J. Climatol. 26: 345-381 (2006).

6742-26, Session 5
Geostatistical analysis of tree-size distributions in the Southern Kalahari, obtained from remotely-sensed data
A. Moustakas, A. Chorti, D. T. Hristopulos, Technical Univ. of Crete (Greece) We propose using geostatistical methods for the spatial analysis

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

55

Conference 6742: Remote Sensing for Agriculture, Ecosystems, and Hydrology

6742-28, Session 5
Effects of spectral resolution and signal/noise ratio in the retrieval of solar-induced chlorophyll fluorescence and reflectance by passive remote sensing in the oxygen absorption bands
L. Palombi, G. Agati, G. Cecchi, D. Lognoli, V. Raimondi, G. Toci, Istituto di Fisica Applicata Nello Carrara (Italy) Retrieval of chlorophyll (Chl) fluorescence signals using passive detection of radiance coming from vegetation is rising large interest because of its potential application in satellite mapping of plant photosynthetic activity. We used an ARC SpectraPro 2300i spectrometer (300 mm of focal length) coupled to a PIMAX (Princeton Instrument) intensified, gateable CCD (Thomson 7895, 512x512 pixels, 24x24 mm2 pitch) to measure solarinduced radiance from leaves of Ficus benjamina L. attached to the plant. At its maximum resolution, the instrument had a spectral coverage of 11.5 nm \@ 500 nm, with a nominal spectral resolution of 0.022 nm/channel. A proper fiber optic bundle, positioned at the input slit of the spectrometer, collected the radiance from the leaf at a 5 cm distance. Measurements were performed on both the 687 and 760 nm atmospheric oxygen absorption bands. To take into account variations in the depth of the O2 bands, each leaf radiance detection was preceded and followed by measurements over a white Spectral (Labsphere) reflectance reference positioned next to the leaf. A computer controlled stage permitted the rapid exchange of sample and reference in front of the collecting fiber optic bundle. To check the capability of the method, Chl fluorescence was retrieved from leaf radiance, at the 687 nm O2 band, during a photoinhibition stress treatment induced by high-light (2500 mE m-2 s-1) irradiation. As expected the Chl fluorescence contribution was largely reduced under photoinhibition and its value normalized to total leaf radiance changed from 32% to 16% after 1 hour of treatment.

6742-30, Session 5
Crop type classification by Hyperion data and MTMF algorithm
H. Fahimnejad, S. R. Soofbaf, A. Alimohammadi, M. J. Valadan Zouj, K.N. Toosi Univ. of Technology (Iran) Discrimination of crop types has been considered as one of important challenges in agricultural applications of remote sensing. Recently available hyperspectral data provide valuable opportunities for mapping and monitoring of crop types. Therefore, practical evaluation of the utility of these data is of prime importance. In this research, capabilities of the hyperspectral data of Hyperion sensor acquired from an agricultural area in Ahmadabad, south of Tehran for cover type mapping have been investigated. These data included 242 Spectral bands between 400 and 2500 nm (level 1B1) with spatial resolution of 30 meters. After application of the standard atmospheric and geometric correction procedures, the corrected data have been analyzed by application of the MTMF algorithm. Results of this research have shown the significant capabilities of the hyperspectral data for crop type and agricultural land use mapping. Some spectrally similar crops in multi spectral data can be discriminated by the use of hyperspectral data. However, lack of a reliable data of spectral properties of different crop types is a serious problem. In addition, qualities of results show high dependence on the size of agricultural fields and complexity of the patterns of crop cultivation in the area.

6742-31, Session 6
Integrating temporal and spectral information from low-resolution MODIS and high-resolution optical satellite images: two Hungarian case studies
D. Kristof, D. Neidert, Z. Nagy, K. Pinter, Szent Istvn Univ. (Hungary) The aim of this study is to investigate the possibilities of integrating both temporal and spectral information provided by MODIS Land Surface Reflectance (LSR) products together with high resolution satellite images. MODIS LSR products provide an estimate of the surface spectral reflectance, and offer daily revisit capabilities. Nevertheless, the low spatial resolution of MODIS results in a number of issues to be dealt with (such as gridding artifacts and non-rectangular point spread function, among others). An increasing number of methods can be found in the literature to deal with these hindrances and are used and tested during our work. This work consists of two case studies. In the first case study, we use very high resolution IKONOS images along with daily MODIS time series to monitor agricultural practice over a case study area in Hungary. The applied methods include image segmentation, classification, linear unmixing and trend analysis. Our preliminary results show that low-resolution data can be efficiently used for this purpose. In the other case study, we have compared a five-month series of daily MODIS reflectance data to the in situ measurements taken by the radiometers of a flux tower mounted on a grassland study site in the eastern part of Hungary. Moreover, five high-resolution SPOT-5 satellite images, taken at different dates during the vegetation period, were used to analyze fine spatial patterns of the vegetation. To make the SPOT images intercomparable and comparable with MODIS and flux tower measurements, their radiometric correction was carried out using linear radiometric rectification techniques with the corresponding spectral bands of the sameday MODIS reflectance products. Our results show that despite some difficulties and limitations, this simple one-step fitting can yield accurate and consistent radiometric correction in many cases.

6742-29, Session 5
Ecosystem productivity and dynamics issued from multispectral and hyperspectral satellite imagery
A. Kyparissis, N. Markos, S. Stagakis, E. Levizou, Univ. of Ioannina (Greece); O. Sykioti, National Observatory of Athens (Greece) Low resolution images from MODIS and SPOT multispectral satellites are used for extracting indices correlated with major parameters of productivity, for two deciduous forests (Fagus sylvatica, Quercus sp.) and one shrubland dominated by the semi-deciduous Phlomis fruticosa. Ground ecophysiological measurements were conducted for three growing periods (20052007) and are used for indices evaluation as well as input parameters for an ecosystem productivity model. The results of the ground-based productivity model are compared to the 8day MODIS GPP product, showing that MODIS algorithm underestimates productivity and does not closely follow ecosystem dynamics. In an attempt for a more precise productivity product a new light-use efficiency model based on satellite and meteorological data is designed and presented. Moreover, hyperspectral images from CHRIS/PROBA are used for a more detailed study of the semi-decidual Phlomis fruticosa ecosystem. Ground ecophysiological measurements from two growing periods (2006-2007) are used for evaluation purposes. Images are geometrically corrected and atmospherically adjusted. The reflectance spectra obtained are used for extracting indices related to numerous plant physiological parameters. Fast responsive plant processes, such as the function of the photosynthetic apparatus, the photoprotective response to stress factors (low or high temperature, lack of precipitation) and the detailed pigment content of leaves (chlorophyll a, chlorophyll b, carotenoids) may well be followed by such indices issued from hyperspetral data, offering great advantage over multispectral images for ecosystem remote sensing.

56

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6742: Remote Sensing for Agriculture, Ecosystems, and Hydrology

6742-32, Session 6
EO-1 Hyperion and ALI bands simulation to Landat 7 ETM+ bands and comparison
K. G. Nikolakopoulos, Institute of Geology & Mineral Exploration (Greece); G. A. Skianis, D. A. Vaiopoulos, Univ. of Athens (Greece) Numerous satellite sensor systems have been launched during the last twenty years and satellite data are increasingly being used in regional or global vegetation monitoring. The observation of global vegetation from multiple satellites requires much effort to ensure continuity and compatibility due to differences in sensor characteristics and product generation algorithms. More recently the launch of hyperspectral sensor like Hyperion make the compatibility problem even more difficult as the very narrow hyperspectral bands need to be simulated to the broader multispectral bands before proceed to any further comparison. In this study we tried to compare multispectral (Landsat ETM+ and EO-1 Advanced Land Imager) data with hyperspectral (Hyperion) data for the vegetation cover mapping of Milos Island. All the data were collected the same day within one-minute time. As a result the atmospheric conditions were exactly the same and that make the data ideal for comparison. The performance of the EO-1 Hyperion imaging spectrometer with the Advanced Land Imager (ALI) and the Landsat 7 ETM+ sensor was compared using a method that aggregated portions of the Hyperion 10 nm bands to simulate the broader multispectral bands of ALI and ETM+. The general process was to calculate a weighted sum of the Hyperion bands that covered each Landsat band. The weights used in the sum were derived, by comparing the spectral response of the hyperspectral bands with the respective multispectral band. The Normalized Difference Vegetation Index was used for the comparison of the three data sets and the results are presented in this study.

each mesohabitat unit type for each modeled flow. Fish habitat was then estimated and quantified according to the preferences of main target species as brown trout (Salmo trutta). LIDAR data combined with hydraulic modeling allowed the analysis of fluvial habitat in long fluvial segments which would be unfeasible with traditional survey. On the other hand, the use of LIDAR data requires a detailed bathymetry and substrate characterization of the submerged areas. LIDAR habitat assessment at mesoscale level avoids the problems of time efficiency and upscaling (Maddock, 1999) and is a recommended approach for large river basin management.

6742-36, Session 6
Use of airborne hyperspectral imagery to investigate the influence of soil nitrogen supplies and variable fertilization to winter wheat growth
X. Song, Beijing Academy of Agriculture and Forestry Sciences (China) and National Engineering Research Ctr. for Information Technology (China); G. Yan, Beijing Normal Univ. (China); J. Wang, L. Liu, National Engineering Research Ctr. for Information Technology (China); X. Xue, C. Li, W. Huang, Beijing Academy of Agriculture and Forestry Sciences (China) Advanced technology in airborne detection of crop growth by remote sending can help optimize the strategies of fertilization, and help maximize the grain output by adjusting field inputs. In this study, Push-broom Hyperspectral Image sensor (PHI) was used to investigate the influence of soil nitrogen supplied and variable-rate fertilization to the growth of winter wheat. The objective was to determine to what extent the reflectance obtained in the 112 visible and near-infrared (NIR) wavebands (from 410nm to 832nm) might be related to differences of variance of soil nitrogen and variable-rate fertilization. Management units were arranged at Beijing Precision Farming Experimental Station. Three flights were made during the wheat growing season. Several field experiments, including the crop sampling, soil sampling and variable-rate fertilization were carried out in the field. Data were analyzed for each flight and each band separately. Some spectrum indices were derived from PHI images and statistical correlation analysis were carried out between the spectrum indices and soil nitrogen, variablerate fertilization amount. In addition, the spectrum indices difference between elongation stage and grain filling stage are calculated and the correlation analysis was also carried out. The analysis results indicated that the reflectance of winter wheat is significantly influenced at certain wavelength by the soil nitrogen and the variable-rate fertilization. The soil nitrogen effect was detectable in all the three flights. Differences in response due to soil nitrogen variance were most evident at spectrum indices, such as , INFLEX, Green/Red, NIR-ness, DVI and RDVI. Furthermore, analysis results also indicated that the variable fertilization can reduce the growth difference of winter wheat caused by spatial distribution difference of soil nitrogen.

6742-34, Session 6
Fish habitat characterization and quantification using lidar and conventional topographic information in river survey
M. Marchamalo Sacristn, Univ. Politcnica de Madrid (Spain) and Hydrobiology Research Group, Univ. Politcnica de Madrid (Spain); M. Bejarano, D. Garca de Jaln, R. Martnez Marn, Univ. Politcnica de Madrid (Spain) This study presents the application of LIDAR data to the evaluation and quantification of fluvial habitat in river systems, coupling remote sensing techniques with hydrological modeling and last findings on ecohydraulics. LIDAR techniques offer a powerful tool applicable to a wide range of hydrological applications, such us flood modeling (Cobby et al., 2001; Dal Cin et al., 2005; Roberts et al., 2007) , water balance (Wigneron et al., 1998; Schmugge et al., 2002) and fluvial geomorphology (Evan-Canfield et al., 2005). Fish habitat studies depend on the quality and continuity of the input topographic data. Conventional fish habitat studies are limited by the feasibility of field survey in time and budget. This limitation results in differences between the level of river management and the level of models; river management mostly operates at catchment or river sector level, while modeling uses the much smaller site level (Borsanyi et al., 2004). In order to facilitate upscaling processes from modeling to management units, intermediary methods between the micro- and the macroscale level were developed (Habitat Mapping (Maddock & Bird, 1996); MesoHABSIM (Parasiewicz, 2001)). Borsnyi et al. (2004) proposed a method based on the characterization and mapping of main river hydromorphological units defined according to the surface pattern (wave height), gradient, velocity and depth. LIDAR data of regulated River Cinca valley (Ebro Basin, Spain) was acquired in the low flow season, maximizing the recorded instream dry area. A topographic survey was carried out in a fluvial reach of the same segment recording position and substrate. DEM meshes obtained from LIDAR and conventional surveys were used as the input for hydraulic simulation for a range of flows. Velocity and depth outputs were combined with gradient data to produce a map reflecting the availability of

6742-38, Session 7
Irrigation mapping for agricultural drought monitoring in north China plain using MODIS remote sensing data
H. Tang, Z. Qin, M. Gao, J. Qiu, Chinese Academy of Agricultural Sciences (China) Drought is a critical factor threatening farming harvest in North China Plain, the most important agricultural region in China. Irrigation mapping consists of an undividable aspect of agricultural drought monitoring. In this study, we present our systematic investigation of mapping the irrigation area for drought monitoring in the plain using MODIS remote sensing data. Since farming in the region is a typical irrigation-supported system of winter wheat, followed by summer maize, spectral behaviors of winter wheat cropping system in MODIS data have been examined for construction of algorithm to identification of the system in the MODIS images. NDVI for green vegetation has been used to integrate with irrigation rate for algorithm development. Administrative border vector map at county level is used to overlay on the NDVI images for computation of cropping area. High spatial resolution ASTER data have been used to validate the mapping accuracy of irrigation area from MODIS data. Our results indicate that irrigation area can be

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

57

Conference 6742: Remote Sensing for Agriculture, Ecosystems, and Hydrology


properly mapped using the MODIS data. The mapping accuracy can meet the requirements of drought monitoring to administrate the available water resources for efficient utilization. Total irrigation area of the region is about 1.2 million hectares in 2006. This figure is smaller than that from statistical yearbook. The mapping results indicate that the spatial variation of irrigation area is very obvious in the region. More intensive irrigation can be observed on the eastern Hebei Province, the southern Shandong Province and the eastern Henan province. Agricultural water demands for various areas have been estimated on the basis of the irrigation mapping for anti-drought campaigns. components and model of biological object self radiation. Real-time monitoring system consists of subsystems: multiangular scanning, reflected or/and self radiation scanning and hardware-software processing scanned data from agricultural field of crops. Subsystem of hardware-software scanned data processing consists of multichannel correlation receiver, receiver to PC junction and program of analyzing and indication of scanning results. Multichannel correlation receiver operates correlation comparison of test signals and signals obtained during scanning process. Test signals are obtained previously for certain conditions of crops (kind of crop, its age, contained components and parameters of planting).

6742-40, Session 7
Spatial scale transferring study on leaf area index retrieval by using remotely sensed data
W. Zhang, Regional Ctr. for Temperate East Asia (China); S. Zhong, S. Hu, Nanjing Univ. (China) In order to effectively integrate various land-surface processes and environmental variables retrieved by different spatial resolution remotely sensed images, scaling issues induced by transferring for one resolution data to another have drawn more and more attentions. Taking Zhangye Oasis in the arid, semiarid rangeland of the Heihe River Basin situated in remote region of northwestern China and an temperate, humid inland watershed located in the centre part of China as experimental study sites, using the LAI map derived from the Landsat ETM+ image and NASA published MODIS LAI product for the study regions, an experimental study for up scaling the 30m spatial resolution LAI to 900m was conducted to quantitatively analyzes the introduced errors in the process of LAI up scaling with the final goal of establishing a new and much effective up scaling scheme based on NDVI pixel decomposing. It was found that the error smaller than 5% caused in LAI up scaling can be attributed to the non-linearity of algorithms utilized, and its scale is highly positive correlated with the terrain heterogeneities, terrain heterogeneity is the predominate causing for the errors introduced in LAI scaling transferring. The maximum error caused by terrain heterogeneity can be reached to about 75 %, however, the pure pixel LAI scale transferring is not affected by scaling. On the basis of above mentioned analyses, a newly proposed LAI up scaling algorithm in this study, based on pixel decomposing of the NDVI accounting for terrain heterogeneity in two key procedures, was successfully employed for the up scaling of 30 m spatial resolution LAI to 900 m with fairly satisfied resolution. This algorithm was used to correct NASA published MODIS LAI product and excellent results were obtained, which was back up with the fact that the corrected MODIS LAI product well consistent with the high spatial resolution LAI map in the study regions.

6742-42, Session 7
Assessment of crop yield estimation methods by using satellite imagery and ground observation
S. Shafian, M. R. Mobasheri, K.N. Toosi Univ. of Technology (Iran) Because of the population increment and the need for food security in national and regional scale, which increases the need for field level yield statistics. Therefore crop yield estimation is very important in national and regional scale. Crop yield is necessary, particularly in countries that depend on agriculture as their main source of economy. Such predictions warn the decision makers about potential reduction in crop yields and allow timely import and export decision. Therefore policy of agricultural_economic and yield price are influence by the accuracy and speed crop yield estimation. There are two methods for yield estimation. Conventional methods and remote sensing methods. conventional method are often complicated, costly, time consuming and they can not be run in large scale. Therefore it is necessary to use cheaper/faster methods for crop yield estimation. Remote sensing data has the potential and the capacity to provide spatial information at global scale; of features and phenomena on earth on an almost real-time basis. They have the potential not only in identifying crop classes but also of estimating crop yield. In this paper first, conventional methods and their disadvantages are discussed. Then crop yield estimation methods based on remote sensing data, their advantages and disadvantages are discussed.

6742-63, Session 3:
29 year climatology of satellite derived global land surface moisture
M. Owe, R. de Jeu, T. Holmes, NASA Goddard Space Flight Ctr. (USA) A 29-year global climatology of satellite derived land surface moisture is being developed by the NASA Goddard Space Flight Center and the Vrije Universiteit Amsterdam. The data consist of surface soil moisture, derived from both historical observations and currently active satellite microwave sensors, including Nimbus SMMR, DSMP SSM/I, TRMM TMI, and AQUA AMSR-E. The data span the period from November 1978 through 2007. The surface moisture retrievals are made with the Land Parameter Retrieval Model, which was developed jointly by researchers from the above institutions. The various sensors have different technical specifications, including primary wavelength, radiometric resolution, and frequency of coverage. It is expected that the data will be made available for download by the general science community early in 2008. Specifications and capabilities of different sensors, and how they affect soil moisture retrievals are presented. Global retrieval maps, time series, and comparisons between different sensors and frequencies are also presented.

6742-41, Session 7
Real-time monitoring of growth and biophysical properties of crops in millimeter and optical ranges
Y. V. Savenko, V. I. Vodotovka, National Technical Univ. of Ukraine (Ukraine) The paper presents the results on investigation of real-time monitoring of growth and biophysical properties of crops in millimeter and optical ranges. It is described an original realtime monitoring system investigates properties of biological objects in millimeter and optical range both in active and passive mode of scanning. It has been also used for remote sensing different biological object and geoscience task as well. In addition to the monitoring there have been investigated treatment technologies in millimeter and optical ranges are provided as correction of own properties of object as for medical, ecological and agricultural purposes. A model for agricultural field is rectangular with sides a and b with regular structure of crops strips (period T1, width of strip d and height of strip h). Scanning model consists of model of spatial scanning of field; model of reflecting of millimeter and laser radiation from binary heterogeneity as soil-crop; model of multiangular scattering from crop and especially from their

58

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6742: Remote Sensing for Agriculture, Ecosystems, and Hydrology

6742-64, Session 7:
Optimal land use/cover classification using remote sensing imagery for hydrological modeling in a Himalayan watershed
S. Saran, Indian Institute of Remote Sensing (India); G. Sterk, Wageningen Univ. (Netherlands); S. Kumar, A. Bharti, Indian Institute of Remote Sensing (India) Spatially distributed models of watershed hydrological processes require information on the spatial patterns of terrain, soils and land use/cover. Land use/cover is an important watershed surface characteristic that affects evapotranspiration, infiltration, surface runoff and erosion. Thus, almost any physically-based hydrological model uses some form of land use data or parameters based on these data. Many of these models divide the watershed into Hydrological Response Units (HRU), which are spatial units with expected similar hydrological behaviour. The division into HRUs requires good-quality data on land use/cover and its spatial pattern. A detailed analysis of the effects of the thematic accuracy of land use/cover is necessary before any attempt on using the land use/cover map for hydrological modelling. This paper presents a two-step approach to attain an optimal land use/cover map based on remote sensing imagery for a Himalayan watershed in northern India. The basic purpose of this two-step approaches was to eliminate as much as possible potential error from the land use/cover classification, especially for the class agricultural land. The first step in the approach was a digital classification using a decision tree classifier to extract knowledge from the spatial data in the form of classification rules. . This extracted knowledge was further used for improving the classification accuracy. The results obtained from the decision tree were compared with a maximum likelihood classification, and it was shown that the decision tree classifier resulted in a 10% more accurate classification. The satellite data used for this study are IRS P6 LISS-III (multispectral) and IRS IC PAN (panchromatic). The obtained land use/cover map was not sufficient for the delineation of HRUs. For that purpose, the class agricultural land needed to be sub-divided into terraced (paddy) fields and non-terraced fields. Moreover, for accurate hydrological modelling, it was also necessary to consider the hydrological responses for different crop cover during the monsoon season. So, the second step in the land use/cover classification was done to delineate more classes within the class agricultural land derived from step 1. Therefore we adopted a visual classification approach using optical data fused with ENVISAT ASAR data with HH polarisation and wide incidence angle (I6). This second step with level II classification system resulted into better classification accuracy within the agricultural land class. These detailed land use/cover classes need to be further combined with topography and soil type to derive HRUs for the physically-based hydrological modelling.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

59

Conference 6743: Remote Sensing of the Ocean, Sea Ice, and Large Water Regions 2007
Monday-Tuesday 17-18 September 2007 Part of Proceedings of SPIE Vol. 6743 Remote Sensing of the Ocean, Sea Ice, and Large Water Regions 2007

6743-01, Session 1
Multiresolution optical properties of rough sea surface in infrared
K. Caillault, S. Fauqueux, ONERA (France); C. Bourlier, Univ. de Nantes (France); P. Simoneau, L. Labarre, ONERA (France) In order to fulfill the requirement in background signatures models for the assessment of optronic systems performances, an analytical model of infrared optical properties of windroughened Gaussian sea surface is presented. Spatial variability from 1-meter to several kilometers is taken into account, as well as sub-metric variability (capillarity). For the considered spectral bandwidth, aggregation relations between geometrical optics and global optical properties are already known for centered Gaussian sea surface. But for resolutions under the sea surface correlation length, processes have non-zero mean value and literature expressions are incomplete. Thus optical properties expressions are extended to consider any surface length. They include hiding and shadowing functions and multiple reflections are neglected. Determination of the aggregation of optical surface properties requires the knowledge of surface slope statistical properties: slope mean vector-value and slope covariance matrix. These parameters are obtained by means of a new two-scale model. It consists in dividing the surface in two roughness scales, and in superimposing the small-scale variability (smaller than the surface patch) to the large-scale one. It is achieved by combining a statistical approach with the geometrical surface generation. Theoretical validity of this new approach has been verified. Our model has also been implemented for the generation of synthetic sea surface radiance images, showing its ability to produce good quality ocean scenes in various contextual conditions.

The infrastructure of the permanent facility on Gavdos includes tide gauges, permanent GPS satellite receivers, meteorological and oceanographic instruments, a DORIS satellite beacon, an electronic transponder, communications systems for the transmission of data, etc. Given the importance of the Gavdos installations, both from the scientific point of view and as a model satellite research centre, the continuation of its operation for the collection, analysis and distribution of data is necessary, as is the extension of applications to other satellite missions. The project is now continuing under the Ocean Surface Topography Mission program with funding from NASA and the Greek government. Current plans include the establishment of an identical setup at a site on the main island of Crete, at Kastelli, situated exactly under the descending Jason-1 pass 018. This gives us access to a second site and use of the altimeter measurements made to the north of Crete, in the Aegean Sea. It will thus allow the collection of additional information on the circulation and currents of the area between the Cape Maleas and Western Crete (e.g. Cretan cyclone). The project is now producing results on the basis of the new GDRs and extending our efforts to include the ENVISAT and GFO missions. We are also planning to repeat the co-location at the TUC facility site with the French Transportable Laser Ranging System that established an initial link of the entire GAVDOS network of sites with the ITRF2000 frame, in order to update the link to the global Terrestrial Reference Frame.

6743-04, Session 1
Cramer-RAO lower bounds for sinusoidal models from Topex/Poseidon data in the Indian Ocean
M. Arias Ballesteros, J. J. Alonso del Rosario, P. VillaresDurn, J. Gmez-Enri, M. Cataln Prez-Urquiola, I. Labrador Costero, C. Medina, Univ. de Cdiz (Spain) The Indian Ocean is a quite complex basin where strong currents, rings and fronts happen everywhere. The altimetry studies use to be quite difficult in this area because the energetic of the processes that take place in the areas where the currents break against the continents and to other currents. Because of these the authors have derived the Cramr-Rao lower bound (CRLB) for the estimations of the cosine and sine amplitudes of multi-tone sinusoidal model for TOPEX/Poseidon satellite altimetry data sets covering the Indian Ocean. The CRLB depends on the variance of the White Gaussian Noise that it is computed by Modern Parametric Autoregressive Adaptative Spectral Analysis. Determining CRLB parameters it is possible to establish the minimal error associated to any model built to work in the study area what improves the intrinsic bias of the generated time series. The noise that appears in the altimetry data depends strongly on the errors from the atmospheric and geophysical corrections, so the role of the inverted barometer and tidal corrections are also analyzed. The results is summarized as: a) the spatial structure of the order of the parametric model considering the application or not of the above corrections and its relationship to the surface dynamical system of currents in Indian Ocean; b) the spatial structure of the variance of the WGN in the area and its meaning; c) the CRLB for the Geoids estimators and the CRLB for the estimators of the semi-annual and annual waves.

6743-02, Session 1
Remote polarization measurements of the airsea interface
G. D. Gilbert, J. S. Schoonmaker, I. M. Petrosyuk, Y. Podobna, Advanced Coherent Technologies LLC (USA); J. J. Dirbas, PAR Government Systems Corp. (USA) Narrow band polarization measurements were taken from a bridge in San Diego Harbor using the Advanced Coherent Technologies Multimission Adaptable Narrowband Imaging Spectrometer (MANTIS) multichannel imaging system. MANTIS was capable of simultaneously collecting four channels of imagery through a narrowband green (532 nm) filter together with linear polarizers oriented at 0, 45, 90, and 135 degrees. This configuration enabled the collection of the first three Stokes Vector elements. The data is being gathered used to explore methods of calculating the sea surface Mueller Matrix. Methods, measurements, and models are presented.

6743-03, Session 1
Absolute calibration results for the radar satellite altimeters from the Eastern Mediterranean GAVDOS project
S. P. Mertikas, Technical Univ. of Crete (Greece); E. C. Pavlis, Univ. of Maryland/Baltimore County (USA) A permanent satellite facility on the island of Gavdos, Crete, Greece has been established to carry out calibration of satellite altimeters and also to determine and continuously monitor the absolute sea level. The objectives of the permanent facility have been (1) to carry out accurate measurements of the distances between the station and the satellite altimeters, (2) to ensure the absolute height of the surface of the sea, as this is implemented by the altimeters measurements on a world level, and (3) to monitor and observe any errors whatsoever in the satellite altimeters.

60

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6743: Remote Sensing of the Ocean, Sea Ice, and Large Water Regions 2007

6743-05, Session 1
High resolution multispectral photogrammetric imagery: enhancement, interpretation and evaluations
A. C. Roberts, C. R. Bostater, Jr., T. Becker, Simon Fraser Univ. (Canada) A variety of aerial mapping cameras were adapted and developed into simulated multiband digital photogrammetric mapping systems. Two multiband cameras (IIS 4 band and Itek 9 band) and 3 pairs of mapping and reconnaissance cameras were evaluated for digital spectral performance and photogrammetric mapping accuracy in an intertidal environment. Aerial films (24cm X 24cm format) tested were: Agfa color negative and extended red (visible and near infrared) panchromatic, and; Kodak color infrared and B&W (visible and near infrared) infrared. All films were negative processed to published standards and digitally converted at either 16 (color) or 10 (B&W) microns. Excellent precision in the digital conversions was obtained with scanning errors of less than one micron. Radiometric data conversion was undertaken using linear density conversion and centered 8 bit histogram exposure. This resulted in multiple 8 bit spectral image bands that were unaltered (not radiometrically enhanced) optical count conversions of film density. This provided the best film density conversion to a digital product while retaining the original film density characteristics. A series of underwater geometric line targets were imaged across a 3.5 meter tidal range for shell sand, eelgrass and sandstone benthic environments. Imagery varied from completely exposed targets (very low tide) to greater than 3 meters depth for all targets. Spectral and photogrammetric evaluations were undertaken for substrate and bathymetry determinations (spectral) and target spatial accuracy, stereoscopic depth evaluations and capillary and gravity wave line target distortions. Results indicate that multispectral photogrammetric systems offer improved intertidal feature mapping capability.

conditions where specific absorption spectral variation is not very high, and that application of multi-spectral algorithms can be more efficient for the retrieval of fluorescence contributions in coastal areas.

6743-08, Session 2
A physics-based approach to deriving optical water quality parameter concentrations from hyperspectral satellite data: a case study of Sacca di Goro, Italy
S. Salviato, J. Barbaro, F. Braga, Istituto di Scienze Marine (Italy) A remote sensing based methodology for monitoring coastal water quality was applied on Sacca di Goro (Po River Delta). This lagoon is one of the most important aquaculture systems in Italy but in the last decade the anthropogenic eutrophication processes and anoxic crises were intensified. For this case study a physics-based approach was adopted for mapping the water quality parameters by optical behavior. This approach is based on radiative transfer theory in the water body. The Semi-Analytical Model for Bathymetry, Unmixing, and Concentration Assessment (SAMBUCA) developed by the Environmental Remote Sensing Group, CSIRO-Land and Water of Canberra (Australia), was adopted for the project. By inverting this model, it was possible to assess, pixel by pixel, chlorophyll concentrations, dissolved organic matter and non-algal particle concentrations and to determine the bottom substratum composition and depth. From June to October 2005, in collaboration with the Province of Ferrara, a series of in situ measurements were carried out monthly to perform a spectral characterization of Sacca di Goro. Information on the spectral behavior of the bottom substratum, the apparent and inherent optical properties and the active optical component concentrations (chlorophyll, dissolved organic matter, and suspended matter) along the water column was collected. This information was used to parameterize the SAMBUCA model and to validate the water quality products obtained by inverting two Hyperion images acquired on 14 and 30 August during two tidal cycles.

6743-06, Session 2
Impact of scattering and absorption of photosynthetic pigments on fluorescence retrieval algorithms for coastal waters
S. Ahmed, A. Gilerson, J. Zhou, S. Hlaing, I. Ioannou, W. Jerez, B. M. Gross, F. Moshary, City College/CUNY (USA) Fluorescence Line Height (FLH) algorithms are effective for fluorescence retrieval in the open ocean where elastic reflectance in the fluorescence zone does not deviate much from the baseline. In coastal waters, FLH algorithms are significantly complicated by the overlap of the fluorescence and elastic reflectance peaks. To test accuracy of MODIS, MERIS and other FLH algorithms, we compared numerical simulations using an extensive synthetic database suitable for case II waters, with results of extensive field measurements of reflectance, absorption and attenuation spectra by us in the Chesapeake Bay, as well as satellite FLH data from several areas that typically show low correlation between [Chl] and FLH. Our synthetic datasets were created using the HYDROLIGHT radiative transfer code with IOPs connected to parameterized microphysical models in accordance with procedures used to generate the IOCCG dataset, but with some added improvements. These included higher (1 nm) spectral resolution, a wider range of parameters more typical of coastal waters, including chlorophyll specific absorptions with significant variations in spectral shapes and magnitude. HYDROLIGHT simulations of elastic reflectance using measured attenuation/ extinction spectra, followed by subtraction from measured reflectance, permitted retrieval of the fluorescence contribution to the latter, for comparisons with the data set simulations. We find relatively small fluorescence contributions to surface reflectance for mineral concentrations \> 5 mg/l because of strong attenuation in the excitation zone and enhanced elastic reflectance making fluorescence detection unrealistic. For lower mineral concentrations, we find that some combinations of NIR observation bands permit reasonably good FLH retrievals in

6743-09, Session 2
Subsurface unmixing with application to underwater classification
M. Vlez-Reyes, Univ. de Puerto Rico Mayagez (USA) This paper will discuss the problem of unmixing of hyperspectral imagery in subsurface aquatic remote sensing. Unmixing in optically shallow waters is set as a nonlinear inversion problem where the properties of the water column and the mixing fractions of a set of endmembers are retrieved together in hyperspectral remote sensing. The differences between land and aquatic unmixing in particular the role of the water column in distorting the spectral signature of the bottom and how different methods in the literature address this inverse problem are discussed. Examples using simulated and real hyperspectral imagery are presented.

6743-11, Session 3
Sea surface temperature (AVHRR) and ocean colour (MODIS) seasonal and interannual variability in the Macaronesian islands of Azores, Madeira and Canarias
A. M. Martins, Univ. dos Aores (Portugal); A. S. B. Amorim, Regional Directorate of Fisheries of Madeira (Portugal); M. P. Figueiredo, Univ. dos Aores (Portugal); R. J. Souza, Regional Directorate of Fisheries of Madeira (Portugal); A. P. Mendona, I. L. Bashmachnikov, Univ. dos Aores (Portugal); D. S. Carvalho, Regional Directorate of Fisheries of Madeira (Portugal) Within the framework of ORPAM (03/MAC/4.2/A2) project, five years (2001-2005) of Sea Surface Temperature (SST) and five years (2002-2006) of Ocean Colour (OC) satellite observations permitted to establish and compare the mean surface temporal variability among the Azores, Madeira, and Canarias regions.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

61

Conference 6743: Remote Sensing of the Ocean, Sea Ice, and Large Water Regions 2007
Satellite derived SST data (in C) were obtained from a time series of AVHRR data (from HAZO HRPT station in the Azores) and near-surface OC data (chlorophyll a in mg m-3) were obtained from a time series of MODIS data NASA/GSFC). Results show strong, but similar, SST seasonal variability patterns among the 3 regions with SST values ranging approximately between 15 and 27 C during winter and summer months, respectively. The largest SST differences among the 3 regions are observed during wintertime, with the Azores region representing the lowest SST values. Interannual SST variability is observed among the regions, with the largest variation occurring in Canarias. Ocean Colour seasonal variability patterns seem to vary inversely with SST. In the Azores, spring blooms dominate, while smaller autumn blooms are also observed. In Madeira, only spring blooms dominate. In Canarias, variability is high but not as evident as the other two regions, with the largest OC values observed during wintertime. Interannual OC variability is also observed and is mostly similar among the 3 regions, with the largest variation occurring in the Azores followed by Canarias (0.05 and 0.03 mg m-3, respectively). In general, our results suggest a strong latitudinal effect in the studied regions, particularly, among the Azores and Madeira regions. Canarias surface variability is less evident, principally in terms of OC. This most likely reflects the contribution of e.g. influence of African coastal upwelling and river discharges, OC algorithms failure to precisely estimate chlorophyll a in Case 2 waters; and/or winter mixing in the region.

6743-14, Session 3
Integrated coastal zone management plan and coastal zone information system for Udupi Coast, west coast of India
G. S. Dwarakish, S. A. Vinay, D. M. Shetty, J. B. Pai, K. Mahaganesh, National Institute of Technology Karnataka (India); U. Natesan, Anna Univ. (India) Coastal areas are under great pressure due to increase in human population and industrialization/commercialization and hence these areas are vulnerable to environmental degradation, resource reduction and user conflicts. Coastal Zone Management aims to achieve sustainable development, reduce vulnerability and maintain ecological balance. In the present study, a coastal stretch of approximately 90km extending from Mulki to Kundapur, in Udupi District, along Karnataka Coast, West Coast of India is considered as the study area. The study foresights to come up with a Integrated Coastal Zone Management Plan (ICZMP) by considering various parameters such as shoreline change detection, coastal slope, geomorphology, sea level rise and ocean parameters (tides, wind, waves, currents) and thus arrive at Coastal Vulnerability Indices (CVI). Using CVI values, critical erosion areas are also identified. The work also involves creation of maps/information database on vulnerable areas of coastal degradation, inundation areas during floods/tsunami using Digital Elevation Model (DEM) and changes in Land Use/Land Cover pattern required for ICZMP. CRZ maps are prepared to monitor developmental activities violating CRZ notification. The Study adopts RS and GIS tools extensively for analysis. Changes in Shoreline have been assessed considering three sets of data: 1967-1987, 1987-2000 and 2000-2006. Bathymetry maps are prepared using Naval Hydrographic Charts. Ground truth data has been collected through field validation using GPS. Digital Image Processing has been carried out using ERDAS Imagine 9.0 software. Overlay analysis and Map Composition has been done using ArcGIS. The results of shoreline change detection showed that the coastal stretch is in dynamic equilibrium which undergoes cyclic changes of erosion and accretion. It also showed shifting of river mouth over a period of time. All the outputs are incorporated in a user friendly Coastal Zone Information System (CZIS) interface using V.B.6.0. The system allows retrieval of required information and also has a provision to modify or upgrade database from time to time.

6743-12, Session 3
Comparing airborne high resolution hyperspectral imagery with aerial photgrammetric imagery for mapping submerged features in coastal shallow waters
C. R. Hall, Dynamac Corp. (USA); C. R. Bostater, Jr., Florida Institute of Technology (USA) Identification and mapping of submerged features in shallow coastal environments is of interest for natural resource management, coastal land development, and coastal security. A variety of airborne cameras and multi- and hyperspectral sensors are available to acquire remotely sensed information about the coastal environment but few studies compare and contrast their utility for specific applications. On May 5, 2003, aerial photographs at scales of 1:24000 and 1:9600 were collected for the Sebastian Inlet area of the Indian River Lagoon, Florida. During the following weeks, 1.2 m resolution hyperspectral imagery was collected for the same region along with extensive ground based data on depth, bottom type, and density and species composition of submerged aquatic vegetation (SAV) including drift algae. In this paper, we present a comparison between imagery obtained from traditional aerial photography and a modern digital hyperspectral sensor with respect to mapping SAV distributions and identification of individual taxa. Aerial photographs were scanned (RGB) into a digital format and overlaid after registration with the digital hyperspectral data. The hyperspectral data covered the 400 1000 nm wavelength region with approximately 3nm wide bands. Results demonstrate the capabilities of the 2 methods for SAV mapping when compared to detailed ground truth transects of SAV density and species composition.

6743-15, Session 4
The quasi-analytical algorithm of ocean inherent optical properties using two reference wavelengths
D. Pan, B. Yan, T. Mao, X. He, State Oceanic Administration (China) The water total absorption coefficient a( ) and total backscattering coefficient bb() are the key of inherent optical properties in ocean color remote sensing, which can be further used to retrieve the other inherent optical properties, such as the absorption coefficients of pigment, detritus and gelbstoff, and the beam attenuation coefficient, and further to retrieve the concentration of water components, dissolved and particulate organic carbon, etc. In this paper, a two reference wavelengths quasi-analytical algorithm is developed to retrieve the total absorption coefficient a() and total backscattering coefficient bb(). Firstly, the algorithm procedure of a() and bb( ) is described briefly. Secondly, the two reference wavelengths of 510nm and 555nm of total absorption coefficients (0) (which exclude the absorption of pure water) and particulate backscattering coefficient (0) are discussed in detail. Finally, three optical dataset were use to validate this algorithm, including the in situ dataset of China Yellow Sea and East China Sea in the spring of 2003, the synthesis dataset and the global in situ dataset from IOCCG. The results show that the two reference wavelengths quasi-analytical algorithm is quite promising both for Case I water and Case II water.

6743-13, Session 3
Comparison of airborne hyperspectral and photo images with synthetic images
C. R. Bostater, Jr., Florida Institute of Technology (USA) Comparisons are presented which demonstrate the utility of utilizing airborne hyperspectral images, high spatial resolution photogrammetric images and advanced radiative transfer model synthetic images. Results show the value of each of the above airborne remote sensing data types for use in helping to detect man-made targets and natural bottom features in harbors, waterways and coastal marine inlets. Methods are presented to demonstrate how sensor and model calibration is central to producing high quality data and images for coastal mapping and detection applications.

62

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6743: Remote Sensing of the Ocean, Sea Ice, and Large Water Regions 2007

6743-16, Session 4
The effect of aerosol scattering on the ocean color polarization correction
X. He, T. Mao, D. Pan, State Oceanic Administration (China); Z. Hao, Nanjing Univ. of Information Science & Technology (China) The effect of the polarization response is remarkably for some of the ocean color remote sensors, such as MODIS. The ocean color polarization correction algorithm of MODIS had been operational, and the variation of the polarization correction varies from 0.7%(at near infrared bands) to 5.4%(at blue light bands) for the Aqua MODIS. However, the current implementation of the MODIS polarization correction algorithm ignores the polarization component produced by aerosol scattering. In order to assess the effect of the aerosol scattering, we used the coupled ocean-atmospheric vector radiance transfer model-PCOART to simulate the top-of-atmosphere (TOA) linear polarization radiance at two case, one was the pure Rayleigh atmosphere which absent aerosol, the other case had the aerosol thickness of 0.2. The results showed that the aerosol scattering had negligible effect on the blue wavelength (443nm) except in the sun glint, however, the effect on the near infrared wavelength (865nm) was remarkable. Furthermore, we compared the TOA linear polarization radiances form POLDER remote sensed data and the simulated data using PCOART with the pure Rayleigh atmosphere, and the results also showed that aerosol scattering had negligible effect on the blue wavelength (443nm), but had large effect on the near infrared wavelength (865nm). Finally, we developed the improved ocean color polarization correction algorithm for MODIS, which considered the effect of the aerosol scattering on the TOA linear polarization radiance. By using the POLDER polarizing remote sensing data, this improved algorithm was validated, and the results showed that the accuracy of the polarization correction was improved not only at the blue wavelength, but also at the near infrared wavelength.

correction, compensation, or removal. Retrieval of atmospheric water content from remote sensing data is very important for atmospheric correction. This paper introduces and derives the algorithm about retrieving water content of atmosphere from AISA+ (Airborne Imaging Spectrometer for Application) data based on Modtran code. AISA sensor is put on the aviation platform covering the East Sea in China on July 29, 2005. AISA+ is Hyperspectral Imaging sensor including 61 bands from visible to NIR, and band 56 and band 59 are in water absorbed region and band 51 is in atmospheric window. So the three bands were used to in following expressions. The authors utilized 2 near-IR water vapor channels in addition to existing windows channels, it will be possible to derive the water vapor amount from AISA+ data in Eastern Sea in China and analyze the retrieving result. Finally, the relationships between the transmittance and atmospheric water content of AISA+ data was computed and analyzed. The maximum atmospheric water content retrieved from AISA image is 1.43 cm and the minimum water content is 0.04 cm, and average atmospheric water content is 0.40 cm. The results are consistent with the real conditions, so the methods are feasible for retrieval of atmospheric water contents from AISA+ data.

6743-19, Poster Session


Optical model for the water characterization of the highly turbid water of the Winam Gulf (Victoria Lake)
F. Santini, R. M. Cavalli, Istituto sullInquinamento Atmosferico (Italy); A. Palombo, S. Pignatti, Istituto di Metodologie per lAnalisi Ambientale (Italy) The study, proposed within the framework of the cooperation with Kenyan Authorities, has been carried out on the Kenyan part of the Lake Victoria. This lake is one of the largest freshwater bodies of the world where, over the last few years, environmental challenges and human impact have perturbed the ecological balance. Pollution and sediments loads from the tributaries rivers and antrophic sources caused a worrying increase of the turbidity level of the lake water. Secchi transparency index has declined from 5 meters in the 1930s to less than one meter in the 1990s. With the aim of providing an inexpensive way to gather information linked to the water clarity and quality, a method for remotely sensed data interpretation, devoted to produce chlorophyll a, coloured dissolved organic matter (CDOM) and total suspended solids (TSS) maps, has been assessed. At this purpose a bio-optical model, based on radiative transfer theory in water bodies, has been refined. The method has been applied on an image acquired on January 2004 by ENVISAT/MERIS sensor just a week after an in situ campaign took place. During the in situ campaign a complete data set for model refinement and products validation has been collected. This data comprise surface radiometric quantity and samples for laboratory analyses. The comparison between the obtained maps and the data provided by the laboratory analysis showed a good correspondence, demonstrating the potentiality of remote observation in supporting a the management of the water resources.

6743-17, Session 4
The spectral absorption coefficient model of phytoplankton in the coastal ocean
Z. Mao, State Oceanic Administration (China) The spectral absorption models of phytoplankton are established based on the in-situ measurements and the accuracy of the models is assessed with the comparisons among the models. A power function is used to study the phytoplankton absorption model with wavelength-dependent coefficients determined by the least squares method. The accuracy of the power model is assessed by three kinds of spectral error indices, which are root mean square error (RMSE), the spectral-average-relative error, the spectral average logtransformed relative error and with the average values of 0.07 m-1, -12.5% and -4.5%, respectively. The accuracy of the models is influenced by many factors such as the equation forms, the parameter values of the models, the phytoplankton species, pigments composition and particle size structure. The performances of the models are obviously different, even the same form of the model with different parameters. Two phytoplankton absorption models of Sathyendranath are validated and other three forms of phytoplankton absorption model are also developed to check whether the models are sensitive to chlorophyll a concentration (Chl) and to the forms of the equations. The results show that it is the parameters of the models that dominate the performances of the models instead of the form of the equations. It is also indicated that the absorption values estimated by the models fall within the ranges of the measured data and the models are difficult to be validated without information of the phytoplankton species.

6743-20, Poster Session


Preliminary results of atmospheric correction of a marine airborne multi-spectrum scanner (MAMS): case study in Chinese East Sea
D. Wang, D. Pan, T. Mao, J. Chen, State Oceanic Administration (China) So far, China Marine Surveillance Force has her own modern aerial equipments for marine law-execute with the advantage of functioning agilely and operation of larger surveillance coverage, providing powerful safeguard all-round, harmonious and sustainable development of coastal economy. On the plane, the marine airborne multi-spectrum scanner (MAMS), made by Shanghai Institute of Technology and Physics CAS, is an important portion. In this paper we will simply introduce a flight experiment using MAMS in Chinese East Sea at first. To retrieve quantitative information from MAMS data, accurately atmospheric correction is a necessary step. So we will focus on applying some atmospheric correction algorithms to the MAMS airborne

6743-18, Poster Session


Retrieval of atmospheric water content based on AISA+ data
Q. Cheng, D. Pan, D. Wang, J. Chen, T. Mao, State Oceanic Administration (China) Many applications require the elimination of atmospheric effects caused by molecular, particulate scattering and atmospheric water content factors, a process known as atmospheric

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

63

Conference 6743: Remote Sensing of the Ocean, Sea Ice, and Large Water Regions 2007
ocean color remote sensing data. Then case study of Zhoushan sea area in East Sea will be discussed and comparison of different correction methods will be done, including MODTRAN, Empirical Line Method, Dark object subtraction, etc. The result demonstrates that the atmospheric correction is effective to remove heterogeneous atmospheric effects and retrieve surface reflectance, and will suggest one promising method for MAMS in Chinese coastal water.

6743-23, Poster Session


Estimation of seawater optical parameters in the northern Tyrrhenian Sea
F. Maselli, Istituto di Biometeorologia (Italy); L. Massi, Univ. degli Studi di Firenze (Italy); C. Santini, Istituto di Biometeorologia (Italy); M. Pieri, Lab. per la Meteorologia e la Modellistica Ambientale (Italy) Global standard ocean colour algorithms may be inefficient to estimate the concentration of seawater constituents in specific cases. Local overestimation or underestimation of chlorophylla, suspended sediments and yellow substance are in fact quite common. To avoid this problem, our research group works on the local calibration empirical or semi-analytical algorithms through comparison to in situ measured data. The concentrations of chlorophyll-a, suspended sediments and absorption of yellow substance, were found for a number of samples near the coast of Tuscany (Italy). An unconventional algorithm was then developed and applied to satellite data (MODIS, MERIS, etc) for the retrieval of water constituents concentrations. This inversion algorithm is based on Spectral Angle Mapping (SAM), which is used to minimize the angle between simulated and measured remote sensing reflectances. The estimated parameters concentrations showed a lower error with respect to that obtained by a standard error minimization criterion. The monthly maps of chlorophyll-a concentrations obtained by applying the proposed algorithm to numerous satellite images confirm the oligotrophic nature of the Tuscan Sea, where high values of seawater constituent concentrations can be found only near the mouths of the main rivers.

6743-21, Poster Session


Wind field retrieval under high wind conditions by combined scatterometer and radiometer data
J. Zou, Q. Zhu, J. Chen, Y. Ding, State Oceanic Administration (China); Z. Chen, Zhejiang Univ. (China); M. Lin, National Satellite Ocean Application Service (China) With the availability of scatterometer data, surface wind vectors can be estimated from the backscatter measured over oceans, guarantee global, long-term monitoring of the winds on the oceans, which make them very valuable for climate studies and other use. At moderate wind speeds, the scatterometer wind speed is reliable. But at higher speeds, scatterometer retrieved wind always appear to underestimate the speeds especially in tropical cyclones, because of deficiencies of the geophysical model function for high winds, effects of rain caused attenuation, wind gradient influence, and the saturation of the backscattering under high wind. As a passive microwave sensor, radiometer do not show the phenomena of saturation under high wind, so it can be appropriate candidate to retrieve high wind. In this paper, combined scatterometer and radiometer data is used to retrieve wind field under high wind condition. Using in situ data and meteorological data as a criterion, comparison is made between the performance of scatterometer and radiometer retrieved wind, result show that it is better to use radiometer data as a replacement of scatterometer in wind speed retrieve. Theoretic analyse about the phenomena is given.

6743-24, Poster Session


Polarization of scattered solar radiation in the atmosphere-ocean spherical system in conditions of wind-water surface interaction
A. B. Gavrilovich, Instytut Fizyki (Belarus) Interpretation of optical observations in the Earth atmosphereocean spherical system resulting from the ground or the orbital space with purpose optical remote diagnostics of medium and prediction of aerosol climatic consequences derived from natural and industrial perturbations, requires use of adequate mathematical models of radiation transfer. The known mathematical models of large-scale process in a atmosphere and ocean be unable to reproduce adequately a real multiparametric picture of radiative field. In this report is given the performance of radiating transfer model developed by us. The radiation transfer in the atmosphere-ocean spherical system is the mathematical problem of radiation transfer theory, which solution is the vectorial magnitude (Stokes vector) with four components for intensity and state of radiation polarisation. The complicated structure aerosol extinction and gas absorption is taken into account, also the switching volcanic layers, unizotropic reflection of terrestrial cover, wind disturbance on a surfase of ocean, and the optical properties of water medium. For problem solution it is used one of modifications of Gfunction method developed by author. We are derived the polar and asimuth distributions of Stokes vector components for a hemisphere directions corresponding to ascending and descending streams of radiation. We changed also parameters of different conditions of wind disturbance on ocean surface because of wind-water interaction. The used colculation method of the field components is adapted to realization of numerical experiments for purposes of remote optical monitoring of natural and antropogeneus perturbations of environment. The calculations polarization characteristics of solar light are compared with experimental measurement data executed by author in areas of central part of Indian ocean.

6743-22, Poster Session


Interannual water level variations in Lake Izabal, Guatemala, Centroamerica, using radar altimetry, and its relationship with oceanographic features
C. Medina, M. Arias Ballesteros, J. Gmez-Enri, J. J. Alonso del Rosario, P. Villares-Durn, M. Cataln Prez-Urquiola, Univ. de Cdiz (Spain) It is well known that ocean-atmosphere circulation and dynamics affect the weather conditions over the continents and the ocean itself. The hydrologic cycle is driven by climatic parameters like precipitation, temperature, evaporation, winds and humidity. Hence, rivers water discharge and lakes water level variations are impelled by climatic conditions also. Lake Izabal is the largest lake in Guatemala (Centroamerica); its main tributary is the Polochic River, which contributes to approximately the 70% of the lakes water. The lakes level dynamics are related to the Polochic Rivers runoff and therefore to the precipitation/evaporation over the rivers catchment area. In this study we used ENVISAT RA-2 Geophysical Data Records orbits over the lake, coupled with in situ water level measurements in order to determine its water level fluctuations. Further, we used the multimission gridded Mean Sea Level Anomalies (MSLA) to determine oceanographic events like Southern Oscillation Cycle in the Pacific Ocean and the heat content in the Caribbean Sea and Western North Atlantic. The precipitation/evapotranspiration historical records over the lakes catchment area and the lakes surface were also analyzed. The Lake Izabal water level fluctuations are driven by the annual cycle of rainy and dry seasons. The most impressive result found was an abrupt lake level rise in July 2006 which is correlated to an abnormal precipitation in June 2006. Theoretically, this strong precipitation increase was forced by La Nia Southern Oscillation events during early 2006.

64

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6743: Remote Sensing of the Ocean, Sea Ice, and Large Water Regions 2007

6743-25, Poster Session


Ocean color atmospheric correction over the coastal region with multi-viewing satellite data
Y. Mitomi, Remote Sensing Technology Ctr. of Japan (Japan) The atmospheric correction for the ocean color remote sensing is the most important technique to retrieve the water leaving radiances, which are less than about 10% of the satellite radiances, and its major uncertainties are occurred by the estimation error of atmospheric aerosols that is highly variable in both space and time. Especially, the correction over the coastal region becomes more difficult because of the absorbing aerosol or turbid water background. To improve the atmospheric correction accuracy in these areas, the atmospheric correction algorithm with multi-viewing observation data was examined. In this paper, we present the application result to POLDER2 multi-viewing observation data, and also the compound product with MODIS.

6743-26, Poster Session


Water pollution analysis from Lidar investigations on Romanian Black Sea coast
J. G. Vasilescu, G. Pavelescu, L. Belegante, National Institute of Research & Development for Optoelectronics (Romania); C. Strechie, National Institute for Marine Geology and GeoEcology (Romania); S. M. Babichenko, A. Lisin, AS Laser Diagnostics Instruments (Estonia) The aim of the paper is to analyze the fluorescence characteristics of the polluted water on the Romanian Black Sea coast. A fluorescence LIDAR, based on an excimer (308 nm) and a dye laser (367, 460 nm) was used in order to map DOM (Dissolved Organic Matter) and chlorophyll variations in a marine area. Onboard ship campaign was performed during spring time to evidence the algae blooming. For seawater pollution analysis physico-chemical parameters and chlorophyll concentration were taking into account. The organic compounds and phytoplankton were characterized in correlation with the band of the water Raman scattering and by their specific fluorescence. The LIDAR data were combined with Acoustic Doppler Profiler System (ADP) measurements of the water velocity. Highly polluted areas were noted and mapped along the ship trajectory.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

65

Conference 6744A: Sensors, Systems, and Next-generation Satellites


Monday-Thursday 17-20 September 2007 Part of Proceedings of SPIE Vol. 6744A Sensors, Systems, and Next-generation Satellites

6744A-01, Session 1
The third cycle of Earth explorers core missions
P. Bensi, J. Bzy, C. Lin, J. Langen, M. Davidson, M. Berger, H. Rebhan, P. Ingmann, P. Silvestrin, European Space Agency (Netherlands) The Living Planet Programme of the European Space Agency encompasses a science-driven strategy for monitoring the Earth from space. The Earth Explorer missions are defined, developed and operated in close cooperation with the science community and focus on the key components of the Earth System: the atmosphere, biosphere, hydrosphere, cryosphere and the Earths interior. The emphasis of these missions is on providing data to advance our understanding of these individual components, their interaction with each other and the impacts that human activities have on natural Earth processes. By involving the science community from the beginning and introducing a peer-reviewed selection process, this on-going user-driven approach has given the Earth science community an efficient tool in their endeavour to better understand and monitor our planet. So far, this process has resulted in six missions currently under development: GOCE, Cryosat, ADM Aeolus, SMOS, Swarm and EarthCARE. The third cycle of Earth Explorers Core Missions was started in 2005 to select the seventh Earth Explorer mission due to launch in 2014/2015. At present there are six candidate missions being assessed at pre-phase A level. These missions were chosen to enter the assessment phase as a result of the Call for Core Mission ideas released by ESA in 2005, which resulted in 24 proposals out of which the following six candidates were selected: BIOMASS - global measurements of forest biomass and extent. TRAQ - (TRopospheric composition and Air Quality) - Air quality monitoring and long-range transport of air pollutants. PREMIER - (PRocess Exploration through Measurements of Infrared and millimetre-wave Emitted Radiation) Understanding the processes that link trace gases, radiation, chemistry and climate in the atmosphere. FLEX - (FLuorescence EXplorer) - Observation of global photosynthesis through the measurement of fluorescence. A-SCOPE - (Advanced Space Carbon and Climate Observation of Planet Earth) - Improving the understanding of the global carbon cycle and regional carbon dioxide fluxes. CoReH2O - (Cold Regions Hydrology High-resolution Observatory) - Detailed observations of key snow, ice and water cycle characteristics. This paper presents an overview of the six candidate missions, describing the scientific objectives and outlining the main aspects of the candidate implementation concept currently under evaluation.

(GSE) program. Additional inputs come from on-going GMES projects funded by ESA, EU (FP5 and FP6) and ESA/EU member states. Three priorities (fast-track services) for the mission have been identified by user consultation working groups of the European Union: Marine Core Services, Land Monitoring and Emergency Services. This paper describes the Sentinel-1 mission, an imaging synthetic aperture radar (SAR) mission at C-band. It provides an overview of the mission requirements, how they were derived and some insight into the preliminary technical concept for the satellite.

6744A-03, Session 1
ADM-Aeolus follow-on missions
M. J. Endemann, W. Veith, P. Dubock, European Space Research and Technology Ctr. (Netherlands); D. Morancais, F. Fabre, EADS Astrium (France); R. Wimmer, P. McGoldrick, EADS Astrium Ltd. (United Kingdom) ADM-Aeolus is a dedicated satellite to provide global observations of vertical wind profiles to demonstrate the capability to eliminate a major deficiency of the Global (Climate) Observing System. Aeolus is considered a precursor for an operational wind profiler system, which is expected to improve Numerical Weather Prediction analyses and forecasts in the tropics and extra tropics. Aeolus is under development for the European Space Agency with Astrium Satellites as prime contractor. Launch is planned for mid 2009 for a 3 year mission. There is likely to be a significant gap between the nominal end of life of Aeolus in late 2012, and the availability of wind profiles from post-EPS instruments in 2019 or later. The presentation will sketch a programme to fill this gap. It is based on copies of the present Aeolus satellite with minor modifications, like more vertical samples, an alternative lineof-sight, and measures for an increased lifetime. The programmatics required to fill the data gap will be discussed.

6744A-04, Session 1
Definition of mission requirements for the follow-on EUMETSAT polar system
P. L. Phillips, P. Schluessel, C. J. Accadia, J. J. W. Wilson, A. Perez-Albiana, S. Banfi, EUMETSAT (Germany) EUMETSAT and ESA have initiated joint preparatory activities for the definition of the Post EUMETSAT Polar System (PostEPS) needed for the 2019 timeframe as a replacement for the current EUMETSAT Polar System. Based on the first outputs of the EUMETSAT Post-EPS User Consultation process initiated in 2005, mission requirements for potential Post-EPS missions have been drafted. Expertise from a variety of communities was drawn upon in order to generate user needs expressed in terms of geophysical variables, for operational meteorology, climate monitoring, atmospheric chemistry, oceanography and hydrology. Current trends in the evolution of these applications were considered in order to derive the necessary satellite products that will be required in the Post-EPS era. The increasing complexity of models with regard to parameterisation and data assimilation, along with the trend towards coupled atmosphere, ocean and land models, generates new requirements, particularly in the domains of clouds and precipitation, trace gases and ocean/ land surface products. Pre-feasibility studies at instrument and system levels will commence in 2007 with support of industry and representatives of the user and science communities. Such studies, planned for completion by end of 2008, aim at defining and trading off possible mission and system concepts and will establish preliminary functional requirements for full or partial implementation of Post-EPS Mission Requirements. Cost drivers and needs for critical R&D will also be identified. The generation of both the user and mission requirements have been supported substantially by the Post-EPS Mission Experts Team and the Application Expert Groups. Their support is gratefully acknowledged.

6744A-02, Session 1
Sentinel-1 ESAs new European SAR mission
E. Attema, M. Davidson, N. Floury, G. Levrini, B. Rommen, European Space Agency (Netherlands); P. Snoeij, Dutch Space B.V. (Netherlands) In Europe the development of space technology is being coordinated by the European Space Agency under tight control of its Member States and their normal democratic procedures. For mature Earth observation applications other than meteorology a new European programme framework for operational satellites is emerging, based on the explicit requirements of services provided to the end-user. The Global Monitoring for Environment and Security (GMES) programme is a joint initiative of the European Union (EU) and the European Space Agency (ESA). As part of the GMES space component, ESA is undertaking the development of a European Radar Observatory (Sentinel-1), a European polar orbiting satellite system for the continuation of SAR operational applications. Sentinel-1 is an imaging radar mission at C-band aimed at providing continuity of data for user services. Special emphasis is placed on services identified in ESAs GMES service elements

66

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6744A: Sensors, Systems, and Next-generation Satellites

6744A-05, Session 1
Meteosat third generation, phase A, optical payload consolidation
D. M. A. Aminou, H. Stark, European Space Agency (Netherlands); R. Stuhlmann, A. R. Rodriguez, EUMETSAT (Germany) ESA and EUMETSAT have initiated joint preparatory activities for the formulation and definition of the Meteosat Third Generation (MTG) geostationary system to ensure the continuity and improvement of the Meteosat Second Generation (MSG) system. The MTG will become the new system to be the backbone of the European operational meteorological services from 2015, in particular, will ensure the continuation of the imagery missions. The first phases were devoted to the definition and consolidation of end user requirements and priorities in the field of Nowcasting and Very Short Term Weather Forecasting (NWC), Medium/Short Range global and regional Numerical Weather Prediction (NWP), Climate and Air Composition Monitoring and to the definition of the relevant observation techniques. The following missions have been analysed and preliminary concepts studied: - High Resolution Fast Imagery Mission (successor to MSG SEVIRI HRV mission) - Full Disk High Spectral Resolution Imagery Mission (successor to the mission of other MSG-SEVIRI channels) - Lightning Imagery Mission - IR Sounding Mission - UV-VIS-NIR Sounding Mission. After pre-phase A mission studies (2003-2006), where preliminary instrument concepts were investigated allowing in the same time to consolidate the technical requirements for the overall system study, a phase A study on MTG has been launched at the beginning of February 2007 for the space segment system feasibility and programmatic aspects to be accomplished during 2007-2008 time frame. The space segment phase A study will cover all elements to the level of details allowing to conclude on the feasibility of the system and to produce cost estimates with a good level of confidence. This paper addresses an overview of the outcome of the MTG space segment progress (spacecraft concept, payload preliminary design studies) accomplished in the frame of the phase A. It namely focuses onto the Imaging and IR Sounding Missions, highlights the platform and resulting instrument concepts, establishes the critical technologies and introduces the study progress towards the implementation of the MTG development programme.

next generation Earth Observation instruments have started, namely High Resolution Wide Swath SAR, IR detectors and coolers, optical components and mechanisms for LEO and GEO and high power Mixed Garnet laser transmitter. With these activities Germany will be able to provide future Earth Observation missions with suitable technologies as an answer to the increasing complexity of user requirements. In this paper the objectives and the strategy of the German Earth Observation Programme will be explained and the main elements, i.e. missions and technology developments as well as the plans for the future will be introduced.

6744A-07, Session 2
New optical sensor systems for high-resolution satellite, airborne and terrestrial imaging systems
A. Eckardt, A. Brner, F. Lehmann, U. Dombrowski, Deutsches Zentrum fr Luft- und Raumfahrt e.V. (Germany) The department of Optical Information Systems (OS) on the institute of Robotics and Mechatronics of the German Aerospace Center (DLR) has more then 25 years experiences with high resolution imaging technology. Technology changes on the Developments of detectors, the significant change of manufacturing accuracy in cobination with the engineering research defines the next generation of spaceborne sensor systems focusing on Earth observation and remote sensing. The combination of large TDI lines, intelligent synchronisation control, fast readable sensors and new focal plane concepts open the door to new remote sensing instruments. This class of instrument are feasible for high resolution sensor systems regarding geometry and radiometry and their data products like 3D virtual reality. Systemic approaches are essential for such designs of complex sensor systems for dedi-cated tasks. The system theory of the instrument inside an simulated environment starts the optimisation process of the optical, mechanical and electrical designs. Single modules and the entire system have to be calibrated and verified. Suitable procedures and procedures must be defied on component, module and system level for the assembly test and verification process. This kind of development strategy allows the hardware in the loop design. The paper gives an overview about the current activities at German Aerospace Center on the field of innovative sensor systems for photogrammetry and remote sensing.

6744A-08, Session 2
TROPOMI: solar backscatter satellite instrument for air quality and climate
J. de Vries, E. C. Laan, Dutch Space B.V. (Netherlands); R. W. M. Hoogeveen, R. T. Jongma, I. Aben, SRON Nationaal Instituut voor Ruimteonderzoek (Netherlands); H. Visser, E. C. Boslooper, TNO (Netherlands); H. K. Saari, VTT Optical Instruments (Finland); M. R. Dobber, P. Veefkind, Q. Kleipool, P. F. Levelt, Koninklijk Nederlands Meteorologisch Instituut (Netherlands) TROPOMI is a nadir-viewing grating-based imaging spectrograph in the line of OMI and SCIAMACHY. TROPOMI is part of the ESA Candidate Core Explorer Mission proposal TRAQ and also of the CAMEO satellite proposed for the US NRC decadal study. A TROPOMI-like instrument is part of the ESA/EU Sentinel 4&5 pre-phase A studies. TROPOMI covers the OMI wavelengths of 270-490 nm to measure O3, NO2, HCHO, SO2 and aerosols and adds a NIR channel and a SWIR module. The NIR-channel (710-775 nm) is used for improved cloud detection and aerosol height distribution. The SWIR module (2305 - 2385 nm) measures CO and CH4 and forms a separate module because of its thermal requirements. TROPOMI is a non-scanning instrument with an OMI-like telescope but optimized to have smaller ground pixels (10 x 10 km2) and sufficient signal-to-noise for dark scenes (albedo 2 %). TROPOMI has the same wide swath as OMI (2600 km). In TRAQs mid-inclination orbit, this allows up to 5 daytime observations over mid-latitude regions (Europe, North-America, China).

6744A-06, Session 2
The German Earth observation programme: building on the success of TerraSAR-X and RapidEye
P. Schaadt, Deutsches Zentrum fr Luft- und Raumfahrt e.V. (Germany) In addition to five years of routine operations of SCIAMACHY on-board of ESAs ENVISAT mission, the launches of the TerraSAR-X and RapidEye missions and the beginning of both their operational phases are the major milestones for the German Space Programme in 2007. These two missions will contribute significantly to the European GMES-Initiative and to the Global Earth Observation System (GEOSS) enhancing the knowledge about state and dynamics of the Earths system. Both missions are implemented under public-private-partnership between government and industry, an innovative and economic scheme for space mission implementation. With the TanDEM-X and the Hyperspectral EnMAP mission this efficient way of sharing competences and responsibilities on one hand and benefits on the other hand will be further followed. In addition, with MetImage Germany started the development of an imaging radiometer for the European post-EPS satellite system of EUMETSAT. This important contribution of the German Earth Observation Community with further five attractive missions to the global system is seen as the adequate equivalent to the political approach for the European and German leadership at the environmental agenda. In parallel technology developments for

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

67

Conference 6744A: Sensors, Systems, and Next-generation Satellites


The paper gives a description of the TROPOMI instrument and focuses on several important aspects of the design, for example the sun calibration and detector selection status. Detector selection is important because the small ground pixels of TROPOMI result in high internal data rates for which use of CCDs will yield larger dissipations. CMOS-type detectors are a step forward in this respect and provide a means to integrate front-end electronics inside the detector chip. CMOS-type detectors can also be made more resistant to proton-induced radiation damage. Sun calibration is important because of the wide range of viewing angles in the mid-inclination orbit and because of spectral features introduced via sun diffusers. Aspects related to the TROPOMI SWIR module are discussed in a separate paper in this conference. and will carry two instruments. JAXA will develop DPR (Dual frequency Precipitation Radar) which is a follow on of PR on TRMM. Another project is EarthCare. It is a joint project with ESA and JAXA is going to provide CPR (Cloud Profiling Radar). Discussions on future Earth Observation programs have been started including discussions on ALOS F/O.

6744A-12, Session 3
Application of ASTER data to hazard evaluation with an example of damage evaluation of the landslide in Pakistan
H. Watanabe, National Institute for Environmental Studies (Japan) There have been a lot of discussion on application of remote sensing data to hazard evaluation. For the effective evaluation of the hazard, the data should have been acquired before the hazard occurrence and should be quickly acquired after it . To fulfill the first condition, it is necessary to have world coverage of data. For the second condition, flexible and timely data acquisition is necessary. ASTER seems to fulfill the both conditions. In case of the giant landslide occurred in Pakistan on October 8, 2005, the data before and just after the landslide are both available. And the damage was quantitatively evaluated by using DEM generated from ASTER stereo pairs obtained before and after it.

6744A-10, Session 2
The microwave humidity sounder (MHS): in-orbit performance assessment
R. Bonsignori, EUMETSAT (Germany) The Microwave Humidity Sounder (MHS) is the high-frequency microwave radiometer of the ATOVS (Advanced TIROS Operational Vertical Sounder) instrument suite of the IJPS (Initial Joint Polar System), the joint EUMETSAT-NOAA programme for operational satellite meteorology. Five MHS models have been built by Astrium Ltd under EUMETSAT contract, two of which are currently operational on the NOAA-18 and Metop-A satellites. The MHS instrument replaces the former AMSU-B in the operational microwave sounder suite. The present paper provides a summary description of the MHS instrument and describes in some detail its in-orbit performance and functionality, along with a comparison with similar instruments. The in-orbit performance has been extensively assessed during the SIOV (Satellite In-Orbit Verification), the first part of the commissioning phase aimed at activating the payload and verifying its operation, and is periodically monitored throughout the mission life. The performance relevant to the SIOV and operational phase of the MHS instruments on both NOAA-18 and Metop-A are presented. With respect to its predecessor AMSU-B, the MHS instrument constitutes a sensible improvement in terms of radiometric sensitivity and calibration accuracy, while allowing full continuity of the acquired data and relevant processing.

6744A-13, Session 3
Accuracy assessments of standard products of ALOS optical instruments and their high level products
T. Tadono, M. Shimada, Japan Aerospace Exploration Agency (Japan); J. Takaku, S. Kawamoto, Remote Sensing Technology Ctr. of Japan (Japan) The Advanced Land Observing Satellite (ALOS, nicknamed Daichi) was successfully launched on January 24, 2006. This paper introduces the results of calibration and validation and image quality evaluation after the launch for two ALOS optical sensor, the Panchromatic Remote-sensing Instrument for Stereo Mapping (PRISM) and the Advance Visible and Near Infrared Radiometer type-2 (AVNIR-2). PRISM consists of three independent panchromatic radiometers and is used to derive a digital surface model (DSM) with high spatial resolution, which is an objective of the ALOS mission. The preliminary results of the initial calibration and validation and image quality evaluations of PRISM and AVNIR-2 have been presented. At that time, we analyzed data acquired during the initial mission check (IMC) and the initial calibration phase (ICP), which was spent about 9 months after the launch. As the results of ICP, the sensor characterizations have been evaluated, and the radiometric accuracies of both PRISM and AVNIR-2 were almost sufficient except for band-4 of AVNIR-2. However, the geometric accuracies were not sufficient due to the satellite attitude have not been precisely determined yet, and the offset components (i.e., sensor alignments) of PRISM could not been evaluated after the launch. We are also developing DSM and ortho-rectified image generation software as a part of validation. Those products are defined as a high level product in JAXA. Thus, geometric calibration is important in generating a highly accurate DSM with stereo pair images of PRISM. Relative radiometric calibration is carried out using acquired images over homogeneous targets such as ocean, deserts, ice and snow areas, and nighttime observation. The absolute radiometric calibration is applying with the cross-calibration method using calibrated satellite images (e.g. MODIS aboard Terra and Aqua satellites, ASTER, and SPOT-5). In this paper, the updated results of calibration and validation as well as stabilities of sensor characterization will be given in this paper.

6744A-11, Session 3
Overview of Japanese Earth observation programs
H. Shimoda, Japan Aerospace Exploration Agency (Japan) Three programs, i.e. TRMM, ADEOS2 and ASTER, are going on in Japanese Earth Observation programs. TRMM and ASTER are operating well. ADEOS2 was failed, but AMSR-E on Aqua is operating. On 24th, Jan. 2006, ALOS (Advanced Land Observing Satellite) was launched successfully from Tanegashima Space Center by HIA launcher. ALOS carries three instruments, i.e., PRISM (Panchromatic Remote Sensing Instrument for Stereo Mapping), AVNIR-2 (Advanced Visible and Near Infrared Radiometer), and PALSAR (Phased Array L band Synthetic Aperture Radar). PRISM is a 3 line panchromatic push broom scanner with 2.5m IFOV. AVNIR-2 is a 4 channel multi spectral scanner with 10m IFOV. PALSAR is a full polarimetric active phased array SAR. PALSAR has many observation modes including full polarimetric mode and scan SAR mode. The spacecraft is operating well as well as all the 3 sensors are operating well. Next generation satellites will be launched in 2008-20012 timeframe. They are GOSAT (Greenhouse Gas Observation Satellite), GCOM-W and GCOM-C (ADEOS-2 follow on), and GPM (Global Precipitation Mission) core satellite. GOSAT will carry 2 instruments, i.e. a green house gas sensor (TANSO-FTS : Thermal and Near Infrared Sensor for Carbon Observation) and a cloud/aerosol imager (TANSO-CAI). TANSOFTS is a Fourier transform spectrometer (FTS) and covers 0.76 to 15 m region with 0.2 cm-1 resolution. TNASO-CAI is a 5 channel pushbroom scanner to observe aerosols and clouds. GCOM-W will carry AMSR2 (AMSR follow on). GCOM-C will carry SGLI (GLI follow on). GPM is a joint project with NASA

6744A-14, Session 3
Greenhouse gases observing satellite (GOSAT) sensor and satellite system
T. Hamazaki, Japan Aerospace Exploration Agency (Japan) Japan Aerospace Exploration Agency (JAXA) is developing Greenhouse gases Observing Satellite (GOSAT). The mission objective of the GOSAT is to observe the global distribution of

68

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6744A: Sensors, Systems, and Next-generation Satellites


CO2 and CH4 and their changes from space. The accuracy of CO2 columnar density is expected between 1-4ppm. GOSAT has Thermal And Near infrared Sensor for carbon Observation (TANSO-FTS) and Cloud and Aerosol Imager (TANSO-CAI). The TANSO-FTS is a Fourier Transform Spectrometer and covers wide range of spectrum including 0.76-14um in 0.2cm-1 spectrum resolution. The 0.76um band is used to observe O2 density and determine the exact path length. The 1.6 and 2.0um bands are used to observe CO2 density. The 1.6um band is also used to observe CH4 density. The 5.5-14um band is used to observe CO2, CH4, water vapor, atmospheric temperature and pressure. CO2 and CH4 altitude distribution data is also available with this band. The number of total observation channel reaches up to 18,000. The TANSO-CAI is used to compensate errors which are caused by the cloud and aerosol. The GOSAT is the joint endeavor with JAXA, Ministry of Environment (MOE) and National Institute for Environmental Studies (NIES). The GOSAT will be launched by the H-IIA launch vehicle in August 2008. GOSAT data will be available worldwide mainly through NIES. Data distribution through ESA, NASA or World Data Center for Greenhouse Gases (WDCGG) is also under coordination. This paper shows the overview of GOSAT project, sensor and satellite design and its performance. is remained in instrument line shape (ILS), which is required by the simulation of observed spectra for TANSO optics. After launching, the ILS of 1.6um band is validated by using onboard 1.55m diode laser. The ILS of the other channels is validated by using solar spectra. The processed data after the onboard calibrations will be evaluated by vicarious method using accumulated observation data. The vicarious calibration for radiance is studied by preparing the appropriate targets on the earth by using current satellite dataset, such as MODIS, which observes the equivalent wavelengths of the GOSAT observation band. The calibration sites can be used for the radiance comparison with the other satellite observation data. As for SWIR targets, deserts, forests, and snowfield will be selected. As for TIR targets, sea surface temperature is well known by other sensor observations or reliable dataset. This presentation shows the overview of processing system of calibration for GOSAT sensors.

6744A-17, Session 4
Development of GOSAT ground data system at NIES and the data processing strategy
H. Watanabe, T. Yokota, T. Matsunaga, K. Hiraki, H. Ishihara, National Institute for Environmental Studies (Japan) GOSAT Project is a joint project of MOE (Ministry of the Environment), JAXA (Japan Aerospace Exploration Agency) and NIES (National Institute for Environmental Studies (NIES). Data acquired by TANSO-FTS (Fourier Transform Spectrometer) and TANSO-CAI (Cloud and Aerosol Imager) on GOSAT will be collected at Tsukuba Space Center \@ JAXA. The level 1A and 1B data of FTS ( interferogram and spectra, respectively ) and the level 1A of CAI ( uncorrected data ) will be generated at JAXA and will be transferred to GOSAT Data Handling facility (DHF) at NIES for further processing. Radiometric and geometric correction will be applied to CAI L1A data to generate CAI L1B data. From CAI L1B data, cloud coverage and aerosol information ( CAI Level 2 data) will be estimated. The FTS data that is recognized to havelow cloud coverage by CAI will be processed to generate column concentration of carbon dioxide CO2 and methane CH4 ( FTS Level 2 data ). Level 3 data will be global map column concentration of green house gases averaged in time and space. Level 4 data will be global distribution of carbon source/sink model and re-calculated forward model estimated by inverse model. Major data flow will be also described. The Critical Design Review of the DHF will be completed in early July of 2007 to prepare the scheduled launch of GOSAT in August 2008. The data products can be searched and will be open to the public through GOSAT DHF after the data validation process.

6744A-15, Session 3
Retrieval performance of GOSAT thermal infrared FTS sensor for measuring gas concentrations and cloud properties
R. Imasu, N. Saitoh, Y. Niwa, The Univ. of Tokyo (Japan) Greenhouse gases Observing SATellite (GOSAT) is a Japanese satellite that aims to observe CO2 concentration from space and contribute to advance the research on the estimation of CO2 surface flux using source/sink inversion technique. This is a joint mission of the Ministry of the Environment (MOE), the National Institute for Environment Studies (NIES), and the Japan Aerospace Exploration Agency (JAXA). The main sensor of GOSAT is a Fourier Transform Spectrometer (FTS) which covers wide spectral range of CO2 absorption bands, namely, 1.6m and 2.0m (SWIR), and 15m (TIR). The former two bands will be used to estimate columnar concentration of CO2, and the latter to retrieve the vertical profile of CO2 concentration in the upper atmosphere above about 700hPa. Also installed on the satellite is an imaging sensor, Cloud and Aerosol Imager (CAI), which will be used to detect clouds and aerosols. Center for Climate System Research (CCSR) has contracted with JAXA to develop algorithms for retrieving CO2 and CH4 concentrations from the FTS-TIR data. We present the total performance of the sensor for retrieving atmospheric parameters such as gas concentrations and cloud optical and microphysical properties on the basis of simulation studies. Particularly concerned are resolving power of CO2 vertical profile and ability to estimating CO2 concentration in the boundary layer by combining FTS-TIR and FTS-SWIR data. Sensitivity to the thermodynamic phase (ice/water) of cloud particles is also discussed.

6744A-18, Session 4
Development of the DPR algorithms and products for GPM
S. Shimizu, R. Oki, M. Kachi, H. Hanado, M. Kojima, Japan Aerospace Exploration Agency (Japan); T. Iguchi, National Institute of Information and Communications Technology (Japan); K. Nakamura, Nagoya Univ. (Japan) Global Precipitation Measurement (GPM) started as an international mission and follow-on mission of the TRMM project to obtain more accurate and frequent observations of precipitation than TRMM. An important goal for the GPM mission is the frequent measurement of global precipitation using a GPM core satellite and a constellation of multiple satellites. The GPM core satellite is also developed by the US and Japan, while the constellation satellites are developed by various countries. The accurate measurement of precipitation will be achieved by the Dual-frequency Precipitation Radar (DPR) installed on the GPM core satellite. DPR consists of two radars, which are Ku-band (13.6 GHz) precipitation radar (KuPR) and Ka-band (35.5 GHz) radar (KaPR). KaPR will detect snow and light rain, and the KuPR will detect heavy rain. In an effective dynamic range in both KaPR and KuPR, drop size distribution (DSD) information and more accurate rainfall estimates will be provided by a dual-frequency algorithm. The frequent precipitation measurement every three hours at any place on the globe will be achieved by eight constellation satellites with microwave radiometers (MWRs). JAXA/EORC is responsible

6744A-16, Session 3
Calibration plan of GOSAT sensors
K. Shiomi, Japan Aerospace Exploration Agency (Japan) Greenhouse gases Observing SATellite (GOSAT) is aimed at observing the greenhouse gases, such as CO2 and CH4, from space. Thermal And Near infrared Sensor for carbon Observation (TANSO) is carried on the GOSAT satellite. TANSO is composed of 2 instruments, Fourier Transform Spectrometer (FTS) for measuring greenhouse gases absorption spectra, and Cloud and Aerosol Imager (CAI) for cloud detection and the correction of cirrus and aerosol interference within the FTS field of view. The interferogram data (L1A) is converted into calibrated spectra data (L1B) at ground processing system. The processing algorithm is developed through analyzing the laboratory experimental data Main procedure of calibration processing is phase correction of interferogram pairs between observation and calibration data. Spectral accuracy is related with stability of sampling laser wavelength. When the laser wavelength is estimated precisely, the calibrated wavelengths are in good agreement with the atmospheric absorption lines. Other issue

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

69

Conference 6744A: Sensors, Systems, and Next-generation Satellites


for the GPM/DPR algorithm development for engineering values (Level 1) and physical products (e.g. precipitation estimation) (Level 2 and 3) and the quality control of the products as the sensor provider. It is also important for us to produce and deliver 3-hourly global precipitation map in real time in order to make useful for various research and application areas (i.e., the prediction of the floods). To secure the quality of estimates, the mission must place emphasis on validation of satellite data and retrieval algorithms. This is a very difficult issue and needs careful consideration and strategy.

6744A-21, Session 4
Design challenge on forthcoming SGLI boarded on GCOM-C
H. Masaru, NEC TOSHIBA Space Systems, Ltd. (Japan); K. Tanaka, Y. Okamura, Japan Aerospace Exploration Agency (Japan); T. Amano, K. Shiratama, NEC TOSHIBA Space Systems, Ltd. (Japan) The Japanese Aerospace Exploration Agency (JAXA) has the plan of the Global Change Observation Mission (GCOM) for monitoring global environmental change. SGLI (Second generation Global Imager) is a mission instrument to be installed on the GCOM-C (one satellite of GCOM Mission: Climate) satellite. SGLI is the optical radiometer observed to the frequent Global, Ocean, Land, Cloud and Ice sphere to help determine the Earths climate change. SGLI is a suite of two radiometers that operate in the Visible to NIR called VNR and the SWIR to TIR called IRS. The VNR is employing a wide swath (1150km) push-bloom scan with line CCD detector and lens optics. IRS is employing a combinational cross-track scan mirror (1400km swath) and mirror collecting optics with cooled PV detector. SGLI observation features are follows. - Image of the Land surface cover at 250m resolutions. - Image of the Ocean surface and the Polar region at 1km resolution. - Three direction polarization observation at visible light. - Land surface Temperature with 500m resolutions. - Ocean surface Temperature with 1km resolution. We report the SGLI preliminary design and special feature. The current SGLI is BBM development phase which is underway to confirm the feasibility of the design.

6744A-19, Session 4
Five years of AMSR-E monitoring and successive GCOM-W1/AMSR2 instrument
K. Imaoka, M. Kachi, A. Shibata, M. Kasahara, Y. Iida, Y. Tange, K. Nakagawa, H. Shimoda, Japan Aerospace Exploration Agency (Japan) Japan Aerospace Exploration Agency (JAXA) has been proposing the Global Change Observation Mission (GCOM). GCOM will consist of two series of medium size satellites: GCOM-W (Water) and GCOM-C (Climate). The mission will take over the Advanced Earth Observing Satellite-II (ADEOS-II or Midori-II). During the Japanese fiscal year 2006, the GCOMW1 satellite (the first generation of GCOM-W series) was intensively evaluated through the Space Activities Commission of Japan and the JAXA internal review processes and approved for moving on to the Phase-C/D. Current target of launch date is beginning of 2012. The Advanced Microwave Scanning Radiometer-2 (AMSR2) is sole mission instrument onboard the GCOM-W1 satellite. Although the simultaneous observation by a microwave scatterometer and AMSR2 is still desired, installation of the scatterometer is not the case at least for the GCOM-W1 satellite. AMSR2 is a successor of the AMSR for the EOS (AMSR-E) provided to the NASA Aqua satellite and AMSR onboard Midori-II with some improvements based on the experiences of AMSR and AMSR-E. They include an improvement of warm-load calibration target and an addition of 7.3 GHz channels to help mitigating radio-frequency interference issue. The AMSR-E instrument is still providing continuous data records more than 5-years. Observed brightness temperatures and retrieved geophysical parameters are being widely used for monitoring environmental changes and for applying to the operational applications such as numerical weather forecasting. We expect a long-term continuity by leading the GCOM-W/AMSR2 to the AMSR-E observation.

6744A-22, Session 4
Japanese cloud profiling radar for EarthCARE
T. Kimura, Japan Aerospace Exploration Agency (Japan); H. Kumagai, National Institute of Information and Communications Technology (Japan) Cloud Profiling Radar (CPR) for EarthCARE satellite is going to be developed under cooperation of Japanese Aerospace Exploration Agency (JAXA) and National Institute of Information and Communications Technology (NICT) in Japan. CPR is a W-band (94.05 GHz) Doppler Radar to observe three dimensional cloud structure and up/downdraft in clouds. A large precise antenna reflector; 2.5 m diameter, long life and high power EIK and low noise amplifier, make it possible to observe most radiatively significant clouds with its high sensitivity; -35 dBZ. The EarthCARE CPR will be the first Doppler radar in space to observe updraft in cloud layer globally. As the integrated observation system of EarthCARE, synergy observation with other three instruments of EarthCARE, i.e. Atmospheric LIDAR (ATLID), Multi Spectral Imager (MSI) and Broad Band Radiometer (BBR), is aimed to reveal aerosol and cloud interaction and relationships with radiation budget. The development phase of CPR is now in extended Phase-A, JAXA and NICT are going to proceed to Phase-B from the end of 2007. Finally EarthCARE with CPR is planned to be launched in 2012 after integration with satellite bus and three sensors, which are going to be developed by ESA.

6744A-20, Session 4
A study on the possibility of land vegetation observation with SGLI/GCOM-C
Y. Honda, Chiba Univ. (Japan) and SORST-JST (Japan); M. Moriyama, Nagasaki Univ. (Japan); A. Ono, Japan Aerospace Exploration Agency (Japan) The fourth IPCC ( the Intergovernmental Panel on Climate Change ) report predicted that earth average temperature after 100 years rose to 6.4 degrees with a the worst case. This influence ( the global change ) will become a serious thing for human society. Future earth observation using satellite data should monitor global climate change, and should contribute to social benefits. Especially, human activities has given the big impacts to earth environment. JAXA (former NASDA) has the plan of the Global Change Observation Mission (GCOM) for monitoring of global environmental change. SGLI (Second Generation GLI) onboard GCOM-C (Climate) satellite, which is one of this mission, is an optical sensor from Near-UV to TIR. Characteristic specifications of SGLI are follows, - 250m resolutions over land and area along the shore - Three direction polarization observation (red and NIR) - 250m resolutions temperature over land and area along the shore These characteristics are useful in many fields of social benefits. For example, multiangular observation and 250m high frequency observation bring new knowledge in monitoring of land vegetation. It is expected that land products with land aerosol information by polarization observation are improved remarkably. We are studying these possibilities by ground data using a radio control helicopter ( FUGINS ).

6744A-23, Session 4
Cloud observations from future Japanese satellite missions
T. Y. Nakajima, H. Ishida, Tokai Univ. (Japan); T. Nakajima, The Univ. of Tokyo (Japan); H. Shimoda, Tokai Univ. (Japan) Observations of cloud optical and microphysical properties from future Japanese satellite mission will be discussed and summarized in this paper. Aerosol indirect effects are the major uncertainty of climate change studies through the significant effects on the changes of cloud optical and microphysical properties. Thus, the global monitoring of cloud and aerosol distributions and their microphysical properties are needed. Historically, the cloud parameters have been retrieved from visible-to-infrared multi-

70

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6744A: Sensors, Systems, and Next-generation Satellites


spectral imagers such as AVHRR/NOAA, GLI/ADEOS-II and MODIS/Terra&Aqua satellites for nearly thirty years. These data are used to detect the variation of cloud properties with the climate changes, and also used to understand the cloud droplet growing process by comparison the retrieved results with cloud model outputs. GOSAT, EarthCARE, and GCOM are three future Japanese satellite missions that are applicable to the cloud observations. These missions have multi-spectral imagers. The characteristics and the mission background are different. In the GOSAT mission, the cloud masking flags will be provided by the Cloud and Aerosol Imager (CAI) to mitigate the uncertainty of CO2 retrieval by the Fourier Transform Spectrometer (FTS). The EarthCARE is the unique mission that equipped cloud radar, lider, broadband radiometer, and multi-spectral imagers on one platform. The multifaceted 3D observations of clouds from the EarthCARE equipments will contribute to reveal the cloud growing process, aerosol effects on the clouds, and radiation budget of cloud and aerosol. The GCOM, the follow-on mission of ADEOS series, is planed and designed to monitor geophysical parameters from land/ocean to atmosphere for more decadal period. The authors will discussed, in this paper, the role of each missions showing some previous results obtained passed and ongoing missions, and suggest some synergy studies focusing cloud observations. (POD) System but with significant enhancements to improve performance and reliability to progress the sensor suite towards an operational capability. The Poseidon-3 Altimeter and one of the POD systems - Doppler Orbitography and Radiopositioning Integrated by Satellite (DORIS) will be provided by CNES. The Advanced Microwave Radiometer (AMR) and the two remaining POD systems, Global Positioning System Payload (GPSP) and the Laser Retroreflector Array (LRA) are provided by NASA/JPL. Three additional new instruments (T2L2, LPT, Carmen-2) will also be onboard, to study radiation in the satellite environment (LPT, Carmen-2), and to make high accuracy comparison of distant clocks by laser link (T2L2) . In addition to the basic scientific interest, these instruments should allow for future enhancements in data quality and accuracy for future ocean altimetry missions. This paper will provide a status of the current mission development and highlight the enhancements in the sensor suite leading toward the future utilization as an operational measurement.

6744A-26, Session 5
Advanced remote-sensing imaging emission spectrometer (ARIES): an instrument concept for a next-generation imager/sounder
T. S. Pagano, M. T. Chahine, F. G. OCallaghan, Jet Propulsion Lab. (USA) The success of the NASA EOS MODIS and AIRS programs have demonstrated the value to the scientific and meteorological community of space based imaging and sounding. Technology advancements over the last two decades enable significant enhancements in the performance of these systems. It is now possible to combine high spectral resolution observations (like AIRS) with moderate spatial resolution (like MODIS) in a single instrument. The ARIES provides over 3000 spectral channels in the 3.6 - 15.4 m spectral region with spatial resolution of 1.0 km while still scanning 55. The ARIES has size and mass less than half that of AIRS or MODIS primarily due to the advancements in focal plane assemblies and wide field optical systems developed under the NASA IIP and in US industry. The combined capability will allow more cloud free observations per unit area, and improve the overall cloud-clearing approach applied on AIRS. ARIES will improve sensitivity to atmospheric water vapor, temperature and trace gases in the boundary layer and facilitate studies of surface and atmospheric interaction for global climate studies. ARIES will also improve regional weather forecast and improve quality of products for assimilation including boundary layer water vapor, temperature and surface emissivity. This paper will discuss the primary requirements for ARIES, the expected science and operational benefits and the technology available today that demonstrates the viability and low risk of the approach

6744A-24, Session 5
NASAs Earth science flight programs
S. P. Neeck, T. F. Hammer, NASA Headquarters (USA) NASAs strategic goal in Earth science is motivated by the fundamental question: How is the Earth changing and what are the consequences for life on Earth? NASAs mission in Earth science, as mandated by the Space Act, is to ... conduct aeronautical and space activities so as to contribute materially to ...the expansion of human knowledge of the Earth and of phenomena in the atmosphere and space. Therefore NASAs role is unique and highly complements those of other U.S. Federal agencies (such as the National Oceanic and Atmospheric Administration, National Science Foundation, U.S. Geological Survey, and Environmental Protection Agency) by continually advancing Earth system science from space, creating new remote sensing capabilities, and enhancing the operational capabilities of other agencies and collaborating with them to advance national Earth science goals. Continuous global observations of variability and change are required to reveal natural variability and the forces involved, the nature of the underlying processes and how these are coupled within the Earth system. NASAs Earth Science Division (ESD) provides these observations through its orbital and suborbital Flight Programs. In the decade 2007-2016, ESD will develop and demonstrate new sensors and interacting constellations of satellites to address critical science questions and enable advances in the U.S. operational capabilities. These and currently operating missions technical and programmatic details and status will be presented.

6744A-27, Session 5
A high-spectral resolution solid state infrared spectrometer for atmospheric air quality measurement
A. E. Roche, J. B. Kumer, J. L. Mergenthaler, R. L. Rairden, Lockheed Martin Advanced Technology Ctr. (USA) Spectrometers, in which a grating is coupled with a two dimensional detector array to provide high resolution spectra without the need for spectral scan mechanisms can be designed in compact, low mass, rugged, configurations. The approach is particularly well suited to long-term stable operation in space borne applications. We are pursuing the use of this technology for space borne tropospheric air quality monitoring, targeting high spectral resolution atmospheric solar reflective and thermal emission spectroscopy in the wavelength range 2 to 5 microns. This is a region where key tropospheric pollutant and greenhouse gases such as ozone, carbon monoxide, carbon dioxide, methane, formaldehyde and water vapor, have strong spectral features. The relatively short wavelengths allow for the use of welldeveloped detector technology, and passive radiative cooling. With sufficient resolving power, sensitivity, and judicious combination of spectra in this range, good information on tropospheric vertical distributions, including boundary layer data, can be obtained.

6744A-25, Session 5
OSTM (Ocean Surface Topography Mission), moving ocean altimetry towards an operational climate measurement
P. V. Vaze, Jet Propulsion Lab. (USA); J. Perbos, Ctr. National dtudes Spatiales (France); M. A. Abid, S. Brown, A. R. Dorsey, A. Kitiyakara, Jet Propulsion Lab. (USA) OSTM (Ocean Surface Topography Mission) will provide continuity of ocean topography measurements that began with TOPEX/Poseidon and is currently being carried out by Jason1. Measurements made by the three missions will allow scientists to better understand ocean circulation, climate change processes, and sea level rise on a multi-decadal scale. OSTM continues the long tradition of international collaboration in ocean altimetry between the United States (NASA) and Europe (CNES) with the addition of the respective operational weather and climate agencies, the National Oceanic and Atmospheric Administration (NOAA) and the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT). The OSTM satellite, Jason-2, will carry an instrument suite similar to the past missions such as an Altimeter, Radiometer, and a Precision Orbit Determination

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

71

Conference 6744A: Sensors, Systems, and Next-generation Satellites


We describe the development of a laboratory prototype of such a spectrometer, focused on the measurement of carbon monoxide spectra near 4.6 micron. The design uses a cryogenically cooled grating and associated optical train, coupled with a cooled 1024 x 1024 pixel HgCdTe array. We will present both laboratory absorption spectra and zenithlooking air emission spectra of CO to demonstrate the spectral, radiometric and spatial characteristics of the instrument. We will briefly discuss the implications for a space borne nadirlooking application.

6744A-31, Session 6
Dynamic range optimisation of CMOS image sensors dedicated to space applications
P. Martin-Gonthier, P. Magnan, F. Corbiere, L. Boucher, M. Estribeau, N. Huger, C. Engel, Ecole Nationale Suprieure de lAronautique et de lEspace (France) Nowadays, CMOS image sensors are widely considered for space applications. Although their performances have been significantly enhanced with the use of CIS (CMOS Image Sensor) processes in term of dark current, quantum efficiency and conversion gain. Dynamic Range (DR), defined as (maximum usable output linear swing)/(noise in dark) remains an important parameter for a lot of applications. Most of the dynamic range limitation of CMOS image sensors comes from the pixel. During programs in collaboration with EADS Astrium, SUPAERO/CIMI laboratory have studied different ways to improve it and test structures have been developed to perform analysis and characterisation. A first way to improve dynamic range will be described, consisting in improving the voltage swing at the pixel output. In collaboration with the foundry, process modifications are made to optimise implant settings and enhance threshold voltage and body effect of MOS transistors in the pixel. Test vehicles and process modifications made to improve voltage swing will be depicted. We have demonstrated a voltage swing improvement more than 30%. A second way to improve dynamic range is to reduce readout noise. In the common photodiode pixel, pixel with 3 MOS transistors, noise comes from reset phase (SQR[kT/CPH] for hard reset or close to SQR[kT/(2*CPH)] for soft reset with CPH : photodiode capacitance). New readout architecture have been developed to perform a correlated double sampling readout. Details on this readout architecture will be given and strong readout noise reduction will be demonstrated by measurements performed on our test vehicle. A third way to improve dynamic range is to control conversion gain value for different pixel. Indeed, in 3 TMOS pixel structure, dynamic range is related of conversion gain through reset noise which is dependant of photodiode capacitance. Decrease and increase of conversion gain for a given pixel pitch (13m) have been performed with different design techniques. A good control of the conversion gain will be demonstrated with variation in the range of 0.2 to 5 of initial conversion gain at a given pixel pitch (13m). In conclusion of the talk, a review of improvements about voltage swing, conversion gain and noise level is made to understand impact on dynamic range.

6744A-28, Session 5
Global precipitation measurement preliminary design
S. P. Neeck, R. K. Kakar, NASA Headquarters (USA); A. A. Azarbarzin, A. Y. Hou, NASA Goddard Space Flight Ctr. (USA) The overarching Earth science mission objective of the Global Precipitation Measurement (GPM) mission is to develop a scientific understanding of the Earth system and its response to natural and human-induced changes. This will enable improved prediction of climate, weather, and natural hazards for present and future generations. The specific scientific objectives of GPM are advancing: Precipitation Measurement through combined use of active and passive remote-sensing techniques, Water/Energy Cycle Variability through improved knowledge of the global water/energy cycle and fresh water availability, Climate Prediction through better understanding of surface water fluxes, soil moisture storage, cloud/precipitation microphysics and latent heat release, Weather Prediction through improved numerical weather prediction (NWP) skills from more accurate and frequent measurements of instantaneous rain rates with better error characterizations and improved assimilation methods, Hydrometeorological Prediction through better temporal sampling and spatial coverage of high-resolution precipitation measurements and innovative hydro-meteorological modeling. GPM is a joint initiative with the Japan Aerospace Exploration Agency (JAXA) and other international partners and is the backbone of the Committee on Earth Observation Satellites (CEOS) Precipitation Constellation. It will unify and improve global precipitation measurements from a constellation of dedicated and operational active/passive microwave sensors. GPM is completing the Preliminary Design phase and is advancing towards launch in 2013 and 2014.

6744A-30, Session 6
COBRA monolithic CMOS image sensors family: linear, multilinear and 2D devices answering space applications
O. Saint-Pe, EADS Astrium (France) Since more than 10 years now, Astrium SAS and Suparo are jointly developing high performance CMOS image detectors for space applications. The continuous efforts spent have now enabled to achieve the development of a family of CMOS detectors able to address several space applications including LEO/GEO Earth observation and Solar System exploration. Based on a CMOS process optimised for imaging applications, these linear, multilinear and 2D image sensors are featuring excellent electro-optical performances. Thanks to the already developed toolbox, Astrium SAS and Suparo have now the capability to develop with low risk CMOS detectors for dedicated space applications. After a short introduction explaining the various architectures of COBRA CMOS image sensors, the authors present data related to their operation, performances and space qualification. The talk also emphasizes the development of test benches and associated procedures to improve accuracy and efficiency for extensive detectors characterisation.

6744A-32, Session 6
Active pixel sensors: the sensor of choice for future space applications?
J. Leijtens, TNO (Netherlands) It is generally known that active pixel sensors (APS) have a number of advantages over CCD detectors if it comes to cost for mass production, power consumption and ease of integration. Nevertheless most space applications still use CCD detectors because they tend to give better performance and have more heritage. To this respect a change may be at hand with the advent of deep sub-micron processed APS imagers.(< 0.25 micron feature size) Measurements performed on test structures at the University of Delft have shown that the imagers are very radiation tolerant even if made in a standard process without the use of special design rules. Furthermore it was shown that the 1/f noise associated with deep sub-micron imagers is reduced as compared to previous generations APS imagers due to the improved quality of the gate oxides Considering that end of life performance will have to be guaranteed, limited budget for adding shielding metal will be available for most applications and lower power operations is always seen as a positive characteristic in space applications, deep sub-micron APS imagers seem to have a number of advantages over CCDs that will probably cause them to replace CCDs in those applications where radiation tolerance

72

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6744A: Sensors, Systems, and Next-generation Satellites


and low power operation are important. Autors: J.Leijtens, A.Theuwissen, P.Rao en X.Wang ECSIM can intercept and classify information and error messages from the simulations. A database is maintained with all the information generated by the system. It is possible to add third-party algorithms or tools to convert, analyse and visualize data extracted from generated products.

6744A-93, Session 6
Infrared sensors for Earth observation missions
A. P. Ashcroft, SELEX Sensors and Airborne Systems Ltd. (United Kingdom) SELEX-S&AS is developing a family of infrared sensors for earth observation missions. The spectral bands cover short-wave infrared (SWIR) channels from around 1.4m to long-wave infrared (LWIR) channels up to 12m. The detectors are designed for instruments with a spectral resolution of approximately 15nm in the SWIR band to 1m in the LWIR band. Our MCT technology has enabled a sensor array design that can satisfy the requirements of all of the SWIR and mediumwave infrared (MWIR) bands with near-identical arrays. This is made possible by the combination of a set of existing technologies that together enable a high degree of flexibility in the pixel geometry, sensitivity, and photocurrent integration capacity. The solution employs a photodiode array under the control of a readout integrated circuit (ROIC). The ROIC allows flexible geometries and in-pixel redundancy to maximise operability and reliability, by combining the photocurrent from a number of photodiodes into a single pixel. Defective or inoperable diodes (or sub-pixels) can be deselected with tolerable impact on the overall pixel performance. The arrays will be fabricated using the loophole process in MCT grown by liquid-phase epitaxy (LPE). These arrays are inherently robust, offer high quantum efficiencies. The use of loophole arrays also offers access to SELEXs avalanche photodiode (APD) technology, allowing low-noise, highly uniform gain at the pixel level where photon flux is very low.

6744A-71, Poster Session


Analysis of the relation between compression method and for performance enhancement of multi-spectral camera (MSC) image data
S. Yong, Korea Aerospace Research Institute (South Korea); S. Ra, Chungnam National Univ. (South Korea) Multi-Spectral Camera(MSC) is a payload on the KOMPSAT-2 satellite to perform the earth remote sensing. The instrument images the earth using a push-broom motion with a swath width of 15 km and a ground sample distance (GSD) of 1 m over the entire field of view (FOV) at altitude 685 Km. The instrument is designed to have an on-orbit operation duty cycle of 20% over the mission lifetime of 3 years with the functions of programmable gain/ offset and on-board image data compression/storage. The compression method on KOMPSAT-2 MSC was selected and used to match MSC input image data with output data. The MSC performance was analyzed and enhanced on KGS(KOMPSAT Ground Station) during LEOP and Cal./ Val.(Calibration and Validation) Phase. In this paper, the sequence of MSC cal./val. are described. The influence on image data due to compression algorithm and relation between compression method and performance enhancement algorithm are analyzed and shown.

6744A-72, Poster Session


An approach for retrieval of atmospheric trace gases CO2, CH4 and CO from the future Canadian micro earth observation satellite (MEOS)
A. P. Trishchenko, K. V. Khlopenkov, S. Wang, Canada Ctr. for Remote Sensing (Canada); R. V. Kruzelecky, W. Jamroz, MPB Communications Inc. (Canada); G. Kroupnik, Canadian Space Agency (Canada) Satellite retrievals are broadly recognized as critically important tool for mapping spatio-temporal distribution of trace gases. Among all trace gases, the carbon dioxide and methane provide the largest contribution to the climate radiative forcing and together with carbon monoxide also to the global carbon budget. New Micro Earth Observation Satellite (MEOS) mission is proposed to obtain information about these gases along with some other missions objectives related to studying cloud and aerosol interactions. The basic set of instruments for MEOS is currently under development with funding support from the Canadian Space Agency. The miniature suit of instruments is being designed to make measurements with reduced spectral resolution (1.2nm) over wide NIR range 1.22um to 2.45um and with high spectral resolution (0.03nm) for three selected regions: oxygen A-band, 1.5um-1.7um band and 2.2um-2.4um band. It is also planned to supplement the spectrometer measurements with high spatial resolution imager for detailed characterization of cloud and surface albedo distribution within spectrometer field of view. Paper will describe the approach for cloud/clearsky identification and column retrievals of above trace gases using combination of coarse and high-resolution spectral data. The combination of high and coarse resolution spectral data is beneficial for better characterization of surface spectral albedo and aerosol effects. Oxygen A-band path length will be used for normalization of trace gas retrievals, which improves stability of results and makes them less sensitive to uncertainties of surface elevation and total pressure.

6744A-69, Poster Session


ECSIM: the simulator framework for EarthCARE
R. Voors, Koninklijk Nederlands Meteorologisch Instituut (Netherlands); D. Donovan, Koninklijk Nederlands Meteorologisch Instituut (USA); J. Acarreta, R. Moyano, F. Pirondini, J. Ramos, DEIMOS Space S.L. (Spain); M. Eisinger, R. Franco, D. Lajas, T. Wehr, European Space Agency (Netherlands) In 2013 an important ESA Core Explorer Mission, EarthCARE is scheduled to be launched. EarthCARE, (Earth, Clouds, Aerosol and Radiation Explorer) will comprise two active (a cloud-profiling radar (CPR) and an high spectral resolution atmospheric lidar (ATLID)) and two passive (a Multi-spectral imager (MSI) and a Broad-Band Radiometer (BBR)) instruments. With these, EarthCARE will enable cloud and aerosol properties retrievals consistent with a Top-of-Atmospheric (TOA) flux accuracy of 10 Wm-2. This will be achieved by simultaneously probing the atmosphere vertically with the active instruments in synergy with the passive instruments. It is important to study and understand the factors that could hinder the proper development of the mission and the realization of its scientific goals and to do so well before launch,. With that purpose in mind, a simulator for EarthCARE (ECSIM) has been developed. ECSIM relies strongly upon a previous prototype developed by ESA / KNMI where a combination of forward and retrieval models (full End-to-End capabilities) have been included. In order to make this tool more useful within the scientific and engineering communities, the prototype simulator has been embedded into a completely reorganised architecture intended to improve a number of aspects: Complex algorithms have been enclosed into logical entities: models. Models are connected in a logical sequence with well-defined interfaces. Users can customize almost every models parameter values with XML files. Model outputs are well documented and stored in easy to access NetCDF files. Complex simulations can be built up with a few mouse clicks. Users can run lengthy simulations in batch iterating through different parameter values.

6744A-73, Poster Session


Spatial synchronization in space-borne/airborne hybrid bi-static SAR
P. Zhou, Y. Pi, Z. Han, L. Fan, Univ. of Electronic Science and Technology of China (China)

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

73

Conference 6744A: Sensors, Systems, and Next-generation Satellites


In determining the locations of illuminators and receivers, it has wide choices in bi-static SAR. A subclass raised in recent years, a hybrid SAR system using a spaceborne transmitter and a receiver mounted on an aircraft, combines large illuminated area of the spaceborne SAR with high resolution and flexibility of the airborne SAR. Nevertheless, the development of the hybrid system is still at an early stage for its great difficulty. One of the challenges is the space-time synchronization of the antenna footprints due to the high difference between the platforms velocities. A scheme has been proposed by U. Gebhardt et al. in EUSAR 2006, the key of which is to slow down the satellite antennas footprint in azimuth so that the receiver can follow the illuminator within a prolonged time period. However, the scheme, according to the authors viewpoint, has two insufficiencies. First, linear trajectory of the satellite, flat earth geometry and parallel trajectories between the platforms are assumed in their paper to decrease the complexity, which degrades the practicability of the algorithm. Second, the imaging algorithm will be very complex owing to the continuously changing in pointing directions of antennas. Two schemes have been proposed and validated in this paper to solve the two imperfections respectively. The first scheme is a modification of Gebhardts scheme, whose mathematical model for antenna steering is modified. The second scheme is a new scheme, whose key is to illuminate wide beam by reducing satellite antennas size. The loss of SNR is compensated by the airborne receivers much shorter distance compared to mono-static spaceborne SAR. strong 9.6 micron infrared ozone band at high spectral resolution, while the concept and technology also have applicability toward measurement of other trace species and other applications. The latest results from laboratory testing and characterization of enabling subsystems and the overall instrument system will be reported, with an emphasis placed on testing performed to evaluate system-level radiometric, spatial, and spectral measurement fidelity.

6744A-33, Session 7
Development of a multichannel 10240-pixel TDI CCD
P. Vu, S. W. Mims, Fairchild Imaging (USA) In this paper, we review the design, modeling, and characteristics of a multichannel time-delay-integration (TDI) CCD image sensor recently developed at Fairchild Imaging which is well suited for low light level multispectral imaging requiring a wide field of view. We describe the trade-off between quantum efficiency and MTF, and present test results of the detector performance. The detector consists of two closely overlapped 5248-pixel long TDI imagers, each featuring 3 individual CCD imaging arrays colocated on the same chip, resulting in each of the 3 operating channels having 10240 columns of 13 micron pixels. The device features up to 128 TDI rows selectable in blocks of 2, 4, 8, 16, 32, 64 and 128 for optimal signal to noise performance under a wide range of light levels. The pixels have 100 percent fill factor, and have been optimized for high charge handling capacity. The vertical shift registers operate using 3-phase clocks, and the horizontal CCD register is 4-phase. The detector has 8 output ports providing 27 MHz data rate per output. Preliminary measurements indicate that the charge handling capacity exceeds 700 ke-, the QE is greater than 45% at 700nm, and the MTF at Nyquist is better than 48% at 700nm.

6744A-74, Poster Session


High precision spaceborne accelerometer for measuring extra low liner accelerations
P. O. Demyanenko, Y. V. Savenko, Kiev Polytechnic Univ. (Ukraine) Solutions of several fundamental and applied tasks require creating high sensitive precise accelerometers with sensitivity in order of 10e-13 g. These several tasks are following: inertial navigation of the Earth and Space; gravimetry nearby the Earth and into Space; geology; geophysics; seismology etc. Accelerometers (gravimeters and gradientmeters) with required sensitivity are not available now. The best accelerometers in the world have sensitivity worth on 4-5 orders. There is developed a new class of fiber-optical sensors (FOS) with light pulse modulation. These sensors have super high threshold sensitivity and wide dynamic range, and can be used as a base for creating of measurement units of physical values as 3D superhigh sensitive precise accelerometers of linear accelerations. The principle of operation of the FOS is organically combined with a digital signal processing. It allows decreasing hardware of the accelerometer due to using a usual air-borne or spaceborne computer; correcting the influence of natural, design, technological drawbacks of FOS on measured results; neutralising the influence of extraordinary situations available during using of FOS; decreasing the influence of internal and external destabilising factors (as for FOS), such as oscillation of environment temperature, instability of pendulum cycle frequency of sensitive element of the accelerometer etc. It has been proposed the 3D accelerometer of extra low acceleration that is based on PFOS. Performances (calculated): Threshold of sensitivity - 10e-9 - 10e-13 g; Dynamic range 10e7 - 10e9; Frequency range - 0 - 10 Hz; Mass - 50 grams; Size: length - 120 mm; diameter - 20 mm

6744A-34, Session 7
Next Ofeq focal plane array
E. R. Yacoby, Y. Cohen, Elbit Systems Electro-Optics ELOP Ltd. (Israel) For the next generation of Ofeq satellites, a new focal plane array has been designed. This paper describes the new detector design and its modes of operation. The anastigmat three-mirror telescope (ATMT) design does not allow any refractive elements in the focal plane butting concept, hence a new mechanical butting solution was applied and qualified, and this will be described. The mechanical design also takes into account requirements for Line Of Sight (LOS) stability. In order to reduce analog noise, the detector output will connected directly to the focal plane electronics board using a qualified connector. This technique simplifies the detector replacement process, in case of detector failure during integration. The ATMT design results in an off-axis detector position and due to the wide field of view (FOV) of the camera, distortion effects must be taken into account. Especially in the case of 96 row TDI detectors, the impact on system Modulation Transfer Function (MTF), both in the scan direction and along the detector line direction, will be described. In order to reduce these effects, a curved focal plane is needed; hence, special care was taken in the design of the detector butting angles. The residual MTF contribution of the curved focal plane is presented.

6744A-35, Session 7
Versatile 1024x256 SWIR-HgCdTe hyperspectral imaging sensor for applications with low photon fluxes and high framerates
M. Haiml, H. Bitterlich, A. Erni, M. Finck, K. C. Hofmann, H. Lutz, M. Mai, H. Nothaft, I. Rhlich, J. C. Wendler, R. Wollrab, J. Ziegler, AIM Infrarot-Module GmbH (Germany) Many current remote sensing programs rely on hyperspectral imaging systems. The fast growing market for spectrally and spatially resolved (3D) data require high bandwidth data storage and processing capabilities at system level and optimized sensor units in the focal plane. Particularly interesting is the short-wave infrared (SWIR) region from 0.9 to 2.5 m. AIM has developed a dedicated hyperspectral imaging (HSI) sensor called GENSIS (generic short-wave infrared sensor). This

6744A-75, Poster Session


Overview of laboratory testing results for an imaging Fabry-Perot interferometer
A. M. Larar, W. B. Cook, M. A. Flood, NASA Langley Research Ctr. (USA); E. Burcher, Swales Aerospace (USA); C. Boyer, NASA Langley Research Ctr. (USA); J. J. Puschell, Raytheon Space and Airborne Systems (USA) An airborne imaging Fabry-Perot Interferometer (FPI) system was developed within NASAs Instrument Incubator Program (IIP) to mitigate risk associated with implementation of such a device in future space-based atmospheric remote sensing missions. This system is focused on observing tropospheric ozone through measuring a narrow spectral interval within the

74

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6744A: Sensors, Systems, and Next-generation Satellites


1024 x 256 sensor module with a cut-off wavelength of 2.5 m is based on a 24 x 32 m pitch grid with an optically active area of 24.6 x 8.2 mm2. The read-out circuit (ROIC) is optimized for HSI applications, using a CTIA input stage for low photon fluxes from 100 photons per integration time (S/N = 1) up to 2.4e+6 photoelectrons (full well capacity). Depending on the spectral signature of interest, each spectral line can be set to low or high gain or can be deselected completely. The adjustable gain results in photon-noise or kTC-noise limited operation over the whole dynamic range. The line deselection function allows either for frame rates of several tens of kHz for small regions of interest or to reduce the video data rate already on-chip. The maximum rate for the read out of a full frame of 1024 x 256 pixel is 245 Hz. Operation mode can be set to stare while scan (SWS) and integrate then read (ITR), both in snap-shot mode. The integration time can be varied from 1 s up to 6.5 ms for 10 MHz pixel clock. For space applications at FPA temperatures in the range form 100 K to 220 K, the GENSIS module is integrated in a cooler dewar assembly, with a specially developed compact pulse-tube flexure-bearing cryo-cooler with a long lifetime and a low vibration level. We will give an overview of Gensis functionality and present the latest electro-optical performance data.

6744A-38, Session 7
Development of a long wave infrared detector for the SGLI instrument
A. Dariel, P. Chorier, N. Reeb, B. Terrier, M. Vuillermet, P. M. Tribolet, Sofradir (France) The Japanese Aerospace Exploration Agency (JAXA) has the plan of the Global Change Observation Mission (GCOM) for monitoring of global environmental change. SGLI (Second Generation Global Imager) on board GCOM-C (Climate), which is one satellite of this mission, is an optical sensor from NearUV to thermal IR. SGLI will provide high accuracy measurements of the atmosphere (aerosol, cloud ...), the cryosphere (glaciers, snow, sea ice ...), the biomass and the Earth temperature (sea and land). As part of SGLI, a Long Wave IR Detector (LWIRD) sensitive up to about 13 m is necessary. NEC Toshiba Space, prime contractor for the Second Generation Global Imager (SGLI), has selected SOFRADIR for its heritage in space projects and MCT detectors to develop this detector. Sofradir is indeed a major supplier for European space industry based on the use of a qualified Mercury Cadmium Telluride (MCT) technology for detectors from 0.8 to 15 m. This mature and reproducible technology has been used for 15 years to produce thousands of LWIR detectors with cut-off wavelengths between 9 and 12 m. This detector includes two detection circuits for detection at 10.8 m and 12.0 m, hybridized on a single CMOS readout circuit. Each detection circuit is made of 20x2 square pixels of 140 m. In order to optimize the overall performance, each pixel is made of 5x5 square sub-pixels of 28 m and the readout circuit enables sub-pixel deselection. The composition of the MCT material and the photovoltaic technology are adapted to maximize response for the requested bandwidths: cut-off wavelengths of the 2 detection circuits are 12.5 and 13.5 m at 55K (in order to get a low dark current level). This detector is packaged into a sealed housing for integration into a Dewar at 55K. In this paper, after introduction of the main technical requirements, the design of this space detector is described, including the main trade-offs performed regarding performance optimization. Then first electro-optical results are presented with perspectives for this space detector development.

6744A-36, Session 7
Monolithically-integrated near-infrared InGaAs and mid-infrared QWIP detector array
S. V. Bandara, S. D. Gunapala, J. K. Liu, Jet Propulsion Lab. (USA) Quantum well infrared photodetector (QWIP) technology has shown remarkable success by realizing large-format focal plane arrays in both broad-bands and in multi-bands. The spectral response of these detectors based on the III-V material system are tailorable within the mid- and long-wavelength infrared bands (~3-50 m). Recently, we have extended this wavelength limit by monolithically integrating a near-infrared (visible - 2.1 m) p-i-n photodiode with a mid-infrared (3-5m) (QWIP). This multi-band detector involves both intersubband and interband transitions in III-V semiconductor layer structures. Each detector stack absorbs photons within the specified wavelength band while allowing the transmission of photons in other spectral bands, thus efficiently permitting multi-band detection. Flexibility in many design parameters of these detectors allows for tuning and tailoring the spectral shape according to application requirements, specifically for spectral imaging instruments. In this presentation, we will discuss the recent results, focal plane array design issues and optimization of detector device structures for each application.

6744A-39, Session 7
Radiation tolerance analysis of IASI infrared detectors in the 3.4-15.5 m spectral range
F. Bernard, H. Geoffray, D. Blumstein, O. Gilard, Ctr. National dtudes Spatiales (France); D. Coppens, Noveltis SA (France); F. Cayla, SISCLE (France) IASI (Infrared Atmospheric Sounding Interferometer) is an infrared atmospheric sounder. The IASI instrument is currently operating on the Metop-A satellite (launched in October 2006). The core of the instrument is composed of a Fourier transform infrared spectrometer. The detection chain of the spectrometer includes 3 bands to cover the 3.4 to 15.5 m spectral range. For each band, the IR detection is made by a 2 x 2 pixels array operating at ~93K. This paper presents an analysis of the radiation tolerance of the infrared detectors for each band. On ground, radiation tests have been performed to address sensitivity to gamma-rays and protons radiations. A special care has been taken to keep the detectors at cold temperature during tests. Performance evolution (responsivity, relative spectral response, noise,...) before and after radiations are given. First in orbit impacts of the radiations are also reviewed.

6744A-37, Session 7
Focal plane arrays from UV up to VLWIR
E. M. Costard, J. Truffer, O. Huet, L. Dua, A. Nedelcu, J. Robo, X. Marcadet, N. Brire de lIsle, Thales Research & Technology (France); H. Facoetti, Thomson-CSF (France); P. F. Bois, Thales Research & Technology (France) Since 2002, the THALES Group has been manufacturing sensitive arrays using QWIP technology based on GaAs and related III-V compounds, at THALES Research and Technology Laboratory. In the past researchers claimed many advantages of QWIPs. Uniformity was one of these and has been the key parameter for the production to start. Another widely claimed advantage for QWIPs was the so-called band-gap engineering and versatility of the III-V processing allowing the custom design of quantum structures to fulfil the requirements of specific applications such as very long wavelength (VLWIR) or multispectral detection. In this presentation, we present the performances of both our first 384x288, 25 m pitch, MWIR (35m) / LWIR (8-9 m) dual-band FPAs, and the current status of QWIPs for MWIR (< 5m) and VLWIR (\>15m) arrays. As the QWIP technology cannot cover the full electromagnetic spectrum, we develop other semiconductor compounds for SWIR and UV applications. We present here the status of our first FPA realization in UV with GaN alloy, and at 1.5m with InGaAs photodiodes.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

75

Conference 6744A: Sensors, Systems, and Next-generation Satellites

6744A-40, Session 7
Latest pulse tube coolers developments of air liquide for space
J. Buquet, T. Trollier, J. Tanchon, G. Aigouy, A. Ravex, P. Crespi, Air Liquide (France) Thanks to important development efforts completed and partial ESA funding, AL/DTA is now in position to propose two Pulse Tube cooler systems in the 40-80K temperature range for coming Earth Observation missions such as Metesoat Third Generation, Sentinel 3, etc... The two pulse tube coolers thermomechanical units are yet qualified against environmental constraints. The Cooler Drive Electronics is also an important aspect specifically regarding the active control of the cooler thermomechanical unit during the launch phase, the active reduction of the vibrations induced by the compressor and of course the electrical interfaces with the compressor. This paper details the presentation of the two Pulse Tube Coolers together with the Cooler Drive Electronics aspects

technology. As a matter of fact, Sofradir is involved in several projects for future space missions (SPIRALE, Bepi Colombo, MTG, SGLI...). This paper proposes an overview of Sofradir technology capabilities and experience for design of custom space detectors. In particular this paper presents latest developments for space applications with new results in visible, long wavelength and space qualification of infrared detectors.

6744A-94, Session 7
AlGaN-based focal plane arrays for selective UV imaging at 310nm and 280nm and route toward deep UV imaging
J. Reverchon, J. Robo, J. Truffer, Thales Research & Technology (France); J. Duboz, Consultant (France) The fast development of nitrides has given the opportunity to investigate AlGaN as a material for ultraviolet detection. Such camera present an intrinsic spectral selectivity and an extremely low dark current at room temperature. It can compete with technologies based on photocathodes, MCP intensifiers, back thinned CCD or hybrid CMOS focal plane arrays for low flux measurements. AlGaN based cameras allow UV imaging without filters or with simplified ones. Few results on camera have been shown in the last years, but the ultimate performances of AlGaN photodiodes couldnt be achieved due to: parasitic illumination of multiplexers, responsivity of p layers in p-I-n structures, or use of cooled readout circuit; preventing large development of this technology. We will present results on focal plane array of 320x256 pixels with a pitch of 30m for which Schottky photodiodes are multiplexed with a readout circuit protected by black matrix at room temperature. Theses focal plane present a peak reponsivity around 280nm and 310nm with a rejection of visible light of four decades only limited by internal photoemission in contact. Then we will show the capability to outdoor measurements of plasma monitoring. The noise figure is mainly due to readout noise of the multiplexer and we will investigate the ultimate capabilities of Schottky diodes or Metal-Semiconductor-Metal (MSM) technologies to detect extremely low signal. Furthermore, we will consider deep UV measurements on single pixels MSM at 32nm in a front side illumination configuration. Finally, we will define technology process allowing backside illumination and deep UV imaging.

6744A-41, Session 7
Large format delta-doped silicon imagers with high QE in the UV/optical/NIR for remote observations
S. Nikzad, M. E. Hoenk, J. Blacksberg, T. J. Jones, Jet Propulsion Lab. (USA); S. E. Holland, Lawrence Berkeley National Lab. (USA) There is an increasing need for large format imaging and detector arrays. For example, future NASA missions will require large format high performance imaging arrays in a mosaic or monolithic focal plane to achieve wide field of view, highresolution imaging (and spectroscopy). The basic array design for each application might require CMOS or CCDs unique features. One of crucial keys to achieve highest performance possible in silicon imaging arrays is back illumination. We have developed complete end-to-end post fabrication processes at JPL that have allowed us to produce high performance back illuminated imaging arrays of widely varying designs including n-channel CCDs, p-channel CCDs, and CMOS arrays. We have developed a low temperature processes for B and Sb delta-doping of silicon which can be used to form a thin (< 3 nm) back-surface contact for fully processed back-illuminated, high purity PIN arrays, n-channel CCDs and CMOS arrays, or p-channel CCDs. Non-equilibrium growth by Molecular Beam Epitaxy (MBE) is used to achieve very high dopant incorporation in a thin, surface-confined layer. Optimization of this process has enabled the growth of a thin highly conductive back surface electrode on p-channel CCDs and PIN arrays. The temperature is kept below 450 C throughout the entire process, which is required for compatibility with fully processed and functional Al-metallized devices. This approach enables full depletion of thick, high purity silicon detectors while achieving high quantum efficiency in UV/Optical/ NIR and low dark current. We will present our latest results on delta-doping of large-format high purity arrays, including QE, imaging, dark current, and uniformity.

6744A-42, Session 8
MIBS: past, present and future
J. Leijtens, A. J. Court, E. van der Meche, B. de Goeij, TNO (Netherlands) The microbolometer spectrometer breadboard MIBS is a prism spectrometer that uses an uncooled microbolometer detector array and has been designed for the ESA EarthCARE mission. In order to demonstrate its feasibility a breadboard has been build, and tests have been performed that show good correlation between predicted and achieved results. Although application for EarthCARE has become uncertain due to geodistribution issues, it is feld that this instrument (which is small enough to give grown up performance to a micro satellite) has a lot of application potential for applications like weather forecasting and forest fire detection. The presentation will elaborate on performance predicted, measurements performed, results achieved and future applications.

6744A-76, Session 7
Latest development for space applications at Sofradir
P. Chorier, P. M. Tribolet, Sofradir (France) Sofradir started to work in the field of space applications and especially in the earth observation domain in the beginning of the 1990th. Thanks to the work done with the support of the French Ministry of Defense and the European Space Agency, Sofradir has acquired a large know-how and became a major supplier for European space industry. Nowadays, Sofradir technologies offer possibilities to develop a large panel of high reliable detectors like long linear arrays or two dimensional arrays covering bandwidth from visible to 15 m based on qualified Mercury Cadmium Telluride (MCT)

6744A-43, Session 8
Tests and qualification activities for the GOSAT interferometer flight model
L. E. Lvesque, M. A. Soucy, L. M. Moreau, F. Doyon, D. Duquette, ABB Inc. (Canada); J. Tanii, NEC TOSHIBA Space Systems, Ltd. (Japan) The Greenhouse gases Observing SATellite (GOSAT) is designed to monitor the global distribution of carbon dioxide (CO2) from orbit. It is a joint project of Japan Aerospace Exploration Agency, the Ministry of Environment (MOE), and the National Institute for Environmental Studies (NIES). JAXA is responsible for the satellite and instrument development, MOE is involved in the instrument development, and NIES is responsible for the satellite data retrieval. It is scheduled to be launched in 2008. As existing

76

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6744A: Sensors, Systems, and Next-generation Satellites


ground monitoring stations are limited and still unevenly distributed, the satellite observation has advantages of global and frequent observations. The objective of the mission is in response to COP3 (Kyoto Protocol): Observation of Green House Gases (GHGs) including CO2 with 1% relative accuracy in sub-continental spatial resolution and to identify the GHGs source and sink from the data obtained by GOSAT in conjunction with the data from the ground instruments, with simulated models. In order to detect the CO2 variation of boundary layers, the technique to measure the column density and the retrieval algorithm to remove cloud and aerosol contamination are investigated. The simultaneous observation of methane (CH4), which is the second largest contribution molecule, is studied. A Fourier transform spectrometer with high optical throughput and spectral resolution has been developed for the GOSAT mission. This paper presents an overview of the tests and qualification campaigns of the TANSO interferometer flight model at ABB Bomem as well as the key interferometer performances.

6744A-46, Session 8
A small imaging spectrometer for a microsatellite STSAT-3
J. H. Lee, Kongju National Univ. (South Korea); T. S. Chang, Korea Advanced Institute of Science and Technology (South Korea); K. I. Kang, Korea Advanced Institute of Science And Technology (South Korea); S. W. Rhee, Korea Aerospace Research Institute (South Korea) SATSAT-3, which is a ~120kg micro-satellite, is the third experimental microsatellite of STSAT series designated in the Long-Term Plan for Koreas Space Development by the Ministry of Science and Technology of Korea. The STSAT-3 satellite is initiated at the May of 2007 and will be launched into a lower earth orbit (~ 600km) at 2010. This paper presents a brief introduction of STSAT-3 program and then introduces its secondary payload i.e. a small and light (~5kg) imaging spectrometer. The payload will be used for environmental monitoring such as yellow dust cloud blown from China to South Korea.

6744A-44, Session 8
A generic interferometer design for the next generation weather sounders on GEO satellites
F. Grandmont, J. G. Giroux, M. A. Soucy, H. L. Buijs, ABB Inc. (Canada) A new generation of sensors with hyperspectral capability is being considered for future geosynchronous earth orbiting (GEO) platforms. The scientific interest for more spectral channels suggest that multi-bandpass filter approaches are replaced by those providing true spectral capabilities. Interferometer based sounders offer the possibility to combine large format 2D imaging and high spectral resolution into a single measurement and thus tremendously increase the amount of information available to improve weather predictions models. The so-called throughput or Felgett advantage that made Fourier Transform Spectrometry well known in the 70s is now exploited through what is now known as the multi-channel advantage; the ability to acquire simultaneous spectra on every pixels of a standard array detector. Starting from commonly accepted high level science requirements, a set of judicious instrument parameters are derived to better exploit the technique to the profit of scientific return while preventing secondary aspects to drive instrument costs substantially. A generic design developed specifically to address the weather sounding applications will be presented along with the trade space and rational behind design choices. The heritage from other past space instruments developed by ABB Analytical (formerly Bomem Inc.) onto which this concepts builds up will be highlighted. Experimental results on key instrument life driving parameter that raise the technology readiness level of the concept will also be discussed.

6744A-60, Session 8
Preliminary study on Earth observation sensor from geostationary Earth orbit
Y. Y. Yui, H. Imai, Y. Tange, T. Kimura, Japan Aerospace Exploration Agency (Japan) We have been studying the feasibility and possibility of earth observation sensors from geostationary earth orbit. The main purpose of the sensor is to monitor large scale natural disasters such as landslides, flood, volcano eruption, etc., with spatial resolution of 20m at nadir at optical wavelength (500nm). The sensor specifications and configurations will be shown and critical technical challenges will be discussed.

6744A-47, Session 9
Radiometric calibration status of Landsat-7 and Landsat-5
J. A. Barsi, Science Systems and Applications, Inc. (USA); B. L. Markham, NASA Goddard Space Flight Ctr. (USA) Launched in April 1999, Landsat-7 ETM+ continues to acquire global data, though in a reduced capacity. The Scan Line Corrector failure and the recent switch to Bumper Mode operations have affected the geometric integrity of the data, but the radiometry has been unaffected. The best of the three on-board calibrators for the reflective bands, the Full Aperture Solar Calibrator, has indicated slow changes in the ETM+, but this is believed to be due to contamination on the panel rather then instrument degradation. Regular observation of desert targets in the Saharan and Arabian deserts confirms the lack of change in the ETM+ reflective band response. The thermal band continues to be stable and well-calibrated since an offset error was corrected in late-2000. Launched in 1984, Landsat-5 TM also continues to acquire global data; though without the benefit of an on-board recorder, data can only be acquired where a ground station is within range. Though the satellite is running on multiple back-up systems, the data continues to fill gaps in the ETM+ archive. Historically, the calibration of the TM reflective bands has used an on-board calibration system with multiple lamps. The calibration procedure for the TM reflective bands was updated in 2003 based on the best estimate at the time, using only one of the three lamps. Since then, the Saharan desert sites have been used to test this calibration model. Problems were found with the lamp based model of up to 20% in some bands for some years. Using the Saharan data, a new model was developed and implemented in the US processing system in April 2007. The thermal band was found to have a calibration offset error of 0.68K at 300K based on vicarious calibration data between 1999 and 2006. The offset error was corrected in the US processing system on April 2007 for all data acquired since April 1999.

6744A-45, Session 8
A multispectral 10-metre resolution camera for earth observation
K. F. Middleton, Rutherford Appleton Lab. (United Kingdom) We present a 10 metre resolution camera designed for use in low-earth sun-synchronous orbit. Images are acquired over four spectral bands in the visible and near infrared using a pushbroom configuration and four linear CCD detectors. A refractive optical design, using radiation tolerant glasses, results in a compact, lightweight, yet high performance system, and the opto-mechanical structure is designed to maintain alignment and image quality under typical launch loads. High speed electronics allow simultaneous readout of all four detectors over spacewire links. We discuss the design and construction of the camera, and environmental and optical testing of the engineering model instrument. Image data from ground-based tests are presented.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

77

Conference 6744A: Sensors, Systems, and Next-generation Satellites

6744A-48, Session 9
An overview of Aqua MODIS after five years of on-orbit operation and calibration
W. L. Barnes, Univ. of Maryland/Baltimore (USA); X. Xiong, NASA Goddard Space Flight Ctr. (USA) Launched in May 2002, the Aqua MODIS has been successfully operated on-orbit for nearly five years, making continuous observations that have provided the science community and users worldwide with many high quality data products that have significantly contributed to studies of the Earths climate and environmental changes. MODIS observations are made via 36 spectral bands ranging from the visible (VIS) to long-wave infrared (LWIR), including 20 reflective solar bands (RSB) with wavelengths from 0.41 to 2.2 micrometers and 16 thermal emissive bands (TEB) from 3.7 to 14.4 micrometers, and three (nadir) spatial resolutions: 250m (2 bands), 500m (5 bands), and 1km (29 bands). Many of its operational activities (OA) have been designed and developed to support the sensors on-orbit calibration and characterization. The MODIS on-board calibrators, noticeably improved over those of its heritage sensors, include a solar diffuser (SD), a solar diffuser stability monitor (SDSM), a blackbody (BB), a spectro-radiometric calibration assembly (SRCA), and a space view (SV) port. The SD/SDSM system is used for RSB calibration on a regular basis varying from weekly to tri-weekly. The BB is operated continuously for TEB calibration. The instrument spectral (RSB) and spatial (TEB and RSB) characterizations are provided by the SRCA bi-monthly or quarterly. This paper presents an overview of Aqua MODIS five-year on-orbit operation and calibration with emphasis on the performance of its on-board calibrators and the their impact on the L1 data products and calibration quality. Results discussed in this paper include TEB and RSB detectors noise characterization, short-term stability and corrections to long-term response changes.

6744A-50, Session 9
On-orbit monitoring of MODIS thermal emissive bands response versus scan angle
X. Xiong, NASA Goddard Space Flight Ctr. (USA); A. Wu, Science Systems and Applications, Inc. (USA); W. L. Barnes, B. W. Guenther, Univ. of Maryland/Baltimore (USA) MODIS is a cross-track scanning radiometer that uses a twosided scan mirror to make Earth view (EV) observations over an angular range of +/-55 degrees relative to instrument nadir. The 16 thermal emissive bands (TEB) are calibrated on a scanby-scan basis using a quadratic algorithm that includes corrections to the sensors response versus scan angle (RVS). Currently there are two MODIS instruments on-orbit. The Terra MODIS was launched in December 1999. Its TEB RVS was not well characterized during instrument pre-launch testing. Because of this the Terra spacecraft executed a pair of pitch maneuvers which enabled a full on-orbit characterization of the MODIS TEB RVS. For Aqua MODIS, launched in May 2002, the TEB RVS was successfully measured pre-launch and, consequently, post-launch pitch maneuvers were not required. In this paper we describe a TEB RVS measurement approach developed using observations made during MODIS EV sector rotation (SR). This is a regularly-scheduled operation originally designed for the MODIS lunar observations. During this operation, the EV data collection is delayed (via flight software commands) such that the EV data is taken when the sensor is viewing the internal scan cavity. We will show that this approach can accurately determine TEB RVS differences between the two mirror sides and track TEB RVS changes on-orbit. Results from both Terra and Aqua MODIS and their potential applications will be presented and compared.

6744A-52, Session 9
The aerospace imaging interferometer ALISEO: further improvements of calibration methods and assessment of interferometer response
A. Barducci, Istituto di Fisica Applicata Nello Carrara (Italy); F. Castagnoli, Consiglio Nazionale delle Ricerche (Italy); D. Guzzi, P. Marcoionni, I. Pippi, Istituto di Fisica Applicata Nello Carrara (Italy) In recent years the Italian Space Agency partially supported the study of an aerospace imaging interferometer for Earth Observation (EO). The instrument, named ALISEO (Aerospace Leap-frog Imaging Stationary Interferometer for Earth Observation), belongs to the class of the stationary interferometers due to their simple (light) optical layout and the on-fly reconfiguration ability represent a promising architecture for future EO sensors. ALISEO is now being considered as the principal payload for a novel technological optical mission based on micro-satellite (MIOSat). This new mission is constituted by a micro-satellite equipped with advanced payloads, among which there are an extensible telescope, a high-resolution panchromatic camera, a MachZehnder MEMS punctual interferometer, and ALISEO. MIOSat is expected to provide the scientific community with the technological demonstration of this new remote sensing technology. ALISEO operates in the common-path Sagnac configuration, and it does not employ any moving part to generate phase delay between the two allowed ray paths. The sensor acquires target images modulated by a pattern of autocorrelation functions of the energy coming from each scene pixel, a fringe pattern that is fixed with respect to the instruments field of view. The complete interferogram of each target location is retrieved introducing relative source-observer motion, which allows any image pixels to be observed under different phase delays. In this paper recent laboratory measurements performed with ALISEO are described and discussed in the framework of the aforementioned MioSat mission. In order to calibrate the opticalpath-difference (OPD) axis of the raw interferograms, a set of measurements have been carried out by illuminating a double planar diffuser system with several coloured He-Ne lasers. Standard reflectance tiles together with diffusers doped with Holmium and Rare Earths have been used for validating the wavelength calibration of the instrument and proving the reliability of the reflectance retrieving procedure. Finally, we discuss an algorithm for raw data pre-processing aimed at retrieving the at-sensor radiance spectrum, including exploiting

6744A-49, Session 9
Applications and results of MODIS lunar observations
X. Xiong, NASA Goddard Space Flight Ctr. (USA); J. Sun, Science Systems and Applications, Inc. (USA); W. L. Barnes, Univ. of Maryland/Baltimore (USA) Both Terra and Aqua MODIS have been making periodic lunar observations since their launch in December 1999 and May 2002, respectively. The primary objective of the MODIS lunar observations is to monitor on-orbit response changes for the reflective solar bands (RSB) and to track long-term radiometric calibration stability. The Moon is an extremely stable surface reflectance reference and, together with the lunar radiometric model developed recently by the US Geological Survey (USGS), the applications of lunar observations for Earth remote sensing missions have been greatly enhanced and expanded. In particular, calibration inter-comparisons among sensors on different platforms and operated in different time frames have been enabled in the VIS/NIR/SWIR spectral regions. In this paper, we discuss various applications of MODIS lunar observations, such as monitoring the RSB radiometric stability and calibration consistency between the two sensors, evaluating the calibration differences among detectors within the same spectral band, and tracking the on-orbit changes of band-to-band registration (BBR). Other applications designed for the thermal emissive bands (TEB) and special sensor characterization purposes are also illustrated. Results from multi-year Terra and Aqua MODIS lunar observations demonstrate clearly that the two sensors have been calibrated consistently to within 1% for most RSB compared to the 2% radiometric calibration requirement. In addition the spatial characterization results derived from lunar observations are in good agreement with those derived from the instruments onboard calibrator. The applications and results of MODIS lunar observations presented in this paper provide useful references for other sensors that also make use of lunar surface observations.

78

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6744A: Sensors, Systems, and Next-generation Satellites


the issues of dark signal subtraction, spectral instrument response compensation, effects of vignetting, and Fourier backtransform algorithm. scattered light and describe an algorithm that has been developed to correct an instruments response to this scattered light. By characterizing an instrument and applying corrections for the scattered light, the measurement errors of the instrument are reduced by more than one order of magnitude. Examples are given to demonstrate the efficacy of the approach and implications for remote sensing instruments are discussed.

6744A-53, Session 10
Modeling spectral effects in Earth-observing satellite instruments
R. A. Barnes, Science Applications International Corp. (USA); J. J. Butler, NASA Goddard Space Flight Ctr. (USA) We present a simple, spreadsheet-based model to examine the effects of the spectral response functions of individual instrument bands on their measurements of top-of-theatmosphere radiances. The model uses spectral radiances at 1 nm resolution from the near ultraviolet to the shortwave infrared at wavelengths from 300 nm to 2500 nm, convolving them with the spectral responses of the bands to calculate band-average spectral radiances. For on-orbit calibration purposes, the model uses nominal solar irradiance and lunar albedo spectra to provide saturation, diffuser, and lunar radiances for the bands. For prelaunch calibration purposes, the model uses a 2850K Planck function, normalized to a maximum value of unity, to approximate the spectral shape from a laboratory integrating sphere source. For Earth-exiting radiances, the model uses nominal radiance spectra over a blue ocean, a desert, and a grassland. These spectra are provided with the effects of atmospheric trace gas absorption removed. In addition, the model includes a trace gas transmittance spectrum that can be modified as a function of airmass. Currently, a spectrum with an airmass of 2.4 is used. In the model, this transmittance spectrum is combined with the three Earth-exiting radiance spectra to provide top-of-theatmosphere radiance spectra both with and without trace gas absorption features. We use the model to investigate three types of spectral response features. The first study involves the out-of-band response from one of the Sea-viewing Wide Field-of-view Sensor (SeaWiFS) bands. Using the model, we demonstrate a technique to correct for the effect of that response on measurements of Earth-exiting radiances. The second study shows the effect of in-band spectral differences in an instrument band with multiple detectors. In this example, the effects are small, but differ with the type of Earth scene and with the amount of atmospheric trace gas absorption. On-orbit corrections for portions these detector-to-detector spectral differences are possible. However, at some level these differences will cause a residual striping in the bands measurements that cannot be removed. The final study examines measurements by a proposed multispectral grating-based spectrometer of the wavelength region near 760 nm, where there is a substantial absorption feature from atmospheric oxygen. Based on the bandwidth and wavelength spacing of the instruments bands, we investigate the use of the absorption feature to provide a wavelength calibration for the instrument. This model provides a tool for use in the design of new satellite instruments. In addition, it is possible to use the model to help mitigate the effects of actual spectral response features in instrument bands after those features are revealed during prelaunch characterization.

6744A-56, Session 10
The marine optical buoy (MOBY) radiometric calibration and uncertainty budget for ocean color satellite sensor vicarious calibration
S. W. Brown, B. C. Johnson, National Institute of Standards and Technology (USA); M. Feinholz, S. Flora, M. Yarbrough, D. Peters, Moss Landing Marine Labs. (USA); K. J. Voss, Univ. of Miami (USA); J. L. Mueller, San Diego State Univ. (USA); D. K. Clark, Marine Optical Consulting (USA) For the past decade, the Marine Optical Buoy (MOBY), a radiometric buoy stationed in the waters off Lanai, Hawaii, has been the primary ocean observatory for vicarious calibration of satellite ocean color sensors, including the U.S. Sea-viewing Wide Field-of-view Sensor (SeaWiFS) and Moderate Resolution Imaging Spectrometers (MODIS, Terra and Aqua) instruments. The MOBY vicarious calibration of these sensors supports the international effort to develop a global, multi-year time series of consistently calibrated ocean color data products. A longstanding radiometric ocean color uncertainty goal is to determine satellite-derived normalized water-leaving radiance (LWN) with a combined standard uncertainty of 5%; a critical component in the uncertainty budget is the uncertainty in the vicarious calibration. We will present the MOBY uncertainty budget for water-leaving radiance and discuss additional considerations related to the vicarious calibration of satellite ocean color sensors. Finally, we will discuss approaches being undertaken to reduce the uncertainties in MOBY in situ waterleaving radiance data.

6744A-57, Session 10
Summary of calibration and validation for KOMPSAT-2
D. Lee, Korea Aerospace Research Institute (South Korea) KARI has been doing Calibration and Validation activities for the KOMPSAT-2 (KOrea Multi-Purpose SATellite-2) after launch at July 28th 2006. The Cal/Val for KOMPSAT-2 has already been done for only requirements of KOMPSAT-2, and we need more Cal/Val activities to enhance the KOMPSAT-2 image quality for Users. The con-tent of KOMPSAT-2 Cal/Val defined before launch has been changed after launch. After launch, we could find out the new and different phenomenon and our mistakes from analyzing the KOMPSAT-2 image data. Firstly, we tried to define the status of KOMPSAT-2 (e.g. the sequence of MS band, Line of Sight, Pointing accuracy, etc.). Sec-ondly, every KOMPSAT-2 Cal/Val parameter was validated, and then KOMPSAT-2 was calibrated with the vali-dated and uploaded initial value of it (e.g. Non-Uniformity Correction table, Misalignment between body and sensor etc.). Finally, we could do to finalize the KOMPSAT-2 image data processing system for Users (e.g. De-noise, MTF compensation, Metadata, Users Manual, etc.)

6744A-54, Session 10
Scattered light characterization and correction algorithm for hyperspectral remote sensing imaging systems
K. R. Lykke, S. W. Brown, R. R. Bousquet, Y. Zong, National Institute of Standards and Technology (USA); G. Meister, Futuretech Corp. (USA); R. A. Barnes, NASA Goddard Space Flight Ctr. (USA) Improperly imaged, or scattered, optical radiation within an instrument is difficult to properly characterize and is often one of the dominant residual sources of measurement error. Scattered light can originate from the spectral components of a point source and from spatial elements of an extended source. The spectral and spatial scattered light components are commonly referred to as stray light and are described by an instruments spectral line spread function (LSF) and point spread function (PSF), respectively. In this talk, we present approaches that characterize an instruments response to

6744A-58, Session 10
Radiometric calibration and validation for KOMPSAT-2
D. Lee, Korea Aerospace Research Institute (USA) The Radiometric Cal/Val for the KOMPSAT-2 (KOrea MultiPurpose SATellite-2) has six parameters; Dynamic range, TDI level, Linearity, Video processor gain/offset, Non-uniformity correction table, De-noising, Electric gain/offset and Absolute radiometric Cal/Val. Before launched, Cal/Val site, equipments, code and procedure for the Radiometric Cal/Val have been defined and developed. After launched, KARI Cal/Val team has done the Radiomet-ric Cal/Val for the KOMPSAT-2, and then can guarantees the KOMPSAT-2 radiometric image qualities for Users.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

79

Conference 6744A: Sensors, Systems, and Next-generation Satellites

6744A-59, Session 11
MEDUSA: an ultra-lightweight multi-spectral camera for a HALE UAV
T. Van Achteren, B. Delaur, Flemish Institute for Technological Research (Belgium) The ESA-PRODEX funded MEDUSA project aims to develop a light weight high resolution multi-spectral earth observation instrument, which will be embarked on a solar-powered high altitude long endurance (HALE) UAV, operated at stratospheric altitudes (15 to 18km). :The MEDUSA instrument is designed to fill the gap between traditional airborne and spaceborne instruments regarding resolution and coverage. It targets applications such as crisis management and cartography, requiring high resolution images with regional coverage, flexible flight patterns, high update rates and long mission lengths (weeks to months). The MEDUSA camera is designed to operate at a ground resolution of 30 cm at 18 km altitude in the visible spectrum (400-650 nm), and a swath of 3000m. The central part of the payload is a focal plane assembly consisting of two frame sensors (PAN and RGB). The wide swath is realized with a custom designed highly sensitive CMOS sensor of 10000x1200 pixels. A GPS receiver and Inertial Measurement Unit (IMU) provide accurate position and attitude information. A direct downlink allows near-real time data delivery to the user. The on-board data processing consists mainly of basic image corrections and data compression (JPEG2000). The challenge lies mainly in fulfilling the requirements within the extreme environmental and physical constraints of the HALE UAV. Compared to traditional airborne and spaceborne systems, the MEDUSA camera system is ultra light weight (about 2 kg) and is operated in a low pressure and low temperature environment. System modelling and simulation is used to make careful trade-offs between requirements and subsystem performances. On 27th November 2006 the phase C/D for the design, production and test of the camera has started at VITO with the support of 9 industrial partners. The MEDUSA camera is expected to transmit its first images in the summer of 2008.

6744A-62, Session 11
MIMA a miniaturised Fourier spectrometer for Mars ground exploration: part II optical design
S. Fonti, G. Marzo, R. Politi, Univ. degli Studi di Lecce (Italy); G. Bellucci, Consiglio Nazionale delle Ricerche (Italy); B. Saggin, Politecnico di Milano (Italy) The Mars Infrared MApper (MIMA) is a FT-IR miniaturised spectrometer which is being developed for ESA ExoMars Pasteur mission. MIMA will be mounted on the rover mast and so it must be compact and light-weight. The scientific goals and its thermo-mechanical design are presented in two companion papers. In this work the optical design will be reviewed and the results of the tests performed on some optical components will be presented. The design has faced challenging constraints mainly linked to the requirement of keeping the performances good enough to fulfil the scientific objectives of the mission, while, at the same time, it was imperative to keep the overall size and weigh within the allocated resources. In addition the instrument must be able to operate in the very harsh environment of the Martian surface and to withstand, without permanent damage, even harsher conditions as well as the severe dynamic loads expected at landing on Mars. The chosen solution is a single channel double pendulum interferometer, covering the spectral range between 2 and 25 micron, crucial for the scientific interpretation of the recorded spectra, with a resolution variable between 10 and 5 cm-1. Since the spectral range is too wide to be covered by a single detector, it has been decided to use two different detectors, mounted side by side, in a customised case. Such innovative solution has obviously pros and cons and the optical design has been driven by the need to reduce the inconveniences, while maintaining the advantages.

6744A-63, Session 12
MIMA, a miniaturised infrared spectrometer for Mars ground exploration: part III, thermomechanical design
B. Saggin, E. A. Alberti, L. Comolli, M. Tarabini, Politecnico di Milano (Italy); G. Bellucci, Consiglio Nazionale delle Ricerche (Italy); S. Fonti, Univ. degli Studi di Lecce (Italy) The Mars Infrared MApper (MIMA) is a FT-IR miniaturised spectrometer which is being developed for ESA ExoMars Pasteur mission. MIMA will be mounted on the rover mast and so it must be compact and light-weight. The scientific goals and its optical design are presented in the previous papers, the focus of this work is on the thermo-mechanical design and testing. The design faces challenging constraints both from the environment and the allocated resources. The temperatures during operation are ranging from 120C to +30C with the presence of a low density but thermally effective atmosphere. Severe dynamic loads are expected during launch and mostly at landing on Mars. The overall size is limited to and envelope of 140 mmx140 mm x120 mm and the mass to less than 1 kg. The expected performances of this instrument should be comparable with those of much heavier ones built in the past. An instrument compliant with these constraints has been conceived, introducing many innovative solution with respect to the past experiences and making use of intensive modelling and testing to prove the survival to the harsh environment. Among the most challenging problems the mounting of the brittle KBr optics and the matching of its thermal expansion coefficient with that of the supporting aluminium structure, in a temperature interval of more than 200 C. Most of the components have undergone thermovacuum tests in the low temperature range because none of them was expected to be used in the -100 C range.

6744A-61, Session 11
MIMA, a miniaturized spectrometer Fourier infrared spectrometer for Mars ground exploration: part I, concept and expected performance
G. Bellucci, Consiglio Nazionale delle Ricerche (Italy); S. Fonti, Univ. degli Studi di Lecce (Italy); B. Saggin, Politecnico di Milano (Italy); D. Biondi, P. Cerulli, Consiglio Nazionale delle Ricerche (Italy); E. A. Alberti, Politecnico di Milano (Italy); G. Marzo, Univ. degli Studi di Lecce (Italy); F. Altieri, L. Zasova, Consiglio Nazionale delle Ricerche (Italy) The Mars Infrared MApper (MIMA) is a FT-IR miniaturised spectrometer which is being developed for ESA ExoMars Pasteur mission. The Martian Infrared MApper Fourier Spectrometer is designed to provide remote measurements of mineralogy and atmosphere of the scene surrounding a Martian rover and guide it to key targets for detailed in situ measurements by other rover experiments. Among the main scientific objectives of the MIMA instrument are to assist the rover in rock/soils selection for further in-situ investigation and to identify rocks and soils on the Martian surface which provide evidence of past/present biological activity. The instrument is also designed to measure the water vapour abundance and vertical distribution and its diurnal and seasonal variation, dust opacity, optical properties, composition, diurnal and seasonal variation. The instrument is a double pendulum interferometer providing spectra covering the 2 - 25 m wavelength domain with a resolving power of 1000 at 2 m and 80 at 25 m. The radiometric performances are SNR \> 50 in the near infrared and a NEDe = 0.002 in the thermal region. The instrument design is very compact, with a total mass of 1 kg and an average power consumption of 5 watt.

6744A-64, Session 12
Breadboarding activities of the TROPOMI-SWIR module
R. W. M. Hoogeveen, R. T. Jongma, P. J. J. Tol, A. M. S. Gloudemans, I. Aben, SRON Nationaal Instituut voor Ruimteonderzoek (Netherlands); J. de Vries, Dutch Space B.V. (Netherlands); H. Visser, E. C. Boslooper, TNO (Netherlands); M. R. Dobber, P. F. Levelt, Koninklijk Nederlands Meteorologisch Instituut (Netherlands)

80

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6744A: Sensors, Systems, and Next-generation Satellites


The TROPOMI instrument concept is part of the TRAQ mission proposal to ESA in response to the Call for Ideas in 2005. TRAQ (TRopospheric composition and Air Quality) has been accepted for a further pre-phase A study for the next Earth Explorer core Mission. A very similar instrument has been proposed for the CAMEO platform to the US National Research Council decadal study. TROPOMI is a nadir-viewing grating-based imaging spectrometer using the Dutch OMI and SCIAMACHY heritage. It includes an UV-VIS-NIR module that consists of three UVVIS channels continuously covering the 270-490 nm range to determine O3, NO2, HCHO, SO2, aerosols and a NIR-channel covering 710-775 nm for cloud detection and information on the aerosol height distribution using the oxygen A band. TROPOMI also includes a SWIR module covering 2305-2385 nm that mainly focuses on determination of CO and CH4 total columns. All species are measured with sensitivity down to the Earths surface, thus addressing issues of anthropogenic emissions and their impact on air quality and climate. In the TRAQ mission, unique diurnal time sampling with up to 5 daytime observations over mid-latitude regions (Europe, NorthAmerica, China) is realized by using a non-sun-synchronous, medium-inclination drifting orbit and a 2600 km wide observational swath. Several more general aspects related to the TROPOMI instrument are discussed in a separate paper in this conference. This paper focuses on the development of the SWIR module. A breadboard model (BBM) has been designed and constructed which is as much as possible functionally flight representative. Critical technologies to be demonstrated with the BBM are the SWIR HgCdTe-based 2D focal plane array, the on-board SWIR calibration LED, and in particular, the in-house developed silicon-based immersed grating that allows a hugely reduced instrument volume. In the presentation the results of a performance analysis of the TROPOMI-SWIR channel will be discussed, as well as results of the detector characterization program on a representative off-the-shelf FPA, and details of the photolithographic production of the immersed grating. quasioptical antenna, a superconducting phase-locked Flux Flow Oscillator (FFO) acting as Local Oscillator (LO) and SIS harmonic mixer (HM) for FFO phase locking. The latest results from the pre-flight test and calibration campaign will be presented.

6744A-66, Session 12
Performance of the imaging spectral signature instrument (ISSI) breadboard
U. Kantojrvi, H. K. Saari, K. Viherkanto, VTT (Finland); E. Herrala, Specim Spectral Imaging Ltd. (Finland); B. Harnisch, European Space Agency (Netherlands) With hyperspectral pushbroom imaging spectrometers it is possible identify ground pixels by their spectral signature. The Imaging Spectral Signature Instrument (ISSI) concept performs optical on-board processing of the hyperspectral data cube to identify pixels with a pre-defined and programmable spectral signature. An aircraft compatible breadboard of the ISSI has been designed. It consists of an imaging telescope, which images an object line on the entrance slit of a first imaging spectrograph, which disperses each pixel of the object line into its spectral content on a liquid crystal spatial light modulator. This component is programmed with a spatial transmission behaviour, which is constant along the spatial pixels and equal to the spectral filter vector of the searched specific signature along the spectral pixels. A second inverted spectrograph reimages the transmitted flux into a line of pixels on a CCD detector. ISSI operates at wavelengths between 500nm and 900nm. The spectral filter vector can be selected for 800 spatial pixels with a spectral resolution of 4 nm and almost 8 bit modulation capability. The selection of the filter vector to be programmed on the LCD is dependent on what kind of a learning set is available of the signature backgrounds. The signatures and backgrounds are acquired with ISSI in a spectrometer operating mode. The breadboard will undergo a test program consisting of calibration and verification of the spectral, spatial and radiometric performance and target identification capability.

6744A-65, Session 12
Balloon-borne heterodyne stratospheric limb sounder TELIS ready for flight
R. W. M. Hoogeveen, P. A. Yagoubov, G. de Lange, A. A. J. de Lange, SRON Nationaal Instituut voor Ruimteonderzoek (Netherlands); V. P. Koshelets, Institute of Radio Engineering and Electronics (Russia); B. N. Ellison, Rutherford Appleton Lab. (United Kingdom); M. Birk, Deutsches Zentrum fr Luftund Raumfahrt e.V. (Germany) TELIS (Terahertz and submm Limb Sounder) is a three-channel balloon-borne heterodyne spectrometer for atmospheric research. The observational techniques of TELIS can be compared to the presently flying MLS instrument on board NASAs EOS-Aura satellite, but TELIS is packed with new generation cryogenic heterodyne technologies: novel low-noise cryogenic heterodyne mixer detectors and their local oscillators, novel low-noise cryogenic high-frequency amplifiers, novel back-end spectrometer. TELIS will fly on the MIPAS-B2 gondola. The two instruments together will yield the most complete set of stratospheric constituents. TELIS is a cooperation between the European institutes DLR (PI-institute) RAL and SRON. The first flight is foreseen in November 2007. Many atmospheric trace gases have their rotational transitions in the sub millimeter and THz range, yielding a very rich spectrum. Limb sounding results in very accurate vertical profiles. The three TELIS receivers will operate simultaneously. The 500 GHz channel is developed by RAL and will produce vertical profiles of BrO, ClO, O3 and N2O. The 1.8 THz channel is developed by DLR and will mainly target the OH radical, and will also measure HO2, HCl, NO, NO2, O3, H2O, O2 and HOCl. Finally the 550 - 650 GHz channel is developed by SRON and IREE and will measure profiles of ClO, BrO, O3, HCl, HOCl, H2O and its 3 isotopomers, HO2, NO, N2O, HNO3, CH3Cl and HCN. In the presentation the science and technology of TELIS will be discussed with emphasis on the channel developed by SRON. It contains a Superconducting Integrated Receiver (SIR), which combines on a 4x4 mm2 chip the low-noise SIS mixer and its

6744A-67, Session 12
Spatial heterodyne spectrometer for FLEX
A. D. Scott, COM DEV International Ltd. (Canada) and York Univ. (Canada); A. Bell, COM DEV International Ltd. (Canada) A spatial heterodyne spectrometer (SHS) has significant advantages for high spectral resolution imaging over narrow Fraunhoffer bands compared to traditional solutions. Given comparable optical tendue at R~6500, a field-widened SHS will have a throughput-resolution product ~170 x larger than an air-spaced etalon spectrometer, and ~1000 x larger than a standard grating spectrometer. The monolithic glass Michelson design and lack of moving parts allows maximum stability of spectral calibration over the FLEX mission life.

6744A-68, Session 12
The initial checkout of multi-spectral camera (MSC) system
S. Yong, H. Choi, Korea Aerospace Research Institute (South Korea); S. Ra, Chungnam National Univ. (South Korea) Multi-Spectral Camera(MSC) is a payload on the KOMPSAT-2 satellite to perform the earth remote sensing. The instrument images the earth using a push-broom motion with a swath width of 15 km and a ground sample distance (GSD) of 1 m over the entire field of view (FOV) at altitude 685 Km. The instrument is designed to have an on-orbit operation duty cycle of 20% over the mission lifetime of 3 years with the functions of programmable gain/ offset and on-board image data compression/storage. KOMPSAT-2 was launched on July 28 2006 and stared early operation including initial activation checkout. During IAC, MSC was check and validated through prepared procedure. In this paper, the checkout preparation, configuration and sequence so on are described. The results of the checkout including analysis will be shown.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

81

Conference 6744A: Sensors, Systems, and Next-generation Satellites

6744A-95, Session 8:
Sorption cooling: a valid extension to passive coolers in space missions
J. Doornink, Dutch Space B.V. (Netherlands); J. Burger, Marcel ter Brake, Univ. of Twente (Netherlands) Passive cooling has shown to be a very dependable cryogenic cooling method for space missions. Several missions employ passive radiators to cool down their delicate sensor systems for many years, without consuming power, without exporting vibrations or producing electromagnetic interference. Passive coolers tend to get increasingly bulky if the cooling temperatures are below 80 K. Due to the Stefan-Boltzmann law, multi-stage concepts are required and the radiator area increases rapidly as the cooling temperature drops. Therefore a technology is sought that enhances the performance of passive cooling at lower temperatures, while preserving as much as possible of its advantages. In that sense, sorption cooling has been found to be an appropriate technology. Sorption cooling employs a closed cycle Joule-Thomson expansion process to achieve the cooling effect. Sorption cells perform the compression phase in this cycle. At a low temperature and pressure, these cells adsorb the working medium, at a higher temperature and pressure they give off the medium and thus produce a flow to the expander in the cold stage. The selected sorption process is of the physical type, which is completely reversible and does not suffer from degradation. Sorption coolers include few moving parts, export neither mechanical vibrations nor electromagnetic interference and are potentially very dependable due to their simplicity. The required cooling temperature determines the type of working medium to be applied. Sorption coolers are used in conjunction with passive cooling for heat rejection at different levels. This paper starts with a brief discussion on applications of passive coolers in different types of orbits and gives an overview of the projected mass of active and passive coolers, depending on the cooling temperature. Next the working principle of sorption cooling is summarized. The DARWIN mission is chosen as an example application of sorption and passive cooling. The 4K DARWIN cryocooler architecture has been optimized recently and one of the results is a drastic reduction of the radiator area needed by the sorption cooler. The application field of this type of sorption cooling in space missions is currently being expanded by examining the performance of alternative working media, suitable for different cooling temperatures in the range up to 150 K.

82

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6744B: Global Earth Observing System of Systems, Implementation Strategies and Applications
Thursday 20 September 2007 Part of Proceedings of SPIE Vol. 6744B Global Earth Observing System of Systems, Implementation Strategies and Applications

6744B-94, Poster Session


Characterization and modeling of an improved method of phase detection scheme for displacement optic sensors
N. N. Puscas, Univ. Politehnica Bucuresti (Romania); H. W. Gnewuch, A. G. Podoleanu, D. A. Jackson, Univ. of Kent (United Kingdom) Based on the method presented in references [1] and [2] to overcome the DC drift in RF subcarrier phase detection schemes used in fibre optic sensors we characterized an improved method (open loop) for the measurement of small displacements and vibrations. The method is characterized by low distortions in the modulation process, good signal-to-noise ratio and low cost. We used an optical heterodyne system for detection. For two values of the injection currents of the laser diode Hitachi HLP 1400 used in the experimental work and based on the theoretical model of the ideal receiver we obtained for the measurements of small distances values in the range 0.7 m 4.5 m. Also, in this paper a modelling of the optical noise for the above mentioned optical sensor is presented and analysed. For the small range distances above mentioned we obtained the values for the signal-to-noise ratios in the range 6.6&#61620;10-7 dB 6.9 &#61620; 10-7 dB. The system can be exploited in the measurement of small distances, vibrations and seismic detection. References [1] G. Thursby, F. Dong, B. Culshaw, A novel method for overcoming DC drift in RF subcarrier phase detection scheme for fibre optic sensors, Proc. SPIE, paper WeP-15, OFS-16, 360 (2003). [2] H. Gnewuch, N. N. Puscas, D. A. Jackson, A. Gh. Podoleanu, Improved method of phase detection scheme for displacement optic sensors, Journal of Optoelectronics and Advanced Materials, Vol. 8, No. 1, p. 387-391, 2006.

will address near-term, mid-term and long-term outcomes to respond to the GEOSS 10 Year Implementation Plan 2 year, 6 year and 10 year targets. Particular emphasis is given to the near-term outcomes to present at the GEO Ministerial Conference in November 2007. This paper will present an overview of the CEOS Constellations, including the background, strategy and planning, and implementation.

6744B-78, Session 13
The CEOS constellation for land surface imaging
G. B. Bailey, U.S. Geological Survey (USA); M. Berger, European Space Agency (Netherlands); H. Jeanjean, Ctr. National dtudes Spatiales (France); K. P. Gallo, U.S. Geological Survey (USA) A constellation of satellites that routinely and frequently image the Earths land surface in consistently calibrated wavelengths from the visible through the microwave and in spatial detail that ranges from sub-meter to hundreds of meters would offer enormous potential benefits to society. A well-designed and effectively operated land surface imaging satellite constellation could have great positive impact not only on the quality of life for citizens of all nations, but also on mankinds very ability to sustain life as we know it on this planet long into the future. The primary objective of the Committee on Earth Observation Satellites (CEOS) Land Surface Imaging (LSI) Constellation is to define standards (or guidelines) that describe optimal future LSI Constellation capabilities, characteristics, and practices. Standards defined for a LSI Constellation will be based on a thorough understanding of user requirements, and they will address at least three fundamental areas of the systems comprising a Land Surface Imaging Constellation: the space segments, the ground segments, and relevant policies and plans. Studies conducted by the LSI Constellation Study Team also will address current and shorter-term problems and issues facing the land remote sensing community today, such as seeking ways to work more cooperatively in the operation of existing land surface imaging systems and helping to accomplish tangible benefits to society through application of land surface image data acquired by existing systems. Results from 2007 LSI Constellation studies designed to establish initial international agreements, develop preliminary standards for a mid-resolution land surface imaging constellation, and contribute data to a global forest assessment will be highlighted.

6744B-77, Session 13
CEOS constellations: space agency response to GEOSS
B. Ryan, U.S. Geological Survey (USA); S. Briggs, European Space Agency (USA); T. Inada, Japan Aerospace Exploration Agency (Japan); D. Vidal-Madjar, Ctr. National dtudes Spatiales (France); S. Ward, Symbios Communications (Australia); C. Ishida, Japan Aerospace Exploration Agency (Japan) The Committee on Earth Observation Satellites (CEOS) Constellations is a new mechanism to co-ordinate space-based Earth observation missions to respond to the challenges of the Global Earth Observation System of Systems (GEOSS). The CEOS Constellations is aimed at: moving space agency coordination efforts from the generic to the specific; improving considerably the extent to which the combined outputs of the various space agency programs are relevant to key applications, such as climate and other GEOSS Societal Benefit Areas (SBA); facilitating participation of smaller contributors, through the adoption of a series of standards and guidelines which satisfy key GEOSS requirements; and thereby contributing to realization of the space-based observations required by the GEOSS 10 Year Implementation Plan. Since agreement on the draft Constellations process paper and initiation of four prototype constellations (Ocean Surface Topography, Land Surface Imaging, Precipitation, Atmospheric Composition) at the 21st CEOS Pleanry in Buenos Aires, November 2006, interested CEOS members have established constellation study teams. These teams are addressing the consolidation of user requirements, combined capabilities and gaps, key standards to be adopted by satellite systems instruments, ground systems, data and products, data principles, supporting inter-agency agreements etc. The studies

6744B-79, Session 13
The CEOS constellation for ocean surface topography
S. Wilson, National Oceanic and Atmospheric Administration (USA); F. Parisot, EUMETSAT (Germany) Precision measurements from satellite radar altimeters have revolutionized our understanding of ocean circulation and the critical role the oceans play in climate, and it is giving us new insight into the interaction between the ocean and the atmosphere that influences extreme weather events such as hurricanes. The TOPEX/Poseidon and Jason missions have provided highly accurate measurements of sea surface topography. Other altimeter missions have provided additional observations extending coverage to higher latitudes and finer spatial scales. As an essential in-situ complement, the Argo Program, is nearing full global coverage by 3,000 profiling floats, with completion anticipated later in 2007. Oceanography is now becoming operational in the same sense that meteorology has been for the past few decades. The goal of the constellation is to sustain the fundamental climate data record for global sea level, as well as seasonal, interannual and decadal variability, and to ensure continuity of service for both operational and research users.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

83

Conference 6744B: Global Earth Observing System of Systems, Implementation Strategies and Applications
This Constellation Team is fostering discussions amongst agencies active in satellite altimetry - CNES in France, ESA and EUMETSAT in Europe, ISRO in India, NASA and NOAA in the U.S., and SOA in China - with the objective of developing an overall altimetry system that would promote the use of common approaches. The participation of those agencies and their commitment to this concept will be critical to its success. And that success itself is key, if we are to capitalize on the substantial investments in satellite altimetry by sustaining our demonstrated capability to collect climate data records and ensure continuity of service to all users. Constellations is the title given to this new process, and 4 pilot studies have been initiated in order to pioneer and test the concept. JAXA and NASA were selected as the lead agencies for the study of the development of a CEOS Precipitation Constellation with the support of other CEOS space agency and user community participants. The goals, approach, and anticipated outcomes for the study will be presented.

6744B-82, Session 14
Semantics-enabled knowledge management for global Earth observation system of systems
R. L. King, Mississippi State Univ (USA); S. S. Durbha, Y. Ling, N. H. Younan, Mississippi State Univ. (USA) The Global Earth Observation System of Systems (GEOSS) is a distributed system of systems built on current international cooperation efforts among existing Earth observing and processing systems. The goal is to formulate an end-to-end process that enables the collection and distribution of accurate, reliable Earth Observation data, information, products, and services to both suppliers and consumers worldwide. One of the critical components in the development of such systems is the ability to obtain seamless access of data across geopolitical boundaries. In order to gain support and willingness to participate by countries around the world in such an endeavor, it is necessary to devise mechanisms whereby the data and the intellectual capital is protected through procedures that implement the policies specific to a country. Earth Observations (EO) are obtained from a multitude of sources and requires coordination among different agencies and user groups to come to a shared understanding on a set of concepts involved in a domain. It is envisaged that the data and information in a GEOSS context will be unprecedented and the current data archiving and delivery methods need to be transformed into one that allows realization of seamless interoperability. Thus, EO data integration is dependant on the resolution of conflicts arising from a variety of areas. Modularization is inevitable in distributed environments to facilitate flexible and efficient reuse of existing ontologies. Therefore, we propose a framework for modular ontologies based knowledge management approach for GEOSS and present methods to enable efficient reasoning in such systems. This analysis is exercised in a coastal zone domain.

6744B-80, Session 13
The CEOS atmospheric composition constellation: an example of an integrated Earth observing system for GEOSS
E. Hilsenrath, NASA Headquarters (USA); J. Langen, European Space Agency (Netherlands) The Atmospheric Composition (AC) Constellation is one of four pilot projects initiated by CEOS to bring about technical/ scientific cooperation among space agencies that meet the goals of GEO and comply with the CEOS member agencies national programs. The Constellation concept has been endorsed in the GEO Work Plan, 2007-2009. The AC Constellation goal is to collect and deliver data to develop and improve monitoring, assessment and predictive capabilities for changes in the ozone layer, air quality and climate forcing associated with changes in the environment. These data will support five of the nine GEO SBAs: Health, Energy, Climate, Hazards, and Ecosystems. At the present time ESA, CSA, NIVR, NASA, NOAA and Eumetsat are participating in the Constellation study, and have major assets in orbit including 17 instruments on seven platforms. Constellation Concept study will identify missing capabilities that will result when the present orbiting research satellites missions end and those not taken by the next generation operational missions. Missing observations include very accurate and high spatial resolution measurements needed to be to track trends in atmospheric composition and understand their relationship to climate change. The following are the top level objectives for the AC Constellation Concept Study: Develop a virtual constellation of existing and upcoming missions using synergies among the instruments and identify missing capabilities. Study advanced architecture with new space assets and varying orbits with expectations that new technology could also be brought forward to best meet user requirements Data system interoperability to insure that data are useful, properly targeted, and easily accessible.

6744B-83, Session 14
Multi-sensor approach to address sustainable development
S. Habib, NASA Goddard Space Flight Ctr. (USA) The main objectives of Earth Science research are many folds: to understand how does this planet operates, can we model her operation and eventually develop the capability to predict such changes. However, the underlying goals of this work are to eventually serve the humanity in providing societal benefits. This requires continuous, and detailed observations from many sources in situ, airborne and space. By and large, the space observations are the way to comprehend the global phenomena across continental boundaries and provide credible boundary conditions for the mesoscale studies. This requires a multiple sensors, look angles and measurements over the same spot in accurately solving many problems that may be related to air quality, multi hazard disasters, public health, hydrology and more. Therefore, there are many ways to address these issues and develop joint implementation, data sharing and operating strategies for the benefit of the world community. This is because for large geographical areas or regions and a diverse population, some sound observations, scientific facts and analytical models must support the decision making. This is crucial for the sustainability of vital resources of the world and at the same time to protect the inhabitants, endangered species and the ecology. Needless to say, there is no single sensor, which can answer all such questions effectively. Due to multi sensor approach, it puts a tremendous burden on any single implementing entity in terms of information, knowledge, budget, technology readiness and computational power. And, more importantly, the health of planet Earth and its ability to sustain life is not governed by a single country, but in reality, is everyones business on this planet. Therefore, with this notion, it is becoming an impractical problem by any single

6744B-81, Session 13
CEOS precipitation constellation
S. P. Neeck, NASA Headquarters (USA); R. Oki, Japan Aerospace Exploration Agency (Japan) The outcomes of the 19th Committee on Earth Observing Satellites (CEOS) Plenary held in London in November 2005, recognized that the CEOS Implementation Plan for SpaceBased Observations for GEOSS should: identify the supply of space-based observations required to satisfy the requirements expressed by the 10-year implementation plan for the Global Earth Observation System of Systems (GEOSS); and propose an innovative process whereby the many disparate types of Earth observing programs funded by CEOS Member agencies might contribute to the supply of the required observations. The CEOS Task Force charged with drafting the CEOS Implementation Plan for Space-Based Observations for GEOSS focused its early efforts on the creation of a new planning process which would satisfy the various criteria demanded by member space agencies, and which would hopefully encourage a new phase of specificity and focus in the multi-lateral cooperation efforts undertaken by space agencies under the CEOS umbrella - resulting in improved engagement of all CEOS Members and real implementation results. The CEOS

84

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6744B: Global Earth Observing System of Systems, Implementation Strategies and Applications
organization/country to bear this colossal responsibility. So far, each developed country within their means has proceeded along satisfactorily in implementing their Earth observing needs but it has left a big void in the developing world who have very limited resources to invest in the space measurements. This paper gives some serious thoughts in what options are there in undertaking this tremendous challenge. The problem is multidimensional in terms of budget, technology availability, environmental legislations, public awareness, and communication limitations. Some of these issues are introduced, discussed and possible implementation strategies are provided in this paper to move out of this predicament. A strong emphasis is placed on international cooperation and collaboration to see a collective benefit for this effort. Leonardi, Ministro Della Difesa (Italy) COSMO-SkyMed (COnstellation of small Satellites for Mediterranean basin Observation), is a Dual-Use (Civilian and Defence) end-to-end Earth Observation System aimed to provide data, products and services for a variety of applications, such as Risk Management, Scientific and Commercial and Defence/Intelligence applications. The space segment is composed of four Low Earth Orbit satellites, equipped with a multi-mode high-resolution Synthetic Aperture Radar (SAR) operating at X-band and fitted with innovative data transmission equipment to allow high-speed transmission of data to ground stations. COSMO-SkyMed Mission responds to actual needs of Earth Observation Market, being characterized by full global coverage, all weather, day/night acquisition capability, higher resolution, higher accuracy (geo-location, radiometry, etc.), superior image quality, fast revisit/response time, interferometric/polarimetric capabilities and quicker-and-easier ordering and delivery of data, products (Standards and Higher-Level) and services. The system is characterized by an intrinsic Inter-operability with other EO missions and Expandability towards other possible partners with different sensors typologies in order to provide Earth Observation integrated services to large User Communities and Partner Countries. International cooperation programs are illustrated, such as ORFEO (Optical and Radar Federation for Earth Observation) and SIASGE (ItalianArgentinean Satellite System for Emergency Management). The Dual-Use Mission Planning functionality optimizes system utilization, fulfilling at the same time the needs of different user classes, sharing System Resources. The launch of the first COSMO-SkyMed satellite is foreseen for the first semester of 2007. The remaining 3 satellites will be gradually deployed by the end of 2009.

6744B-84, Session 14
Multi-sensor web system for natural hazard applications
D. P. Ouzounov, S. Habib, NASA Goddard Space Flight Ctr. (USA); G. Cervone, George Mason Univ. (USA); F. S. Policelli, NASA Goddard Space Flight Ctr. (USA) Mutisensor Sensor Web System (MSWS) consists of a portable and widely applicable technology to advance the knowledge and practical use of remote sensing for Natural hazard applications and to maximize the use and sharing of NASA satellite observations. Our rationale for using MSWS is that there are insufficient spatial and temporal coverage using single measurements, and therefore a common platform is needed. The main characteristics of the MWSS prototype are: 1/ use of multiple and already validated physical measurements to be fused into one framework with the latest theoretical models; 2/ enabling model interactions with sensor webs; 3/ provided a feedback on data gaps that may then be acquired from other sources; 4/ provide advanced data mining algorithms to classify, cluster, and find patterns in the data. The initial prototype of MSWS was developed for monitoring the earthquake hazard. MSWS system is a sensor web of existing satellite sensors (MODIS, GOES, METEOSAT, POES, and DEMETER) and ground observations (e.g., Global Positioning System (GPS)) and provides robust continuous monitoring over specific areas of known earthquake hazard. A specialized set of data mining algorithms is employed within the MSWS framework to quickly analyze the data for anomalies which may carry early warning information. We argue that MSWS system is able to perform as early warning system, by bringing different approaches within a common framework.

6744B-87, Session 14
TIGER: supporting the development of water information systems and services in Africa, African water management with EO
F. Palazzo, L. Ghaye, European Space Agency (Italy); S. Iris, Canadian Space Agency (Canada); A. Lipponen, United Nations Educational, Scientific and Cultural Organization (France) Water information systems in Africa are severely degraded, and policies and management decisions in many African countries are currently based on sparse and unreliable information. In 2000, the UN General Assembly included among the Millennium Development Goals (MDGs) a task to reduce by half the proportion of people without sustainable access to safe drinking water. During the 2002 Johannesburg World Summit on Sustainable Development (WSSD), the Heads of States and delegations called on all countries to develop Integrated Water and Resources Management (IWRM) and water efficiency strategies. Achieving this objective in developing areas of the world represents a significant challenge. In 2002, responding to the urgent need for action in Africa, the European Space Agency (ESA), within the context of the Committee of Earth Observation Satellites (CEOS), launched the TIGER initiative. The overall scope is to assist African countries to overcome problems faced in the collection, analysis and use of water related geo-information by exploiting the advantages of Earth Observation (EO) technology. The TIGER initiative has evolved thanks to the contributions of partners such as UNESCO (and its International Hydrological Program), UN Africa-Water group (UN-Economic Commission for Africa), African and international institutions in the context of water resource management (e.g. African Ministers Council on Water (AMCOW)), space agencies (e.g. Canadian Space Agency - CSA), development partners (e.g. African Development Bank) and other international and African organizations (e.g. Council for Scientific and Industrial Research (CSIR), RamsarAfrica, African Departments of Water). The TIGER initiative is implemented through several crosscutting action lines (training, networking and improved access to EO data), aiming to facilitate the development of Water Information Systems and Services in Africa. In this context, the TIGER model is a long-term approach following a three stages

6744B-85, Session 14
Use of remote sensing for decision support in Africa
F. S. Policelli, NASA Goddard Space Flight Ctr. (USA); M. Brown, Science Systems and Applications, Inc. (USA) Over the past 30 years, the scientific community has learned a great deal about the Earth as an integrated system. Much of this research has been enabled by the development of remote sensing technologies and their operation from space. Decision makers in many nations have begun to make use of remote sensing data for resource management, policy making, and sustainable development planning. This paper makes an attempt to provide a survey of the current state of the requirements and use of remote sensing for sustainable development in Africa. This activity has shown that there are not many climate data ready decision support tools already functioning in Africa. There are, however, end-users with known requirements who could benefit from remote sensing data.

6744B-86, Session 14
COSMO-SkyMed program: an advanced dualuse asset for Earth observation
G. Valentini, A. Coletta, Agenzia Spaziale Italiana (Italy); G. Angino, Alcatel Alenia Space (Italy); F. Battazza, F. Caltagirone, Agenzia Spaziale Italiana (Italy); F. Impagniatiello, Alcatel Alenia Space (Italy); A. Capuzi, Agenzia Spaziale Italiana (Italy); S. Fagioli, Alcatel Alenia Space (Italy); R.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

85

Conference 6744B: Global Earth Observing System of Systems, Implementation Strategies and Applications
approach: Research, Pre-operational and Operation stage. This contribution will provide an update on the status of the initiative and its achievements. in order to reduce the costs of the acquisition is presented in this paper. Due to the various available high resolution satellites the accuracy is not limited by the spatial resolution of the imagery, but principally by the accuracy of the direct sensor orientation. The paper also proposes methods to increase the accuracy, to eliminate blunders and to enhance the integrity of the acquired terrain information in order to achieve applicability in aviation.

6744B-88, Session 15
Dissemination and exploitation of grids in Earth science
W. J. Som de Cerff, Koninklijk Nederlands Meteorologisch Instituut (Netherlands); M. Petitdidier, M. Lonjaret, Ctr. National de la Recherche Scientifique (France); L. Fusco, European Space Agency (Italy); L. Hluchy, Institute of Informatics (Slovak Republic) The DEGREE project is a consortium of ES and computer science partners aiming at promoting the uptake of Grid in ES ES needs to deal with enormous amounts of data and large computational needsFor ES e-science can be an essential improvement in research, operation and business, especially when Grid services can be coupled to existing ES services. But is e-science mature and does it cover all ES requirements? DEGREE aims at answering these questions by capturing ES requirements and analyzing Grid solutions. To capture evolving ES requirements for Grid services over 20 application scenarios were collected and grouped into families of applications. This will ease communication, maintenance and tracing of requirements. More ES applications scenarios will be added; updating and analyzing requirement progress will be a continuous effort Three families have been identified, following a scheme with three different levels of complexity. In this case complexity is referring to the complexity of requirements placed on Grid technology by the application, not of the application itself. The first family is called the simple applications, the second the complex applications and the third family is the complex workflow applications. In short specific ES Grid requirements are: Reliability (QoS); real-time and instantaneous access; The need to access licensed software; Data policies on input and output data i.e. complicated security requirements; data is scattered around various institutes in various formats, in various databases and has metadata attached to it in various forms, i.e. data management (accessibility, harmonization) is essential in ES. Workflow orchestration is vital to cover requirements from the third family of applications. Key services for ES are workflow management and data and metadata management. In order to make e-science take the next step in ES, there is a need for standardization on Grid service level so existing and new web services can easily be coupled to the Grid. Portals improving the accessibility to data, computing and results will greatly improve ES research and Grid infrastructure usage.

6744B-90, Session 15
The increase of the efficiency of research tools of active radio and optical sensing
V. G. Oshlakov, T. Eremina, Institute of Atmospheric Optics (Russia) 5. In radio and optically active detection and ranging of objects a signal from an artificial object, at the noise background should be recognized. This noise may be the background created both by a foreign source and the background from the underlying surface, on which the object being studied is located. Using the source polarization of a sensing signal of optical radar and polarization elements of the receiver, we can increase the value of the signal from the studied object as compared with the noise background. The versions of the efficiency increase of research tools of active radio and optical sensing are considered at the following conditions: 1. The background from the external source is known. 2. The scattering matrix of underlying surface creating the background is known. 3. The scattering matrix of underlying surface creating the background is unknown. 4. The external source and underlying surface present the background source: a) the Stokes vector of external source and the scattering matrix of underlying surface are known; b) the Stokes vector of external source and the scattering matrix of underlying surface are unknown; c) the Stokes vector of external source are known and the scattering matrix of underlying surface is unknown and vice versa.

6744B-91, Session 15
The drawing of snow information in north Xinjiang based on MODIS data and its application
P. Huan, Nanjing Univ. (China); S. Fang, Xinjiang Univ. (China) Snow is the important freshwater resource, and it is an important factor, which can affect the global climate. It is one of the most active element on the earth, its character, such as snow area, snow distributing and snow depth, is the important input parameter to global energy balance, climate and ecological model. Drought area is sensitive to the global climate change, and snow, glacier reflect prominently in the drought area climate change, especially seasonal snow, which is not only the most active environment affecting factor, but also is the most sensitive region climate change responding factor. So the significance of monitoring snow depth in Xinjiang is important, on the one hand, which can improve snow-monitoring ability, on the other hand, which can provide credible information in time to local water resource management and ecology construct. This study analyzed the correlativity of each band with snow depth, and compared true snow measuring spectrum with image spectrum, picked out the most sensitive band to snow depth, build inversion snow depth model. At last, integrating the snow depth drawing method and true surveying data in the field, analyzing the snow distributing character, snow change in annual and season at length.

6744B-89, Session 15
Analysis of terrain data based on satellite imagery for aviation purposes
B. Eilmus, G. Heidelmeyer, U. Klingauf, Technische Univ. Darmstadt (Germany) Due to upcoming data driven technologies in the modern aviation the demand for digital terrain data is growing conspicuously. According to the regulations of the ICAO (International Civil Aviation Organization) each contracting state will be obligated to provide electronic terrain data of its territory. This regulation poses a problem particularly for developing and newly industrializing countries. In these countries terrain information meeting the required quality standards is often not available and acquisition methods like airborne photogrammetry and LIDAR are too expensive and therefore not suitable. Also the data resulting from completed and future radar missions like SRTM and TanDEM-X can not solve this problem completely, because terrain data derived from radar sensors may have considerable data voids particularly in mountainous terrain, deserts and iced and snow-covered areas. These voids have to be filled with data of adequate quality. Studies performed at the Institute of Flight Systems and Automatic Control at the Technische Universitaet Darmstadt show that a promising method to support this task is the use of high resolution optical satellite imagery with a stereoscopic coverage. The resulting accuracy of terrain data generated on the basis of satellite imagery without using ground control points

86

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6744B: Global Earth Observing System of Systems, Implementation Strategies and Applications

6744B-92, Session 15
Pump-laser-induced multi-structure photoprocesses of the near-lying singlet- and triplet-excited states in geteroaromatic molecules
A. E. Obukhov, Moscow Mining Institute (Russia) To solve such a problem, one has to reveal general relationships between the structure of excited electronic states and transitions of different spin and orbital nature and photophysical properties of organic molecules. The introductions made it possible to take into account pumping and lasing fields in elementary volume. In this case, one has to determine, e. g. to measure, the spectral parameters of each molecule whose lasing properties are tested. Assumptions should be made if the spectral parameters are unknown. In the latter case, the reability of estimates may be lowered. Figure 1. The now scheme none optical transition from a working sates through active fluctuations {n}, term Tf; {l}, quasi continuum inactive in none optical transition of oscillatory levels of term Tm; , matrix elements spin-orbital interaction, increased on the appropriate integrals of Franc - Condon; , matrix element operators on in-harmonize, &#1043;n, an width of levels active fluctuations due to an oscillatory relaxation; arrows it is shown channels of optical excitation, The circuit none optical , and , and , and , (cm-1) transitions of the spectrum reabsorption, , and , from a working condition through active fluctuations {n} terms ; {l} queasy continuum inactive in none optical transition of oscillatory levels of terms Tm (Tn); arrows it is shown channels of laser pump-pulse excitation. Of the constant of inter-combination conversion are given by modified formula: where is the energy gap between the interacting and states. The ultimate duration of the leading edge of the pumping pulse that permits lasing in the active medium is given, with allowance for reabsorption due to the and transitions, by the formula: The threshold condition of lasing is imposed not on the intensity of the exciting light but on the rate of its growth in time. Reference: [1]. Obukhov. A. E. Effect of Electronically Excited Singlet and Triplet States on the Photophysical Properties in the Series of N, O, S Heterocyclic Compounds, Laser Physics, 2003, Vol. 3, P. 847 - 855. [2]. Obukhov A. E. Excited Singlet and Triplet States and of the New Photoionization Electron-Vibrational Mechanism in the Laser-Active Molecules, Proceedings of SPIE, 2003, Vol. 5121, P. 147 - 156. [3]. Obukhov A. E., The Physical Principles of Simulating the Structure and Photophysical Properties of Complex Heteroaromatic Compounds by Means of the LCAO MO SCF Methods, Laser Physics, Vol. 7 (5), P. 1102 - 1131, 1997.

6744B-93, Session 15
Detection of hydrocarbons using hyperspectral satellite imagery
M. R. Mehdi, National Univ. of Science and Technology (Pakistan); S. S. Butt, LMKR (Pakistan) Improving Management of Energy Resources In our rapidly expanding world, Energy Resources are depleting quickly. If ever there was a need to manage the Energy Resources, it is now. One of the richest industries in the world regarding Energy is the Oil and Gas sector. An enormous amount of money is spent on seismic surveys to identify the probable locations where oil or gas could be found. With the advent of modern sensing and digital processing instruments, Remote Sensing is getting to newer heights. Hyper spectral Satellite images are being used to detect hydrocarbons from space. Many oil and gas reservoirs naturally leak tiny amounts of hydrocarbons at the Earths surface. These are mostly light gases such as Methane and Ethane. Common biological processes can produce Methane (CH4), but Ethane (CH) is only made in underground oil and gas reservoirs. Every object/phenomenon has a unique spectral signature. The study area, Qadirpur (Ghotki), has been selected in collaboration with Oil & Gas Development Company Limited (OGDCL). Hyperspectral image was obtained from Hyperion which is on board Earth Observing-1 (EO-1).It is capable of resolving 220 spectral bands (from 0.4 to 2.5 m) with a 30-meter spatial resolution. The instrument can image a 7.5 km by 100 km land area per image. Three images were acquired and then mosaicked to study the area. Locations of existing well points were identified through GPS surveys. Hyperspectral image was corrected for atmospheric and geometric factors. Spectral signatures were obtained from the imagery for Methane and Ethane which was then matched with known signatures from the digital library of NASA and field collected spectra obtained from surveys. Similar RS/GIS techniques can be used on unexplored sites.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

87

Conference 6745: Remote Sensing of Clouds and the Atmosphere


Monday-Wednesday 17-19 September 2007 Part of Proceedings of SPIE Vol. 6745 Remote Sensing of Clouds and the Atmosphere XII

6745-01, Session 1
Carbon dioxide retrieval from reflected sunlight spectra in the presence of cirrus cloud: model studies
A. I. Bril, S. Oshchepkov, T. Yokota, National Institute for Environmental Studies (Japan) The results of model study for CO2 retrievals from numerically synthesized GOSAT (Greenhouse gases Observing SATellite) observation data are presented. The GOSAT is scheduled to be launched in 2008 to monitor column amounts of CO2 and CH4. A nadir-looking Fourier-Transform Spectrometer (FTS) of Short Wavelength Infrared (SWIR, 1.6 m and 2 m) and 0.76 m oxygen A-band regions will be mounted on GOSAT. To assess CO2 sources and sinks, the monthly averaged CO2 column amounts estimated by satellite-based measurements should have a precision of within 1% or better to provide an advantage over existing ground-based measurement networks. This study focuses on CO2 retrievals in the presence of cirrus clouds. An important feature of this problem is to apply radiance data measured in several spectral channels. In particular, 1.58 m spectral band was utilized for CO2 total column amount retrievals. The cloud correction was performed using an original approach that is based on the application of the equivalence theorem with parameterization of photon path-length probability density function (PPDF). Monte Carlo simulation was used to validate this parameterization for a vertically non-homogeneous atmosphere including an aerosol layer and cirrus clouds. The PPDF parameters were estimated using nadir radiance in the oxygen A-band and in the H2O-saturated area of the 2.0 m spectral band. Application of the proposed approach was shown to provide the required accuracy in the presence of cirrus clouds with optical thickness up to 0.5 for the wide range of observation conditions.

6745-03, Session 1
Accomplishments of the atmospheric infrared sounder (AIRS) and the need for higher spatial resolution in the future
T. S. Pagano, M. T. Chahine, H. H. G. Aumann, Jet Propulsion Lab. (USA) The AIRS instrument was launched in May 2002 onboard the EOS Aqua Spacecraft into a polar sun-synchronous orbit and provides daily global observations of vertical profiles of temperature, water vapor, ozone, carbon monoxide, methane and carbon dioxide. Additional products include cloud and aerosol properties and surface temperature and emissivity. The wide range of parameters available from AIRS have made it invaluable as a tool for weather forecasting and climate modeling. AIRS has improved the 5 day forecast by 6 hours in the NCEP operational system and researchers have identified further improvement pending upgrades to computer systems at NCEP. AIRS data have been used to validate the distribution and transport of water vapor and greenhouse gases in climate models and has identified significant errors currently in the handling of these quantities. Despite the accomplishments of AIRS, we can see limitations in the AIRS due to its coarse spatial resolution that we will examine in this paper by comparing to MODIS and other observations. While IASI and CrIS are the future planned operational systems, their spatial resolution is comparable to AIRS. The enhancements higher spatial resolution will offer are improved yield in cloudy regions, improved boundary layer sensitivity, improvements in the study of surface-air interactions, and the improved sensitivity to sources and sinks of atmospheric gases. An instrument concept called the Advanced Remote-sensing Imaging Emission Spectrometer (ARIES) will address these needs by providing 1km spatial resolution hyperspectral infrared while also improving the spectral resolution of the observations. It will serve as a climate benchmark by providing global daily observations of the hyperspectral infrared spectrum using highly stable observational techniques demonstrated on AIRS

6745-02, Session 1
Ozone depletion in the Austral spring from UV micro-satellite instrument
J. A. Fernandez-Saldivar, C. I. Underwood, Univ. of Surrey (United Kingdom); S. Mackin, Surrey Satellite Technology Ltd. (United Kingdom) The Ozone Mapping Detector (OMAD) on board FASat-Bravo micro-satellite observed backscattered UV radiation to retrieve atmospheric ozone with low-spatial-resolution (150 x 150 km). This relatively coarse resolution with continuous global coverage allowed the observation of the seasonal ozone layer depletion over Antarctica in the austral spring in 1998. Previous analysis of this instrument have shown agreement in the radiances observed by OMAD and NASAs Total Ozone Mapping Spectrometer (TOMS-EP); these have even indicated the detection of an apparently higher ozone content anomaly due to a volcanic cloud of Nyamuragira volcano during its eruption in October 1998 [2]. A new improved version of the simplified algorithm used in OMAD data has been empirically developed and applied to the austral region to determine the total Ozone content. The new data processing allowed the observation of the development of the ozone depletion in 1998 from September to early December when ozone depletion normally occurs. The OMAD results showed good agreement overall when compared with those obtained from TOMS-EP despite their intrinsic instrumental differences. TOMS uses two 1-nm channels (centred around 331-nm and 360-nm), whilst OMAD uses two 10-nm channels (centred around 334-nm and 380nm). Results indicate Ozone contents lower than 150 Dobson Units (DU) in the Antarctic region with absolute errors less than 10 % in the vertical column content and high cross-correlations when compared with TOMS-EP. The value of this low-cost earth observation approach is discussed on the potential of such missions to provide additional atmospheric observations of large-scale phenomena.

6745-05, Session 1
Aerosol robotic network (AERONET) activity within the framework of the International Polar Year
A. Smirnov, Univ. of Maryland/Baltimore (USA); B. N. Holben, NASA Goddard Space Flight Ctr. (USA); N. T. ONeill, Univ. de Sherbrooke (Canada); M. V. Panchenko, Institute of Atmospheric Optics (Russia); M. Sorokin, Science Systems and Applications, Inc. (USA); P. Sobolewski, Institute of Geophysics (Poland); S. M. Sakerin, Institute of Atmospheric Optics (Russia); V. F. Radionov, Arctic and Antarctic Research Institute (Russia); R. Wagener, Brookhaven National Lab. (USA); B. R. Bojkov, Univ. of Maryland/Baltimore (USA); T. Zielinski, Institute of Oceanology (Poland); T. F. Eck, Univ. of Maryland/Baltimore (USA); A. Sinyuk, I. Slutsker, D. Giles, Science Systems and Applications, Inc. (USA) Polar regions are important for climate change studies. Transport of atmospheric pollutants to polar areas changes direct radiation forcing. Cloud formation and polar precipitation are affected by changes in remote aerosol concentrations and local emissions from the open ocean. In the paper we present Aerosol Robotic Network (AERONET) activity in the Arctic and Antarctic regions and over the oceans. AERONET made significant gains establishing sites close to or even beyond the polar circle: in Alaska, in Canada; in Spitsbergen; in Northern Europe; in Siberia, that is subject to arctic haze and summer biomass burning. We hope to expand this activity to the Northern part of Russia, Finland and Greenland. Measurements in those locations will provide a solid data set needed for climate and radiation studies. Aerosol optical depth, columnar water vapor content, aerosol volume size distribution (retrieved from sun and sky radiances) in the total column, real and imaginary parts of the refractive index, and single scattering albedo are the standard data products. Data are publicly available on the Internet (http://aeronet.gsfc.nasa.gov) in the near real-time

88

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6745: Remote Sensing of Clouds and the Atmosphere


mode. Recently hand-held sunphotometers Microtops calibrated at the Goddard Space Flight Center were successfully deployed onboard various research vessels to study aerosol optical properties over the Northern and Southern Oceans and in Antarctica. space instruments have provided a record of global stratospheric and upper tropospheric aerosols that extends over 20 years. Since the demise of SAGE II and SAGE III however, there are presently no space instruments devoted to continuing this aerosol data set. The paper will show that aerosol extinction profile, together with a moment of the size distribution, can be accurately retrieved from limb scatter measurements. The methodology will be described, and retrieval examples will be presented using data from two limb scatter instruments, namely SAGE III and OSIRIS. The retrieved extinction profiles will be compared with SAGE II and SAGE III occultation aerosol products for a series of wavelengths. It will be shown that the retrieval accuracy is good (less than 5%), with precision on the order of 25%. Once operational, the retrieval method will be applied to the data collected by the still-operating limb scatter instruments (namely OSIRIS and SCIAMACHY) in order to extend the aerosol data record into the present time. In the future, the OMPS Limb Profiler instrument, which is presently manifested on NPP (launch date = September 2009) will be used for aerosol research. The OMPS instrument, which is composed of 3 subsystems (namely a total ozone mapper, a UV ozone profiler and a limb scatter ozone profiler) will be described and its capability to retrieve aerosol vertical distribution and microphysical characteristics will be reviewed. Stratospheric and upper tropospheric aerosols play an important role in the Earth radiative budget and their continual monitoring with present and future limb scatter instruments would provide some of the required information needed by the scientific community.

6745-06, Session 1
Aerosol characterization in Morocco with AERONET and intercomparison with satellite data: TOMS, MODIS and MISR
B. Aziza, B. Zouhair, Cadi Ayyad Univ. (Morocco) The work presented in this paper focuses on aerosol characterization in the area of Morocco and the Canary Islands. Ground based measurements are essential. In our case they are materialized by the aerosol optical thickness (AOT) and are provided by the AERONET (Aerosol Robotic Network) network. To fully scan all the area, satellite measurements are certainly the most efficient way. We used TOMS (Total Ozone Mapping Spectrometer) aerosol index (AI) provided by both Earth Probe and OMI platforms. We also used the aerosol optical depths provided by MODIS (Moderate Resolution Imaging Spectroradiometer) and MISR (multiangle Imaging Spectroradiometer) instruments, onboard Terra platform. We have established linear relationships between ground measurements and satellite data. An inter-comparison of satellite data in this region was achieved.

6745-07, Session 1
Remote sensing for site selection of the European extremely large telescope (E-ELT)
H. Kurlandczyk, M. S. Sarazin, European Southern Observatory (Germany) Remotely sensed data can be of great interest for the site selection of astronomical observatories. In particular, candidate sites of the future European Extremely Large Telescope (E-ELT) of 30-60 m diameter from ESO need to be assessed and analytically compared in their observing characteristics. Parameters such as cloud cover and precipitable water vapor which are important for optical and infrared astronomical observations have been assessed with the Meris instrument on the Envisat satellite with a resolution of 1km pixel. A validation of the data was made by comparing Meris data and in situ measurement available from ESO observatories in Chile, La Silla and Paranal, combined with lower resolution values from the GOES weather satellite. A detailed analysis over the available period of 2002 to 2007 of four short listed sites both in the northern and in the southern hemisphere for the E-ELT is presented.

6745-10, Session 1
Optimal estimation applied to the retrieval of aerosol load using MSG/SEVIRI observations
S. Wagner, Y. Govaerts, A. Lattanzio, P. Watts, EUMETSAT (Germany) Using the principle of reciprocity, observations acquired by the SEVIRI radiometer on-board the Meteosat Second Generation satellites provide multi-angular and multi-spectral measurements that can be used for retrieving information on both the atmospheric aerosol load, and the Earth surface. The purpose of the presented new Land Daily Aerosol algorithm developed at EUMETSAT is to derive simultaneously the mean daily tropospheric aerosol load and the land surface properties from the SEVIRI observations. The aerosol load is calculated through the optical depth parameter, for various classes of aerosols over land surfaces, and is inferred from the inversion of a forward radiative transfer model against daily-accumulated observations in the 0.6, 0.8 and 1.6 SEVIRI bands. These daily time series provide the angular sampling used to discriminate the radiative effects that result from the surface anisotropy, from those caused by the aerosol scattering. The algorithm, based on the Optimal Estimation theory, retrieves simultaneously the surface bidirectional reflectance, and the aerosol optical thickness for various classes of aerosols. Results of comparisons with AERONET data are presented to validate the modelling approach and the algorithm that resolves the inversion problem. The retrieval error is analysed, together with the effects on the retrieval quality of updating in time the prior information.

6745-08, Session 1
Use of satellite data for astronomical site characterization
A. M. Varela, Instituto de Astrofsica de Canarias (Spain) The main goal of this work is the analysis of new approaches in order to select the best astronomical sites for hosting the future new generation of extremely large telescopes. The objective is to calibrate data provided by in situ techniques through remote sensing data retrieved from satellite-platforms, in particular, those related with the atmospheric extinction and with the troposphere winds. In addition, from these satellite data is possible to go back at the cloud coverage, water vapour, ozone or the climatic trend. The main problem to use these values is their interpretation and their quantitative calibration. Data analysis need to be complemented with those provided by in situ instruments (telescopes, airborne particles counters, ground meteorological stations, etc.).

6745-11, Session 1
Dust aerosol optical depth retrieval over desert surface, using the SEVIRI window channels
B. De Paepe, Vrije Univ. Brussel (Belgium); S. Dewitte, Royal Meteorological Institute of Belgium (Belgium) Dust aerosols have a significant impact on the earth radiation budget. The aerosol direct radiative forcing is correlated with enhanced optical thickness. Therefore it is useful to retrieve the optical thickness of dust aerosols over desert for assessing the aerosol effect. Dust aerosols have an impact in the thermal infrared wavelengths, what allows to detect aerosols over desert surfaces. To retrieve the aerosol properties over land we have to take into account the surface contribution. The surface radiation depends on the skin temperature, which is characterized by a strong diurnal variation. Therefore it is better

6745-09, Session 1
Stratospheric and upper tropospheric aerosol retrieval from limb scatter signals
D. F. Rault, NASA Langley Research Ctr. (USA); R. P. Loughman, Hampton Univ. (USA) Solar occultation observations made by the SAGE family of

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

89

Conference 6745: Remote Sensing of Clouds and the Atmosphere


to use the surface emissivity, which we assume constant over a time span of 24 hours. The surface emissivity is based on clear sky observations that are corrected for atmospheric extinction and emission. The clear sky image is a composite of pixels that are characterized by the lowest brightness temperature difference between the SEVIRI channels at 8.7 and 10.8 m. Due to the low desert surface emissivity at 8.7 m we can assume that the selected pixel values are obtained for a clear sky day. We use a forward model to simulate the thermal infrared radiation transfer in the dust layer. The apparent surface emissivity in the presence of aerosols is calculated as a function of the geometric angles, the surface emissivity, and the aerosol optical depth (AOD). This is stored in lookup tables (LUT) that are inverted to retrieve the AOD from the apparent surface emissivity. The retrieval algorithm consists of firstly: processing of the clear sky image and computation of the surface emissivity, secondly: processing of the instantaneous image and computation of the apparent surface emissivity, and thirdly: selection of the corresponding LUT and retrieval of the AOD that matches to the observed apparent surface emissivity.

6745-13, Session 2
Cloud top height estimation using simulated METEOSAT-8 radiance
R. Borde, EUMETSAT (Germany); P. Dubuisson, Univ. du Littoral Cte dOpale (France) Atmospheric Motion Vectors (AMVs) are one of the most important products generally derived from all geostationary satellites, and especially from Meteosat at EUMETSAT, because they constitute a very important part of the observation data fed to Numerical Weather Prediction. The height assignment (HA) is still the most challenging task in the AMV extraction scheme. The advent of Meteosat Second Generation provides many new opportunities for improve HA of AMVs. Indeed, existence of a CO2 absorption channel at 13.4 m on SEVIRI instrument enables to use simultaneously the IR/CO2 ratioing methodology in addition to the so-called WV-IRW intercept method for semi-transparent cases. Due to the existence of several Water Vapour and Infrared channels on SEVIRI, each method is implemented in slightly different configuration, and several pressures are then calculated for each AMV. First expectation was to use the agreement of these pressures as a quality check for the final AMV height. Unfortunately, the various methods (STC and CO2 slicing) have clearly their own sensitivity and domain of application, which makes a quality check very challenging. It appeared then necessary to define these domains of application more precisely, for a better use of these methods operationally. This paper presents such results using simulated SEVIRI radiances calculated by the FASDOM radiative transfer code. FASDOM accounts for gaseous absorption as well as cloud scattering and absorption. FASDOM can precisely consider various types of clouds, having various microphysical properties, and then provide the possibility to compare the outputs of the methods knowing precisely the input of the model, especially the pressure of the simulated cloud.

6745-12, Session 2
Experimental studies of infrared scattering and polarization properties of crystalline clouds to improve atmospheric models for remote sensing of Earths atmosphere from space
T. W. Humpherys, Utah State Univ. (USA); V. N. Ivanov, G. Yaskevich, Scientific and Production Association Typhoon (Russia); J. DeVore, A. T. Stair, Jr., Visidyne, Inc. (USA); J. Watson, The Aerospace Corp. (USA); I. Schiller, Visidyne, Inc. (USA); D. V. Chvanov, V. Privalsky, Utah State Univ. (USA) Environmental satellite sensors retrieve cloud properties by analyzing the intensity, spectral dependence, and polarization of scattered sunlight. Modes of data acquisition and sensor data retrieval algorithms should take into consideration fundamental properties of water and ice cloud particles. If such properties are not taken into account, sizable errors may occur in certain cases when interpreting Earth remote sensing data from space. We report results of recent efforts in obtaining metrologically supported experimental data for validation, correction and verification of theoretical models describing scattering and polarization of IR radiation in clouds containing particles in the crystalline phase. The mirror-like crystalline structure of high altitude cloud particles can produce intense glint features viewed from satellite sensors that may interfere with remote sensing of the Earth. By selecting polarization of radiance being observed, we can mitigate such cloud clutter, since cloud-scattered sunlight is generally polarized. Distortion of transmission through the atmosphere due to scattering, when looking through clouds, also entails modeling to correct for propagation losses. Both phenomena require that models be validated and/or refined. On-going experimental investigations are being carried out in full-size cloud chambers of the State Organization SPA Typhoon under conditions comparable with actual atmospheric conditions. A unique instrument and experimental setup has been created at the Institute of Experimental Meteorology (IEM) of SPA Typhoon. It includes the means to form clouds with prescribed characteristics, a system controlling thermodynamic and microstructural parameters of the particles formed, and an instrumentation complex, SPIN, designed for the measurement of polarization characteristics of radiation scattered by cloud particles. Preliminary data have been obtained on optical properties of populations of cloud particles consisting of crystal-columns, with modal size about 10 micrometers. Dependences have been preliminarily examined on the relative amplitude and the degree of polarization of light (with the wave length of 2.5&#61617;0.05 and 3.5&#61617;0.05 micrometers) scattered by the particles at scattering angles from 50 to 1750 , (the angle between the basic axial lines of the incident beam and the receiving optics). The data obtained can be used to develop systems and to define experiments for remote sounding and environmental monitoring of the Earth from space.

6745-14, Session 2
An assessment of the on-orbit performance of the CALIPSO wide field camera
M. C. Pitts, L. W. Thomason, Y. Hu, D. M. Winker, NASA Langley Research Ctr. (USA) The Wide Field Camera (WFC) is one of three instruments in the CALIPSO science payload, with the other two being the Cloud-Aerosol LIdar with Orthogonal Polarization (CALIOP) and the Infrared Imaging Radiometer (IIR). The WFC is a narrowband, push-broom imager that provides continuous highspatial-resolution imagery during the daylight segments of the orbit over a swath centered on the CALIOP footprint. The IFOV of each WFC pixel is approximately 125 m x 125 m when projected on the Earths surface from an orbit altitude of 705 km. The spectral band of the WFC, with a center wavelength of 645 nm and a FWHM bandwidth of 50 nm, is designed to match the Aqua MODIS instruments channel 1. The primary WFC Level 1 products are radiance and reflectance registered to an Earth-based grid centered on the CALIOP ground track. First light WFC images were acquired on May 18, 2006 and routine data acquisition began in early June 2006. An initial science assessment of the WFC on-orbit performance was conducted based on analysis of the first six months of flight data. Comparisons of the WFC measurements with the wellcalibrated Aqua MODIS channel 1 data were performed to evaluate the on-orbit radiometric performance of the WFC. Overall agreement is excellent, especially over bright targets where the WFC measurements typically agree to within a few percent of MODIS. The cloud detection threshold of the WFC has also been investigated through analysis of coincident WFC and CALIOP cloud observations. This paper provides a summary of our overall assessment of the on-orbit performance of the WFC and its cloud detection capabilities.

6745-15, Session 2
Sensitivity of passive measurements in VIS, SWIR and TIR to cirrus microphysical vertical profile: application to cloud remote sensing from MODIS
B. Marchant, G. Brogniez, J. Riedi, L. C. Labonnote, Univ. des Sciences et Technologies de Lille (France); P. Dubuisson,

90

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6745: Remote Sensing of Clouds and the Atmosphere


Univ. du Littoral Cte dOpale (France) For inferring cirrus optical and microphysical properties from satellite imagery, a common assumption is that the radiative properties of a cirrus cloud may be represented by those associated with a specific ice crystal habit, a single particle size distribution and Ice Water Content (IWC). Various algorithm have been developed to retrieve cirrus optical and microphysical properties in the past 20 years. They can be categorized into the techniques based on either thermal infrared or solar reflection measurements. However, in-situ measurements have shown that shapes, sizes and IWC of ice crystals may vary substantially with height within the clouds. Given the different sensitivity of thermal infrared and solar wavelength to cloud microphysic, it is unlikely that a single cloud layer with homogeneous cloud properties can be used to reproduce both type of measurements. Thus, it is necessary to assess the effect of vertical inhomogeneity within cirrus on the radiative transfer calculations and on the retrieval techniques. The purpose of this study is to investigate a microphysical cirrus model composed of different layers in terms of ice crystal habit, size and IWC. The vertical structure will be given by simple analytic formula derived from various prescribed physical constraints. The primary goal of this study is to determine a simple cloud model that can be used to retrieve consistent information from both solar and thermal measurements. For this purpose, we examine the sensitivity of cirrus reflectances and brightness temperature to its vertical description for a suite of MODIS (MODerate-resolution Imaging Spectroradiometer) bands spanning visible, near infrared and thermal infrared wavelengths. Results of this study are presented and potential application to remote sensing of cirrus clouds with MODIS are discussed. and clear pixels. Processing the satellite image with pixel size 6x6 km is accomplished. Clear pixels processing The method of perceptron neural network that allows to obtain the ground albedo, total (aerosol plus molecular) optical thickness, single scattering albedo and phase function parameter from input values of multiangle radiance and solar incident angle. Atmospheric parameters are determined as average for atmospheric column. The calculation of solar radiance with using the MODTRAN-3, code with taking into account multiple scattering is accomplished for neural network learning. A-priory atmospheric characteristics are input as temperature and absorbing gases profiles. All mentioned parameters were randomly varied on the base of statistical models of possible measured parameters variation. Since an input information flow is a random one (because of measurement errors, perturbation ignored and so forth) a risk estimation task appears, that is a task of confidence intervals estimation for output parameters. Apart from it in the project it is implied to carry out analysis of the neural network structure to look after optimal interactions between perceptrons. Optical thickness, single scattering albedo, asymmetry parameter, and surface albedo are retrieved and images are presented on maps Cloudy pixels processing The inverse problem solution in case of cloudy pixels is proposed to obtain on the base of analytical method elaborated by authors. The analytical methodology is based on the inverting the asymptotic formulas of the transfer theory. Only assumption about big optical thickness and weak true absorption of cloud is taken a priori. There is no such restriction as conservative scattering, spectral independence of the optical thickness, infinite optical thickness. The model of horizontally infinite layer is considered. The slight horizontal heterogeneity is approximately taken into account. Processing is accomplished for every available wavelength (in spectral ranges 443- 910nm) and pixel separately. Optical thickness, single scattering albedo and asymmetry parameter are retrieved and images are presented on maps. Optical parameters of cloud may serve for estimating water contents, aerosol pollution and for constructing an adequate optical model of natural clouds.

6745-16, Session 2
Optical characteristics of cirrus clouds at a midlatitude EARLINET station
E. Giannakaki, D. S. Balis, Aristotle Univ. of Thessaloniki (Greece); V. Amiridis, National Observatory of Athens (Greece); S. Kazadzis, Aristotle Univ. of Thessaloniki (Greece) Measurements performed with a backscatter and Raman lidar at Thessaloniki, Greece were used to characterize cirrus clouds and aerosol layers by determining their geometrical and optical properties. This is achieved through the application of different post-processing algorithms. We retrieved the cirrus clouds optical properties by using three independent mathematical methods. In the first method, an iterative procedure was used based on the criterion that forward and backward integration coincide to the desired degree of accuracy. In the second method, the optical depth of the cirrus cloud can be determined by comparing the backscattering signals just bellow and above the cloud if the lidar signals are correctly represent the scattering medium. The third method (Raman) is based on the measurement of the elastic-backscatter signal at 355nm and of the nitrogen inelastic-backscatter coefficients, independently of each other, and thus, of the extinction-to-backscatter ratio. The results are considerably influenced by multiple scattering effects, that not taken into account and this leads to a significant underestimation of calculated cirrus optical depths and lidar ratios. To estimate and correct this effect we have applied a radiative transfer model that calculates the multiple scattering contributions for each cirrus case analyzed. We have compared the resulting optical depths and lidar ratios and found a good agreement between these methods. The comparison has been performed both to the effective and real values of optical depth and lidar ratio.

6745-18, Session 2
Breaking the relations between the scales of scattering theory for main cloud components
I. N. Melnikova, St. Petersburg State Univ. (Russia) Results of the optical thickness retrieval from radiation observations in cloudy atmosphere revealed significant wavelength dependence. The values of the single scattering albedo obtained from these observations show a strong absorption into clouds within the short wavelength region and outside of atmospheric gaseous absorption bands). The single scattering albedo is from 0.95 for polluted cloud up to 0.9995 for clean clouds, which are much smaller than it is expected from Mie theory for the elementary volume of the cloud medium. Similar results were obtained also from the processing satellite POLDER radiative measurements. The optical thickness and scattering coefficient, obtained from numerous airborne and satellite radiative observations, indicate a distinct spectral dependence, namely, optical thickness are twice less at the wavelengths longer than 0.8mkm than at 0.4mkm. These facts contradict results of Mie simulations of light scattering within cloud media. The essential distinctions might be explained if turn to initial axiomatic relations between scales, which are the base for derivations of the scattering theory. Relations between basic scales are not valid within cloud media, thus the direct application of scattering theory results to cloud optics modeling appeared incorrect. The empirical relations for correction of Mie values of the scattering and absorption coefficients are proposed.

6745-17, Session 2
Retrieval of cloud optical parameters from data of reflected radiance multiangle observation
I. N. Melnikova, St. Petersburg State Univ. (Russia) and Russian Hydrometeorological Univ. (Russia); A. V. Vasilyev, St. Petersburg State Univ. (Russia); N. V. Konovalov, M.V. Keldysh Institute of Applied Mathematics (Russia) Satellite image consists from separate picture elements - pixels. During the primary processing of the satellite information they are divided to cloudy (the cloud amounts is one from the pixel characteristics) and cloud free or clear pixels. For solving the problem different methods are applied to processing cloudy

6745-20, Session 3
Applications based on ROSA onboard OCEANSAT_2 space mission
F. Vespe, V. De Cosmo, Agenzia Spaziale Italiana (Italy); A. Zin, Alenia Alcatel Space (Italy)

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

91

Conference 6745: Remote Sensing of Clouds and the Atmosphere


The Radio Occultation Sounder of the Atmosphere (ROSA) is a spaceboard GPS receiver tailored for sensing the atmosphere by applying the limb sounding technique. It was designed and developed by the Italian Space Agency (ASI) and Alenia Alcatel Space (AAS). In 2005 A Memorandum of Understanding was signed between ASI and the Indian Space Research Organization (ISRO) just to put ROSA onboard their next OCEANSAT_2 space mission devoted to Earths remote sensing. The objectives of the activities we plan to perform is twofold. On one hand a detailed analysis has been performed to arrange the antenna pointing toward the anti-velocity direction. In particular a simulation has been performed to understand what are the effects of multipath on the antenna. So we plan to investigate the sensitiveness of the atmosphere profiles, namely pressure, temperature and humidity, to the multipath disturbances suffered by ROSA. On an ather hand we plan to discuss the possible benefits that such space mission will bring in the field of Atmospheric Remote Sensing

6745-23, Session 3
Validation of a high resolution circulation model by lidar measurements of water vapour mixing ratio profiles in the area of Naples
F. Maria Grazia, Univ. Federico II of Naples (Italy); F. Rossella, Univ. degli Studi dellAquila (Italy); P. Gianluca, S. Nicola, Univ. Federico II of Naples (Italy); W. Xuan, Consiglio Nazionale delle Ricerche (Italy) During August-October 2005, several lidar measurements have been performed at many Italian lidar stations during the Launch (International Lindenberg campaign for Assessment of hUmidity aNd Cloud profiling systems and its impact on High-resolution modelling) campaign according with IOP (Intensive Operational Period) schedule. IOP 5 episode (11-12 October 2005) has been selected and investigated with a special attention to the variability of water vapour mixing ratio vertical distribution to better understand the associated circulation in the coastal waters of the Gulf of Naples (Tyrrhenian Sea). A model simulation is performed using MM5V3 model from PSU/NCAR (Grell et al., 1994 and Dudhia, 1993) for this study. The MM5 is a no-hydrostatic model at the primitive equations with a sigma terrain following vertical coordinate. The model has multiple-nesting capabilities to enhance the resolution over the area of interest: the finer grid size used over the Gulf of Naples is 3km. Water vapour mixing ratio observations have been detected in Naples site (4050N, 1411E, 118 m asl) by a water vapour Raman lidar which is also able to measure the backscatter profile at 355 and 532 nm and the extinction profile at 355 nm. Water vapour mixing ratio vertical profiles retrieved by lidar have been compared with the MM5 water vapour mixing ratio vertical profiles. The agreement between the model and the Lidar data is good in the range 1000-4000 m, though the model seems to be not able to capture the spatial sharp variations of water vapour mixing ratio profiles retrieved by lidar.

6745-21, Session 3
A proposed solar UV-B radiometer calibration method: trial using diffuse solar irradiance
S. Takeshita, M. Sasaki, Tokai Univ. (Japan) A solar ultraviolet-B (UV-B) radiometer is generally calibrated against standard equipment using global solar irradiance. A calibration constant of the solar UV-B radiometer is highly depending on the weather conditions and the seasons when calibration is conducted. This phenomenon is caused by the difference in the spectral, cosine, azimuth response and temperature characteristic of the radiometer between tested equipment and standard. An ideal weather condition for calibration is cloudless day, however a quite few days, ten days or so, are available in a year in central part of Japan. According our previous research, the spectral response is a dominant factor for determination of the calibration constant of the UV-B radiometer, and cosine and azimuth response is a minor. Moreover, diffuse UV-B irradiance reaching the earths surface is less sensitive to cloud than the direct one. We tried to use diffuse solar irradiance for calibration of solar UV-B radiometers. Evaluated solar UV-B radiometer was a MS-210W (EKO), and the standard equipment was a same type of the UV-B radiometer. To remove direct solar irradiance, a shading band (MB-11, EKO) was used. Both solar UV-B radiometers with shading band were set horizontally. The experiment using pyranometer (MS-801, EKO) with a shading band was also conducted as a control experiment. By using diffuse solar irradiance for calibration, the small variation in a calibration constant of the solar UV-B radiometer was obtained. The comparison between the solar UV-B radiometer and the pyranometer shows that variation in a calibration constant of the solar UV-B radiometer is larger than that of the pyranometer.

6745-24, Session 3
A European research infrastructure for aerosol study on a continental scale: EARLINET-ASOS
A. Amodeo, G. Pappalardo, Consiglio Nazionale delle Ricerche (Italy); J. Bsenberg, Max-Planck-Institut fr Meteorologie (Germany); A. Ansmann, Leibniz-Institut fr Troposphrenforschung e.V. (Germany); A. Apituley, Rijksinstituut voor Volksgezondheid en Milieu (Netherlands); L. Alados-Arboledas, Univ. de Granada (Spain); D. S. Balis, Aristotle Univ. of Thessaloniki (Greece); C. Bckmann, Univ. Potsdam (Germany); A. P. Chaikovsky, B.I. Stepanov Institute of Physics (Belarus); A. Comeron, Univ. Politecnica de Catalua (Spain); V. Freudenthaler, Ludwig-Maximilians-Univ. Mnchen (Germany); G. H. Hansen, Norwegian Institute for Air Research (Norway); V. Mitev, Observatoire Cantonal de Neuchtel (Switzerland); D. Nicolae, National Institute of Research & Development for Optoelectronics (Romania); A. D. Papayannis, National Technical Univ. of Athens (Greece); M. R. Perrone, Univ. degli Studi di Lecce (Italy); A. Pietruczuk, Institute of Geophysics (Poland); M. Pujadas, Ctr. de Investigaciones Energticas, Medioambientales y Tecnolgicas (Spain); J. Putaud, Joint Research Ctr. (Italy); F. Ravetta, Univ. Pierre et Marie Curie (France); V. Rizi, Univ. degli Studi dellAquila (Italy); V. B. Simeonov, Ecole Polytechnique Fdrale de Lausanne (Switzerland); N. Spinelli, Univ. degli Studi di Napoli Federico II (Italy); D. V. Stoyanov, Institute of Electronics (Bulgaria); T. Trickl, Forschungszentrum Karlsruhe (Germany); M. Wiegner, Ludwig-Maximilians-Univ. Mnchen (Germany) The present knowledge of the aerosol distribution is not sufficient to estimate the aerosol influence on global and regional environmental conditions and climate. This observational gap can be closed by using advanced laser remote sensing. The improvement of the aerosol observation system will contribute to almost all areas of societal benefits listed in the GEOSS (Global Earth Observation System of Systems) Implementation Plan. EARLINET (European Aerosol Research Lidar Network) is the first aerosol lidar network, established in 2000, with the main

6745-22, Session 3
Long-term monitoring of layering of lower atmosphere in urban environment by ceilometer
K. Schfer, S. M. Emeis, C. Jahn, C. Mnsterer, Forschungszentrum Karlsruhe (Germany); C. Mnkel, Vaisala GmbH (Germany) Automatic monitoring of lower atmosphere layering was performed by continuous remote sensing measurements with ceilometer and SODAR in Augsburg. The Vaisala ceilometers LD40 and CL31 were used which are eye-safe commercial lidar systems. Special software for these ceilometers provide routine retrievals of lower atmosphere layering from vertical profiles of laser backscatter data. A comparison was performed with parallel monitoring by a SODAR. From the acoustic and optical remote sensing data the following features are analysed. The SODAR measurements provide the height of a turbulent layer characterized by high acoustic backscatter intensities due to thermal fluctuations and a high variance of the vertical velocity component. Different examples of meteorological conditions will be discussed to demonstrate the possibilities of atmospheric layering retrieval.

92

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6745: Remote Sensing of Clouds and the Atmosphere


goal to provide a comprehensive, quantitative, and statistically significant database for the aerosol distribution on a continental scale. EARLINET is a coordinated network of European stations (25 at present) using advanced lidar methods for the vertical profiling of aerosols. The network activity is based on simultaneous scheduled measurements, a rigorous quality assurance program addressing both instruments and evaluation algorithms, and a standardised data exchange format. Further observations are performed to monitor special events (Saharan dust outbreaks, forest fires, photochemical smog and volcano eruptions). EARLINET-ASOS (Advanced Sustainable Observation System) is a five year EC Project started in 2006, based on the EARLINET infrastructure. The main objectives are: to make EARLINET a world-leading instrument for the observation of the 4-D aerosol distribution on continental scale; to foster aerosol-related process studies, validation of satellite sensors, model development and validation, assimilation of aerosol data into operational models; and to build a comprehensive climatology of the aerosol distribution Acknowledgments The financial support by the European Commission under grant RICA-025991 EARLINETASOS is gratefully acknowledged. Spectroradiometer (MODIS) onboard the NASA Terra and Aqua satellites. Also, we utilize (to some degree) data from the Multiangle Imaging SpectroRadiometer (MISR) onboard Terra, and aerosol data from some other sensors. The main fusion algorithm uses the maximum likelihood technique to merge the pixel values where available, and then the Optimal Interpolation method to fill the remaining gaps. The web-based environment for this data fusion is the NASA Goddard Interactive Online Visualization and Analysis Infrastructure (Giovanni) that provides this data fusion option in addition to other popular online multi-sensor intercomparison capabilities. Giovanni (http://giovanni.gsfc.nasa.gov) allows convenient data manipulation and analysis without downloading large volumes of differently formatted data from different sensors. Various statistical aspects of data fusion are addressed, including sensitivity of the fused data properties to spatial gaps in the original daily images, and spatial and temporal sampling biases.

6745-27, Session 3
Statistical approach to validation of satellite atmospheric retrievals
N. S. Pougatchev, G. E. Bingham, Utah State Univ. (USA); K. St. Germain, D. Seidel, National Oceanic and Atmospheric Administration (USA); F. H. Berger, Deutscher Wetterdienst (Germany) A mathematical model for statistical estimate of the bias and noise in satellite retrievals of atmospheric profiles and a case study are presented. The model allows accurate validation of actual performance of the remote sensing system while in orbit by comparing its measurements to some relevant - validating data sets, e. g., radiosonde network. The following factors are accounted by the model: (i) The satellite and validating systems sample volumes of the atmosphere at times and locations that are not exactly co-located. (i) The true atmospheric states may be correlated or not, e.g. validation against historical records. (iii) The validated and validating systems have different characteristics, i.e., vertical resolution and noise level. All the above factors cause apparent difference between the data to be compared. The presented model makes the comparison accurate by estimating the differences. To demonstrate its practicability we present the case study that involves the radiosonde data from three stations: ARM Tropical Western Pacific (0.5O S, 167O E), ARM Southern Great Planes (37O N, 98O W), and Lindenberg (52O N, 14O E). For each station we considered temperature profile validation scenario and estimated associated errors. The model can be used for assessment and interpretation of the validation results when the above mentioned sources of discrepancies are significant, as well as for evaluation of validation data sources, e.g., GRUAN (GCOS Reference UpperAir Network).

6745-25, Session 3
Improvement of broadband radiance to flux conversion by using the synergy between active and passive remote sensing instruments
C. Domenech, E. Lopez-Baeza, Univ. de Valncia (Spain); D. Donovan, Koninklijk Nederlands Meteorologisch Instituut (Netherlands); M. Bouvet, European Space Agency (Netherlands); H. Barker, Meteorological Service of Canada (Canada) The ESA EarthCARE mission includes the BBR (BroadBand Radiometer), the instrument responsible to provide measurements of broadband radiances over the along-track satellite path. The BBR footprint will be geolocated in space and time with the passive sensor, MSI (Multi-Spectral Imager), and the active sensors, ATLID (ATmospheric LIDar) and CPR (Cloud Profiler Radar) onboard the same platform. The role of the BBR was defined to provide the boundary condition for top of atmosphere flux densities. Thus, the radiance to flux conversion is the main objective for the BBR retrieval algorithms. This conversion has been so far carried out by using specific angular distribution models (ADMs). In this process, every radiance is classified in a unique scene bin of observations characterized by a similar anisotropic behaviour. Each of these scene bins is defined by a range of values distinguishable by the MSI. But MSI can only extract vertically integrated retrievals. Therefore, in multi-layer cloud configurations, scene identification (ID) by means of the MSI retrievals will not distinguish the 3-D structure of the real scenes. These scenes will thus be wrongly identified. Since active sensors are present on the same satellite platform, it would be possible to use their observations to contribute to the BBR scene ID. This work shows a preliminary simulation approach to demonstrate the advantages of this methodology by applying it to multi-layer clouds. The clouds have been built with a stochastic cloud generator model, and the radiative transfer simulations have been carried out with the EarthCARE Simulator, a Monte-Carlo code capable to reproduce the observations of the different mission instruments taking into account the specific characteristics of each sensor.

6745-28, Session 3
The GERB Edition 1 products SEVIRI scene identification
A. Ipe, C. P. Bertrand, N. Clerbaux, S. Dewitte, L. Gonzalez, Royal Meteorological Institute of Belgium (Belgium) The first Geostationary Earth Radiation Budget (GERB) instrument was launched during the 2002 summer together with the Spinning Enhanced Visible and InfraRed Imager (SEVIRI) on board of the Meteosat-8 satellite. This broadband radiometer aims to deliver near real-time estimates of the top of the atmosphere solar and thermal radiative fluxes at high temporal resolution thanks to the geostationary orbit. Such goal is achieved with the L20 GERB processing which generates these fluxes at several spatial resolutions from the directional filtered radiance measurements of the instrument. This processing consists of successive components, one of them being a radiance-to-flux conversion. Such conversion is carried out in the solar region by using the shortwave angular dependency models (ADMs) developed from the Tropical Rainfall Measuring Mission (TRMM) Clouds and the Earths Radiant Energy System (CERES) experiment. As these ADMs are stratified according to specific scene properties, the GERB ground segment will

6745-26, Session 3
Multi-sensor aerosol data fusion: can it be done online?
G. Leptoukh, NASA Goddard Space Flight Ctr. (USA) To increase spatial coverage and produce complete aerosol fields for comparison with models, and facilitate better tracking of aerosol events, aerosol measurements by different sensors can be fused or merged together. In this paper, we focus on merging daily mean Aerosol Optical Thickness (AOT), as measured by the Moderate Resolution Imaging

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

93

Conference 6745: Remote Sensing of Clouds and the Atmosphere


have to rely on a scene identification of SEVIRI data which allows us to select the proper ADM. In this paper, we will briefly justify and describe the implementation of a specific GERB scene identification for the official Edition 1 release of the L20 products. Comparisons between GERB and CERES scene identifications both applied to SEVIRI data will follow. Finally, we will suggest possible improvements based on limitations which could be found.

6745-52, Poster Session


Study on methods of cloud identification and data recovery for MODIS data
X. Wu, Q. Cheng, Zhejiang Gongshang Univ. (China) MODIS data have great potential in rice growth monitoring and yield estimation due to the low cost and high time resolution. Unfortunately, MODIS, which is a kind of visible-infrared sensor, cannot detect land surface through cloud, and cloud-free image is quite rare during rice growth period due to cloudy weather. Therefore, cloud contamination is one of the main obstacles in rice growth monitoring and yield estimation using MODIS data. Based on spectral characteristics of cloud and MODIS channels, taking it into account that MODIS data includes thirty-six bands, especially the infrared channels subdivided, it has realized cloud detection in MODIS images by multi-spectral synthesis method, infrared difference algorithm, index methods and cloud detection index in this paper. The result shows that infrared difference algorithm, index methods analysis are the simple and effective methods to detect cloud. After geometric correction, the cloud-free images are obtained through interpolating using time series MODIS data and ratio value using same date data of different year.

6745-29, Session 3
Integrated cloud-aerosol-radiation product using CERES, MODIS, CALIPSO and CloudSat data
S. Sun-Mack, Science Applications International Corp. (USA); B. A. Wielicki, P. Minnis, NASA Langley Research Ctr. (USA); S. Gibson, Y. Chen, Science Applications International Corp. (USA) The goal of this paper is to provide the first integrated data set of global vertical profiles of aerosols, clouds, and radiation using the combined NASA A-Train data from Aqua CERES and MODIS, CALIPSO, and CloudSat. All of these instruments are flying in formation as part of the Aqua Train, or A-Train. This paper will present the preliminary results of merging aerosol and cloud data from the CALIPSO active lidar, cloud data from CloudSat, integrated column aerosol and cloud data from the MODIS CERES analyses, and surface and top-of-atmosphere broadband radiation fluxes from CERES. These new data will provide unprecedented ability to test and improve global cloud and aerosol models, to investigate aerosol direct and indirect radiative forcing, and to validate the accuracy of global aerosol, cloud, and radiation data sets especially in polar regions and for multi-layered cloud conditions.

6745-53, Poster Session


Joint airborne IASI validation experiment (JAIVEx) cirrus cloud property intercomparison
D. H. DeSlover, R. E. Holz, H. E. Revercomb, D. H. Turner, Univ. of Wisconsin/Madison (USA); D. K. Zhou, NASA Langley Research Ctr. (USA); W. L. Smith, Sr., Hampton Univ. (USA) The study of cirrus clouds using high spectral resolution infrared measurements provides the spectral detail to independently determine a number of cirrus cloud optical properties, including optical depth, number density, and bulk ice crystal effective radius. We will compare results from various methods designed to retrieve cirrus optical properties from the infrared atmospheric emission signature. Emphasis will be placed on the Joint Airborne IASI Validation Experiment (JAIVEx), which is scheduled for April 2007; and is the first coordinated validation campaign for the Infrared Atmospheric Sounding Interferometer (IASI). A pair of aircraft will be flying to support this mission: the NASA WB-57, carrying the NPOESS Airborne Sounder Testbed (Interferometer and Microwave, NAST-I and NAST-M, respectively) and Scanning High-resolution Interferometer Sounder (S-HIS); and the Facility for Airborne Atmospheric Measurements (FAAM) BAe 146-301 aircraft, capable of providing a full suite of cloud physics instrumentation for insitu measurements of aerosols, liquid water and ice particles. Operational satellite-based cloud products (e.g., MODIS and GOES) will also be included for comparison.

6745-30, Session 3
The slow mode of the CERES scanning radiometers
Z. P. Szewczyk, Science Applications International Corp. (USA) The CERES scanning radiometer is designed to measure accurately the solar radiation reflected by the Earth and thermal radiation emitted by the Earth. There are two CERES instruments aboard the Terra spacecraft and two aboard the Aqua spacecraft. The instruments have a response time of 8 to 9 ms. In addition there is a spurious slow mode of response at a low level to radiance input caused by heating up of the plate a detector is mounted on. This slow mode is eliminated by use of a numerical filter whose parameters are characterized by the time response and magnitude. The time response is an order of magnitude longer than the time between sampling and the magnitude is in the order of 3% of the signal. These parameters of the slow mode had been computed using ground calibration data. However, a scan analysis using in-flight data indicates that the slow mode effect has not been entirely eliminated. Its effect on the instrument response is dependent upon the value of the viewing zenith angle, and it is shown to be about 0.4%. This warrants recalculating the value of the parameters with a goal of reducing the slow mode effect to 0.1%. Measurements taken while looking at space and at the internal calibration system provide the required information. By minimizing the difference between the known signal and instrument response, both parameters can be found and applied in the numerical filter. The results are validated using data taken while scanning over the Earth in both directions (left to right and right to left) for at least one month of data at a time. It is shown that the goal of reducing the effects of the slow mode can be achieved for measurements at viewing zenith angles less than 70o. It is also demonstrated that reducing the effect of the slow mode has significant impact on science data products in the range of 2-3Wm-2 for the longwave flux. These effects would be even larger when the radiometer scans from space to Earth or from a warm clear scene to a cold cloud top if the slow mode effect is not eliminated. It is worth mentioning that the work reported in this paper is a part of revision of CERES data products for Edition 3.

6745-54, Poster Session


Cloud climatology in the Canary Islands region using NOAA-AVHRR data
A. Gonzlez, A. Cerdea, J. C. Prez, A. Daz, Univ. de La Laguna (Spain) In this work a threshold technique for cloud detection and classification is applied to 12 years NOAA-AVHRR imagery in order to obtain a cloud climatology in the Canary Islands region (North Atlantic Ocean). Once the clouds are classified, a retrieval method is used to estimate cloud macro- and micro-physical parameters, such as, effective particle size, optical thickness and top temperature. This retrieval method is based on the inversion of a numerical radiative transfer model, based on libRadtran, using artificial neural networks (ANNs). The ANNs, whose architecture follows a Multilayer Perceptron model, were trained with simulated theoretical radiances using backpropagation with momentum method, and their architectures were optimized through genetic lgorithms. The global procedure was performed for both day and night overpasses and, from a set of more than 9000 images, maps of relative frequency were calculated. These results were compared with ISCCP data for the same period. The relationships between the retrieved cloud properties and some climate and atmospheric parameters were also studied. For example, five day back trajectories have been computed

94

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6745: Remote Sensing of Clouds and the Atmosphere


using analyses from the NCEP/NCAR Reanalysis data and were classified in four types depending on the origin of the air parcel. Furthermore, other variables, like galactic cosmic rays flux or North Atlantic Oscillation (NAO) index, were considered.

6745-57, Poster Session


Retrieval of spectral emissivity and surface temperature from airborne infrared hyperspectral sensor, using neural network and spectral smoothness approaches
V. Achard, S. Lesage, L. Poutier, ONERA (France) Infrared hyperspectral imagery gives new opportunities for night observations for military, or security purposes, and for geological studies as rocks have specific infrared absorption bands. Generally, an optimized utilization of spectral information requires to retrieve spectral emissivity, which involves atmospheric compensation and surface temperature and emissivity separation (TES). This paper presents a new method dedicated to airborne hyperspectral sensor that operates in the 3-12 m spectral range, at about 2 km height. It couples neural networks in order to characterize parameters required for atmospheric compensation and a spectral smoothness approach for TES. The network training is performed with radiance spectra simulated with MODTRAN4, and using ASTER emissivities, and the TIGR atmospheric data bank. A sensibility study is carried out in order to specify which atmospheric parameters are needed and how accurate they must be. The impact of various instrumental characteristics such as signal to noise ratio, absolute calibration or spectral calibration, is also studied.

6745-55, Poster Session


Automatic detection of clouds and shadows on SPOT/HRVIR images: application to the images of the African Monsoon Multidisciplinary Analysis (AMMA) project
C. Andr, Institut Pierre Simon Laplace (France); S. Le Hgarat-Mascle, Univ. Paris-Sud (France) In this study, we propose an automatic approach for detecting clouds, cloud shadows and mist present on optical remote sensing images such as SPOT/HRVIR ones. This detection and masking is necessary for land studies from remote sensing data, such as land cover classification and vegetation monitoring. The proposed algorithm is automatic. It is parameterized by only three parameters that are rather robust since their value was kept identical for the processing of all the 39 SPOT/HRVIR images considered (AMMA, African Monsoon Multidisciplinary Analysis, database) that corresponds to various acquisition conditions, seasons, and landscapes. Relatively to the classical threshold techniques, our method makes use of three principal cloud/shadow features that allow the reduction of the false alarm rate: - Clouds can be viewed as connex objects (at least for detection), - Each cloud generates a shadow with similar shape and area, - Knowing the acquisition geometry and the sun position, the relative position of a cloud and its shadow in the image can be known but one parameter (the cloud altitude) The first feature allows the use of hysteresis threshold techniques or growing region. The second and the third feature allow the mutual validation of cloud and shadow detections. The global algorithm includes then three main steps: first, only clouds are detected based on the intercalibration of the MIR/ Green frequency images, and a hysteresis threshold operation of a MIR/Green linear combination image. Second shadows are estimated using MIR image, spatial location estimation from first cloud detection, and growing region technique. Finally, mutual validation is performed from cloud and shadow respective shapes.

6745-58, Poster Session


A semianalytic Monte Carlo code for modelling lidar measurements
E. Palazzi, Consiglio Nazionale delle Ricerche (Italy); I. K. Kostadinov, Consiglio Nazionale delle Ricerche (Italy) and Bulgarian Academy of Science (Bulgaria); F. Ravegnani, A. Petritoli, Consiglio Nazionale delle Ricerche (Italy); D. Bortoli, Consiglio Nazionale delle Ricerche (Italy) and Geophysics Ctr. of Evora (Portugal); G. Giovanelli, Consiglio Nazionale delle Ricerche (Italy) LIDAR (LIght Detection and Ranging) is an optical active remote sensing technology with many applications in atmospheric physics. It can be used to measure densities of certain constituents of the middle and upper atmosphere, to calculate temperatures, to measure wind speed and to provide information about vertical distribution of the aerosol particles. The development of techniques for modelling LIDAR measurements is needed to evaluate the effects of various environmental variables and scenarios as well as of different measurements geometries and instrumental characteristics. A Monte Carlo simulation model can provide an excellent answer to these important requirements. A semianalytic Monte Carlo code for modelling LIDAR measurements has been developed by the Energy Transfer & Minor Gases in the Atmosphere (ETMGA) group of ISAC-CNR. In this radiative transfer model, Monte Carlo calculations of contributions to the total energy received in a LIDAR system have been carried out taking into account the main components of the atmospheric environment and the laser radiation interaction processes of single/multiple scattering, absorption and ground reflection. Independent sources and receiving optical systems with assigned field of view are assumed. To enhance the efficiency of the Monte Carlo simulation, analytical estimates and expected value calculations are performed. Artificial devices (such as local forced collision and splitting) are moreover foreseen by the code, which can enable the user to drastically reduce the variance of the calculation. In this work the developed Monte Carlo model is described and a number of simulations are presented and discussed varying the geometric treatment of the atmosphere, its physical characteristics, the representation of the laser source and the telescope and a number of other meaningful parameters.

6745-56, Poster Session


Cloud factor estimation for insolation using satellite data
K. Han, J. Yeom, Y. Kim, Y. Shu, Pukyong National Univ. (South Korea) Surface Solar Insolation (SSI) is important parameter to interpret ocean-atmospheric interaction, climate change, surface heat flux, and earth radiation budget. A successful retrieval of SSI highly depends on how to describe the cloud attenuation since most of clouds have larger spatial and temporal variability and complicated physical character. Moreover, the accuracy of SSI estimation for cloudy condition is substantially lower than for clear sky. This study aims to generate a neural network-based cloud factor retrieval system, which can improve accuracy of SSI estimation for cloudy condition. In this study, multilayer feed-forward (MLF) neural network (NN) was employed with Levenberg-Marquardt back-propagation (LM-BP) and early stopping method to avoid the over-fitting. The number of hidden nodes was determined by using trial and error method since too complicated network was apt to be over-fitting, while a too simple network structure will have difficult training the network. The validation of the estimated SSI using NN-based cloud factor was performed with pyranometer measurement data obtained from 22 meteorological stations over Korea peninsula. This SSI estimation for cloudy condition showed a good agreement with ground-based measurements (RMSE = 66.0 W/m2). This accuracy indicates that the use of NN-based cloud factor leads an improvement for SSI estimation in comparison with use of previous system of cloud factor (RMSE = 78.03 W/m2).

6745-59, Poster Session


Studying clouds and the Earths radiation balance using CERES and MISR data products
N. A. Ritchey, Science Systems and Applications, Inc. (USA) and NASA Langley Research Ctr. (USA); M. T. Ferebee, NASA Langley Research Ctr. (USA)

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

95

Conference 6745: Remote Sensing of Clouds and the Atmosphere


The Atmospheric Science Data Center (ASDC) at NASA Langley Research Center processes, archives, and distributes data relating to the Earths radiation budget, clouds, aerosols, and tropospheric chemistry. The Clouds and the Earths Radiant Energy System (CERES) and Multi-angle Imaging SpectroRadio-meter (MISR) instruments are key instruments in obtaining and producing several data products that provide global observations of radiation at the top-of-atmosphere (TOA), surface, and through the atmosphere, and cloud properties. The CERES experiment is one of the highest priority scientific satellite instruments developed for NASAs Earth Observing System (EOS). The first CERES instrument was launched in November 1997 on the Tropical Rainfall Measuring Mission Satellite. CERES instruments are also currently flying on the EOS Terra and Aqua satellites. These sets of instruments provide long-term, overlapping radiation budget parameters including broadband shortwave (SW), longwave (LW) and net radiative fluxes for clear sky and cloudy conditions. The MISR instrument was built for NASA by the Jet Propulsion Laboratory (JPL) in Pasadena, California. It was launched into polar orbit on board the Terra satellite from Vandenberg Air Force Base on December 18, 1999. MISR measurements are designed to improve our understanding of the Earths environment and climate with data collected using nine cameras, at different angles and four wavelengths. Data products from MISR contain TOA radiance, aerosol, and cloud properties from regional and global scale environments. The Atmospheric Science Data Center provides data access, services and tools for over 40 projects, including CERES and MISR, in the discipline areas of Earths radiation budget, clouds, aerosols and tropospheric chemistry. Additional information is available from our web site, http://eosweb.larc.nasa.gov. The spectrally-resolved characterization of the atmospheric emission has a fundamental role in the study of the Earth radiation balance, and only a measurement performed in a wide spectral range enables us to separate the contributions to the radiative balance due to the different altitudes, constituents and physical phases. The REFIR-PAD (Radiation Explorer in the Far InfraRed Prototype for Application and Development) Fourier transform spectroradiometer can perform a characterization of the broadband radiative signature of clouds and aerosols, with the only limitation of the need of low levels of water vapour like those that are obtainable in high altitude stations during winter. The capabilities of this kind of measurement have been assessed in a series of test campaign performed in Tuscany during the winter of 2006/2007 during which atmospheric emission spectra have been acquired in various transparency conditions and the evidence of transparency in the far-infrared region below 600 cm-1 has been demonstrated. REFIR-PAD operates in the spectral range extending from 100 to 1400 cm-1 with a resolution of 0.5 cm-1, using roomtemperature detectors and optics and a compact, misalignment-compensated design. The instrument, developed at CNR-IFAC, Florence, has been successfully deployed in several campaigns, both in the ground based zenith-looking geometry and in the nadir-looking balloon borne configuration. The operating spectral range of the REFIR-PAD spectroradiometer encompasses great part of the atmospheric emission spectrum, from the relatively unexplored far-infrared region below 400 cm-1, dominated by water vapour rotational band, to the atmospheric transparency window, where a number of atmospheric instruments are already operating and can provide intercomparison data.

6745-60, Poster Session


Evaluation of adjacency effect for MIVIS airborne images
C. Bassani, R. M. Cavalli, Istituto sullInquinamento Atmosferico (Italy); S. Pignatti, Istituto di Metodologie per lAnalisi Ambientale (Italy); F. Santini, Istituto sullInquinamento Atmosferico (Italy) This work is aimed to atmospherically correct remote sensing data in the solar spectral domain (Visible and Near Infrared) allowing the better assessment of the surface spectral material characteristics. This was obtained by the inversion of the radiative transfer equation for at-sensor signal. In order to detect targets with peculiar spectral characteristics, the atmospheric correction has to take into account the diffuse radiation that constitutes a significant component to the at sensor radiance. The effect of this component (namely adjacency effect), which tends to mask the pixel seen by the sensor, derives principally from the atmospheric scattering due to the aerosol loading in the scene. At this purpose an algorithm, based on 6S calculation, was defined to derive the direct and diffuse component of radiation required to determine the contribution to the pixel reflectance related to the surrounding pixels. The developed algorithm allowed the assessment of this environmental contribution besides the pixel reflectance. This work shows how adjacency effect has a significant role in the correction of remote sensing data, especially if acquired by an airborne hyperspectral sensor like MIVIS. The preliminary analysis of results have highlighted that the adjacency effect is not negligible, mainly when pixels in the scene are spectrally heterogeneous. Such application, on MIVIS scenes, leads to obtain accurate pixel reflectance if compared with ground measurements acquired within testing areas. The analysis of the results shows the accuracy of the process in retrieving the reflectance of different surface materials.

6745-62, Poster Session


Extension of Chandrasekhars formula to a homogeneous non-Lambertian surface and comparison with the 6S formulation
A. Sei, Northrop Grumman Space Technology (USA) The classical Chandrasekhars formula relating the surface reflectance to the top of the atmosphere radiance rigorously applies to a Lambertian surface. For a homogeneous nonLambertian surface in a plane-parallel atmosphere, an extension of this formula was proposed in the 1980s and has been recently implemented in the second simulation of the satellite signal in the solar spectrum (6S) algorithm. To analyze this extension, the rigorous formula of the top of the atmosphere signal is derived in a plane-parallel atmosphere bounded by a homogeneous non-Lambertian surface. Then the 6S algorithm extension is compared with the exact formula and approximations and their validity are pointed out. The methods used for the derivation of the exact formula are classical. They are based on the separation of direct and diffuse components of the radiation fields, on the introduction of the Greens function of the problem, and on integrations of boundary values of the radiation fields with the Greens function.

6745-63, Poster Session


Retrieval of atmospheric water content based on AISA data
Q. Cheng, Zhejiang Gongshang Univ. (China) Many applications require the elimination of atmospheric effects caused by molecular, particulate scattering and atmospheric water content factors, a process known as atmospheric correction, compensation, or removal. Retrieval of atmospheric water content from remote sensing data is very important for atmospheric correction. This paper introduces and derives the algorithm about retrieving water content of atmosphere from AISA (Airborne Imaging Spectrometer for Application) data based on Modtran code. AISA sensor is put on the aviation platform covering the East Sea in China on July 29, 2005. AISA is Hyperspectral Imaging sensor including 61 bands from visible to NIR, and band 56 and band 59 are in water absorbed region and band 51 is in atmospheric window. So the three bands were used to in following expressions. The authors utilized 2 near-IR water vapor channels in addition to existing windows channels, it will be possible to derive the total column water vapor amount from AISA data in Eastern Sea in China and

6745-61, Poster Session


Far-infrared spectrally resolved broadband emission of the atmosphere from Monte Morello and Monte Gomito, near Florence
G. Bianchini, L. Palchetti, A. Baglioni, S. Del Bianco, Istituto di Fisica Applicata Nello Carrara (Italy)

96

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6745: Remote Sensing of Clouds and the Atmosphere


analyze the retrieving result. Finally, the relationships between the transmittance and atmospheric water content of AISA was computed and analyzed. The maximum atmospheric water content retrieved from AISA image is 1.43 cm and the minimum water content is 0.04 cm, and average atmospheric water content is 0.40 cm. The results are consistent with the real conditions, so the methods are feasible for retrieval of atmospheric water contents from AISA data.

6745-66, Poster Session


Fine mode aerosols on a global scale
I. Sano, S. Mukai, Kinki Univ. (Japan); M. Mukai, The Univ. of Tokyo (Japan) Fine mode aerosols influence the microphysical and optical properties of clouds, such as number and size of cloud particles, which may relate the weather change and also climate change on a global scale. This work intends to extract the amount of aerosols in the atmosphere as spectral aerosol optical thickness on a global scale. In order to cover global scale, this work uses satellite data for monitoring the fine mode particles. The dataset of Polarization and Directionality of Earth Reflectances (POLDER) mounted on ADEOS-1, and -2 (Advance Earth Observing Satellite) are adopted. POLDER measures the spectral upwelling radiances (as Stokes components I, Q, and U) at the top of the atmosphere. Therefore, satellite data consists of both atmospheric scattering and ground surface reflection. In general, the latter is larger than the former over the land. This fact, in addition to the uncertainty of the surface reflection behavior, is a reason why the extraction of aerosols over land is very difficult compared to that over ocean. It is known that polarization of ground reflectance is smaller than that of atmospheric scattering. This advantage can be used to retrieve the aerosol information over land area, and hence aerosol optical thickness (AOT) and Angstrom exponent have been already retrieved by using polarization data taken from POLDER based on the relationship between red (0.67 m) and near infrared (0.86 m) wavelengths in the previous work. An algorithm based on the near UV and violet data, which are provided by GLI sensor on ADEOS-2, is available to solve this problem. Furthermore the results of aerosol transported model AGCM are useful for the aerosol type selection for our aerosol retrieval. As results, aerosol distribution, especially fine mode particles, is precisely obtained on a global scale.

6745-64, Poster Session


An aerosol optical thickness retrieval algorithm for MSG data over land: applications in the Mediterranean area
L. Guerrieri, S. Corradini, S. Pugnaghi, R. Santangelo, Univ. degli Studi di Modena e Reggio Emilia (Italy) Meteosat Second Generation (MSG), the new generation of European geostationary meteorological satellites, is a significantly enhanced follow-on system of the previous generation of Meteosat. The Spinning Enhanced Visible and Infra Red Imager (SEVIRI) is the radiometer aboard MSG collecting, each 15 minutes, images of the underneath part of the globe in 12 spectral bands with a spatial resolution of 3 km. In this paper an algorithm retrieving the Aerosol Optical Thickness (AOT) over land using SEVIRI data is presented. The AOT at 0.55 micron is estimated minimizing the difference between measured and computed radiances in the visible channel centred at 0.6 micron; this is done using a Look-Up Table (LUT) obtained with the Second Simulation of the Satellite Signal in the Solar Spectrum (6S) radiative transfer code considering a wide series of possible scenarios. To perform the mentioned minimization it is necessary to know the right satellite and sun geometry, the total ozone abundance and the land surface reflectance at 0.6 micron. Using the readings of five automatic sun-photometers of the Aerosols Robotic Network (AERONET) located in the Mediterranean area (Avignon, Laegeren, Modena, Rome and Lecce) an empirical linear relationship between the Top Of Atmosphere (TOA) reflectance at 1.6 micron and the surface reflectance at 0.6 micron was obtained. The results of the algorithm, compared with the quoted sun-photometric measurements show a standard error of 0.1. An AOT map of the Po Valley (Italy), obtained from an MSG image taken during a typical polluted day, is shown in the paper.

6745-67, Poster Session


Aerosol optical properties variation on different mountain sites in Italy
G. Pavese, Istituto di Metodologie per lAnalisi Ambientale (Italy); F. Esposito, L. Leone, R. Restieri, M. Calvello, G. Grieco, G. Masiello, C. Serio, Univ. degli Studi della Basilicata (Italy) Direct solar irradiance and IR atmospheric radiance have been measured during different measurements campaigns, conducted in mountain sites in Italy, located far from anthropogenic sources. Direct solar irradiance measurements have been obtained by a high resolution (1.5 nm) AVANTES USB2000 radiometer, working in the spectral range 400 nm 900 nm, while down welling IR sky radiance measurements have been collected by means of an MR100 BOMEM Fourier Transform Spectrometer covering the spectral ranges 500 cm1 - 5000 cm-1 or 100 cm-1 - 3000 cm-1, depending on the detector mounted on it and 0.5 cm-1 of resolution. At least two Radiosonde launches per day furnished ancillary data as temperature and water vapour profiles. All the instruments were located on a mobile laboratory, specifically projected to host them. From direct solar irradiance measurement Aerosol Optical Depth, Angstrom parameters and Aerosol Size Distributions have been retrieved, while IR data allow the water vapour and temperature profile retrieval. The measurements rates were 15 minutes for solar irradiance and 12 minutes for IR radiance. Both different orographic characteristics and different airmasses circulation on the measurements sites influence Aerosol Optical Depth values and variation, as verified even by the use of HYSPLYT back-trajectories analysis. Infrared radiances inversion allows the water vapour content retrieval and a correlation between measured AODs, retrieved effective radii and water vapour content has been looked for.

6745-65, Poster Session


Stratospheric ozone and nitrogen dioxide total column and vertical profiles in southern Portugal during 2004-2006
D. Bortoli, A. M. Silva, Univ. de vora (Portugal); G. Giovanelli, Consiglio Nazionale delle Ricerche (Italy) The SPATRAM (Spectrometer for Atmospheric TRAcers Monitoring) instrument has been developed by the collaboration between CGE-UE, ISAC-CNR and ENEA. SPATRAM is a multipurpose UV-Vis-NIR spectrometer (250 - 950 nm). It is installed at the Observatory of the CGE since April 2004 and actually it is utilized to carry-out measurements of the zenith scattered radiation, the so-called Passive mode, in order to retrieve by application of DOAS (Differential Optical Absorption Spectroscopy) methodology - the vertical content of some atmospheric tracers such as Ozone (O3) and Nitrogen Dioxide (NO2). The application of Differential Optical Absorption Spectroscopy (DOAS) methodology to the zenith scattered light data collected with the SPATRAM instrument allows for the detection of stratospheric trace gases involved in the ozone cycle such as NO2, OClO, BrO and the ozone itself. For the continuous NO2 monitoring the [407 - 460] nm spectral region is investigated. For the Ozone retrieval the spectral interval 320-360nm is chosen. The application of the DOAS algorithms to the spectral data recorded during the whole working period allowed for the retrieval of stratospheric Nitrogen Dioxide, and Ozone total column. In addition the application, to the output of the DOAS program, of sophisticated inversion schemes, using the Air Mass Factor (AMF) matrix as the kernel of the inversion algorithm, allowed for the determination of the nitrogen dioxide vertical profiles. The results obtained for Ozone and nitrogen dioxide are presented and discussed

6745-68, Poster Session


The influence of aerosols on tropospheric ozone production
J. R. Slusser, C. Corr, S. Kreidenweis, Colorado State Univ. (USA) In an effort to understand the factors governing the production of photo-chemical tropospheric ozone in large cities,

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

97

Conference 6745: Remote Sensing of Clouds and the Atmosphere


measurements of aerosol optical properties in the UV are presented. These measurements were part of large campaigns TEXAQS, MILAGRO (Houston, TX and Mexico City, resp) We report aerosol optical depths, single scattering albedo and wavelength averaged asymmetry parameter studied at using a UV 7 channel rotating shadowband radiometer. We also make as first atempt at correlating the UV aerosol optical properties with its chemical composition . Slant column density MAX-DOAS measurements of nitrogen dioxide (NO2) and ozone (O3) have been carried out in the Presidential Estate at Castel Porziano (41.7445 N, 12.4012 E) during the period September-November 2006. Castel Porziano Estate is a natural reserve which covers 62 km2 lying some 20 km from the centre of Rome, between the western suburbs of the capital and the urban centres of Ostia and Torvaianica. The instrument features a UV/Visible spectrometer (operating in the range 250-800 nm with a 0.5 nm optical resolution \@ 430 nm) connected by means of an optical fibre to a small telescope with both azimuthal and zenithal movements. The retrieval of information on the vertical distribution of trace gases from the measured slant column densities requires: - the application of sophisticated inversion schemes to derive the vertical profiles of the trace gases under interest using the Air Mass Factor (AMF) matrix as the kernel of the inversion algorithm; - an accurate simulation of the radiative transfer in the atmosphere for AMF calculations concerning target trace gases. AMF is used first to convert slant column densities into vertical column densities and then it is applied to the inversion algorithm. The nonlinear iterative inversion method based on the weighted Chahine algorithm was used to perform the retrieval of NO2 and O3 vertical profiles. Combining the slant column measurements performed from sunset to sunrise at different elevation angles we obtained the time evolution of the trace gas profile during the day. The PROMSAR (PROcessing of Multi-Scattered Atmospheric Radiation) model has been used to simulate the radiation transfer in the atmosphere and to calculate AMFs. This model participated in the recent past to a comparison exercise of radiative transfer models of various international research groups obtaining good results for the simulation of the multiaxis viewing geometries. In this study, the amount of profile information that can be retrieved from ground based MAX-DOAS measurements during September-November 2006 is investigated for NO2 and O3.

6745-69, Poster Session


Evaluation of air quality from space
S. Mukai, I. Sano, Kinki Univ. (Japan); M. Mukai, The Univ. of Tokyo (Japan) It is well known that the increasing emissions of anthropogenic aerosols associated with continuing economic growth in Asia has caused serious air pollution. The aerosol properties of urban atmospheric particles have been analyzed for radiometry data obtained using a multi-spectral photometer located at a NASA/ AERONET station at Kinki University Campus, Higashi-Osaka, Japan, since 2002. The suspended particulate matter (SPM) has been simultaneously measured at the same AERONET site since 2004. Long-term simultaneous monitoring of aerosols and SPM has revealed the characteristics and seasonal variations of the atmospheric particles over the urban city of HigashiOsaka. Our results suggest the presence of a strong correlation between the AOT (Aerosol Optical Thickness) and SPM, which highlights the possibility that the SPM concentration can be estimated from the AOT, and vice versa [1]. Moreover, combining radiometric aerosol information with surface-level particulate mass data appears to be a promising approach for gaining a better understanding of air quality and the atmospheric environment. However, it is difficult to relate SPM data directly to the column AOT. In other words, air pollutants are potentially related the emission and transportation of aerosols, because SPM data approximately represent the mass concentration of atmospheric particles at the surface. Therefore, in order to obtain the precise relationship between the surface-level SPM and satellite-derived AOT, an aerosol transport-radiation model should be derived to estimate the factors that affect this relationship. Factors such as the vertical distribution of aerosol mass concentration reflecting the boundary layer dynamics, and more detailed SPM measurements is investigated in this work. Furthermore, the latest results measured by a new instrument of PM1 set up in January 2007 will provide us with the meaningful information because the smaller particles have the stronger influence on the human health. [1] S. Mukai, I. Sano, M. Satoh, and B. N. Holben, Aerosol properties and air pollutants over an urban area, Atmos. Res., (2006). [doi:10.1016/j.atmosres.2006.02.020]

6745-71, Poster Session


Measurement and analysis of aerosol optical thickness over the East China Sea
X. Deng, Nanjing Univ. of Information Science and Technology (China); D. Pan, State Oceanic Administration (China); Z. Sun, Nanjing Univ. of Information Science and Technology (China); X. He, State Oceanic Administration (China); Z. Hao, Nanjing Univ. of Information Science and Technology (China) Shipboard hand-held sun photometers measurements which were collected over the East China Sea from December 15,2006 to February 11,2007 ,were used to discuss the temporal and spacial distributions of AOT,and the type and source of the aerosol particle over the East China Sea . The results of this study suggest that AOT was normally large in the coastal areas and became small gradually with the distance away from the coastal areas.The value of AOT also was larger in the coastal areas with industry than those without industry. AOT was closely related to the development of pollution source zone and wind transportation. The analysis of many cases showed that the minimum AOT value over sea appeared at noon during a day and its reason could be found in weather data, and AOT was correlated to water vapor. The calculated ngstrom wavelength exponent showed that the radius of aerosol particles was small in the coastal areas and large in the central area of the East China Sea. It can provide the basic data for marine regional climate and remote sensing.

6745-70, Poster Session


Multiple axis DOAS measurements for the retrieval of NO2 and O3 vertical profiles in the Presidential Estate of Caste Porziano, Rome
E. Palazzi, A. Petritoli, F. Ravegnani, I. K. Kostadinov, Consiglio Nazionale delle Ricerche (Italy); D. Bortoli, Consiglio Nazionale delle Ricerche (Italy) and Geophysics Ctr. of Evora (Portugal); S. Masieri, G. Giovanelli, Consiglio Nazionale delle Ricerche (Italy) In this work we present a methodology for the retrieval of the vertical distribution of atmospheric gas pollutants from ground based multiple axis Differential Optical Absorption Spectroscopy (MAX-DOAS). In contrast to ground based observations of zenith-scattered sun light, with MAX-DOAS ground based measurements one can observe scattered sun light from a variety of viewing directions simultaneously. Whereas traditional zenith DOAS instruments are mostly sensitive to trace gases located in the stratosphere, the higher sensitivity of MAX-DOAS to tropospheric trace gases is due to the enhanced optical paths through the lower atmosphere. Moreover, the main advantage of this kind of measurements is given by the opportunity to retrieve the vertical profile of a trace gas with a good time resolution through the combination of observations at different lines of sight.

6745-72, Poster Session


Air pollution monitoring using the open path technique
L. Belegante, National Institute of Research & Development for Optoelectronics (Romania); D. Zisu, National Research and Development Institute for Environmental Protection (Romania); I. Ionel, Politehnica Univ. Timisoara (Romania); D. N. Nicolae, National Institute of Research & Development for Optoelectronics (Romania) The open-path technique is a widely use method for air pollution monitoring, being capable of detecting trace gases down to

98

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6745: Remote Sensing of Clouds and the Atmosphere


ppb concentrations. Using this technique, the DOAS system can measure simultaneously several types of atmospheric molecular compounds having UV absorption spectra, by averaging on a long optical path (hundreds of meters). In this paper, two kinds of comparative measurements by DOAS and in situ detectors (using the point monitoring technique), in high polluted areas from Bucharest and the surroundings, are presented and analyzed. Only point monitoring is a standardized technique and therefore by comparing the two methods, several strengths and weaknesses of the open path technique will be evidentiated. This study is intended to mark out main environmental conditions (e.g. atmospheric fluctuations, topography) responsible for important differences that appear between experimental data acquired with the two techniques. transfer model, to estimate the radiation fluxes and the cloud radiative forcing.

6745-75, Poster Session


Analysis of measurements taken by a ceilometer installed in southern Portugal
M. J. T. Costa, Univ. de vora (Portugal) and Consultant (Portugal); D. Bortoli, V. Costa, Univ. de vora (Portugal); A. M. Silva, Univ. de vora (Portugal) and Consultant (Portugal); F. Wagner, S. Pereira, Univ. de vora (Portugal); J. L. Guerrero-Rascado, L. Alados-Arboledas, Univ. de Granada (Spain) A VAISALA Ceilometer CL31 is operating continuously in the Observatory of the vora Geophysics Centre (CGE) since May 2006. It has a measurement range from 0 to 7.5 km, reporting resolution of 5 m and programmable measurement cycle (from 2 to 120 s). The CL31 ceilometer provides measurements of the cloud base height up to three simultaneous layers and of the profile of the backscatter coefficients, which in the absence of clouds gives a good approximation of the aerosol boundary layer profile The enhanced single lens optics applied in the CL31 ceilometer ensures excellent performance also at low altitudes. The ceilometer measurements are used here to study special aerosol events that reach vora (3834' N, 754' W, 300m a.m.s.l.) such as forest fires and desert dust transports originating from the Sahara desert, which often occur in the south of Portugal, especially in spring and summer. These measurements are important in the way that they can complement radiative transfer calculations to derive aerosol and cloud properties from satellite data, providing the aerosol vertical distribution in the boundary layer, which is sufficient in some situations. The aerosol backscatter coefficient integrated in the lowest layers of the atmosphere is also correlated with data obtained from a TEOM (Tapered Element Oscillating Microbalance) installed in the same place (CGE observatory). The TEOM measures the in situ mass concentration of aerosols (particle mass per unit air volume) near the ground, with aerodynamic diameter lower than 10m (PM-10), with temporal sampling of 10min. Furthermore, ceilometer measurements are also compared with measurements taken with a Lidar during an intensive campaign that took place at the CGE observatory in vora, during the first half of June 2006.

6745-73, Poster Session


Influence of urban aerosol pollution to radiative forcing
A. V. Nemuc, C. L. Talianu, National Institute of Research & Development for Optoelectronics (Romania); S. Stefan, Univ. din Bucuresti (Romania) Daily PM10 concentrations of samples collected at two sites, urban and rural from Romania have been used to estimate the aerosol direct radiative forcing. Using OPAC (Optical Properties of Aerosols and Cloud) model we determined the single scattering albedo, the aerosol optical depth and aerosol up-scatter fraction, aerosols properties needed to estimate the magnitude and sign of direct aerosol radiative forcing. The surface albedo was assumed 0.2 for the urban site and 0.06 for the rural site for all wavelengths. For aerosol scale height we used 1km in winter and 2 km in the summer to calculate the optical depth of the boundary layer. Statistical analysis of the PM10 concentration for both sites show clear seasonal cycle with maxima in the winter. As a consequence of urban atmospheric pollution the radiative forcing for urban site appears strongly modified in comparison with rural site

6745-74, Poster Session


Satellite estimated cloud radiative forcing in the presence of aerosol events over the south of Portugal
D. Santos, M. J. T. Costa, Univ. de vora (Portugal) and Consultant (Portugal); D. Bortoli, Univ. de vora (Portugal) and Institute of Atmospheric Sciences and Climate (Italy); A. M. Silva, Univ. de vora (Portugal) and Consultant (Portugal) The study of cloud properties is central to understand climate. Clouds scatter and absorb solar radiation and emit and absorb thermal radiation, therefore play a significant role on the Earths radiation budget. Aerosol particles affect the Earth-Atmosphere energy budget directly through their interaction with solar radiation and indirectly, through the modification of cloud microphysical properties. The alterations that clouds undertake due to aerosols may have strong implications on their interaction with solar and terrestrial radiation, being for that reason the cloud radiative forcing an important issue for atmospheric and climate study purposes. Portugal is situated on the pathway of the desert dust plumes advected from Sahara desert and, during summer, forest fire smokes are extensively produced mainly in the central mountainous regions of Portugal and transported across the mainland. The aim of this work is to estimate the cloud radiative forcing in the presence of desert dust and forest fire aerosol events, in the area of Portugal. The analysis of the possible alterations undertaken by the clouds through their interaction with aerosols is done for selected episodes of strong desert dust transports and forest fire episodes that occurred in the Continental Portugal area and nearby Atlantic Ocean. The cloud and aerosol local vertical distributions are obtained from measurements taken in vora using a Vaisala CL31 LIDAR ceilometer, operating at 910 nm. The cloud properties, namely the type, optical thickness, effective radius and top temperature, are derived from the inversion of satellite radiation measurements in the visible and near infrared spectral regions. When the cloud properties are achieved, they are then used in combination with a radiative

6745-77, Poster Session


Lidar measurements for the short-term forecast of meteorological stability
J. Polkanov, B.I. Stepanov Institute of Physics (Belarus) Now the short-term forecast of the meteorological situation is actual, in connection with the tendency of growth of its sharp not predicted changes. Consideration of various methods of processing of the received information, used in meteorology, a radar-location, quantum spectroscopy, has allowed to reveal new opportunities of the complex form signal analysis reflecting processes of self-organizing in a terrestrial atmosphere. Such approach can be named the structural analysis of statistically generalized signal. Efficiency of such approach was checked by the example of processing results of laser sounding an atmosphere [1] and numerical modelling [2]. In result it was possible to reveal steady enough and regular turbulent structures in conditions concerning a steady atmosphere of a ground layer, their transformation at change of the meteorological situation [3]. The kind of these structures became complicated with reduction of thermodynamic stability of an atmosphere, and the degree of communication of the revealed structures is felling. It is offered to predict reduction of the meteorological situation stability, as more inertial process, on anticipatory change of the revealed structure character optical uniformities a ground layer concerning a steady atmosphere. It opens opportunities to use a method of optical anomalies detection [4] which occurrence is caused powerful local infringements of the atmosphere thermodynamic stability, not connected with of the daily meteorological situation. They can be caused as sharp atmospheric processes of spontaneous catastrophic character, and technogenic collapse by accidents outside of a zone of measurements when vital need to become

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

99

Conference 6745: Remote Sensing of Clouds and the Atmosphere


an opportunity to predict a trajectory of movement of such process on the basis of similar predictors. The literature: [1] Method of detection of the optical atmospheric inhomogeneities structure. Patent USSR &#8470; 1597816, G01W1/00 from 05.07.88, publ. 08.06.90. [2] Polkanov, Iu. A.; Kudinov, V. N. A possibility for the analysis of the periodic structure of a complex signal. Priborostroenie (ISSN 0021-3454), vol. 32, April 1989, p. 6-11. http:// adsabs.harvard.edu/abs/1989Prib...32....6P [3] Polkanov, Iu. A Matching between a change in the structure of atmospheric optical inhomogeneities and a set of meteorological parameters. Meteorologia i gidrologia, Moskva (ISSN 0130-2906), 1991, no3, pp. 39-48. http://cat.inist.fr/ ?aModele=afficheN&cpsidt=5237241 [4] Method of definition of the optical heterogeneity position in atmosphere. Patent USSR &#8470; 1448907, G01W1/00 from 01.06.86, publ. 01.09.88. http://www.inno.ru/projects/21877/ project22145.shtml solar irradiance. Then an iterative procedure has been started tuning H2O, CO2, CO, O3, and aerosol abundances. In practice, the MODTRAN 4 code is executed several times with different atmospheric parameters (H2O, CO2, CO, O3, and aerosol abundances) until the calculated ground irradiance matches the in-field measurements and the estimated ground spectral reflectance map is free from the related spectral signatures. To test and validate the method the at-sensor radiance images acquired on 15th December 2005 during a coastal zone remote sensing campaign by the new airborne sensor HYPER / SIMGA have been utilized. The sensor has been operating in the 0.4 - 2.5 m spectral range with 768 bands and a resolution of 2.4 - 5.4 nm. First results are presented and discussed taking into account the feasibility of avoiding in-field measurements.

6745-33, Session 4
Characterization of tropical atmosphere through wide-band emission spectra acquired by a balloon-borne uncooled FTS spectroradiometer
G. Bianchini, B. Carli, U. Cortesi, L. Palchetti, Istituto di Fisica Applicata Nello Carrara (Italy) On June 30th, 2005 the REFIR-PAD (Radiation Explorer in the Far InfraRed - Prototype for Application and Development) Fourier transform spectroradiometer performed the first wideband spectral characterization of the top-of-atmosphere emitted radiation in the far-infrared with an uncooled instrument. The nadir emitted radiance has been measured down to 100 cm-1, thus covering a spectral interval that, until now, was nearly unexplored, and up to 1400 cm-1, including the well characterized atmospheric window region, in which it is possible to perform comparison and intercalibration with operative instruments. The measurements were performed at an altitude of 34 km, from a stratospheric balloon launched in tropical region, near Teresina (Brazil). The acquired spectra have a spectral resolution of 0.5 cm-1. It should be noted that despite the operating spectral range extending to the far-infrared region, REFIR-PAD does not require any cooled components, thanks to the use of room temperature pyroelectric detectors and an optical scheme that compensates for the instrument selfemission. This work shows the results of the analysis of the spectra, focusing on the far infrared portion of the atmospheric emitted radiance. The retrieval of the vertical profiles of water vapour and temperature during the flight is presented. The vertical resolution of the retrieval is 2 km in the upper troposphere lower stratosphere (UTLS) region, and lower at higher altitudes. The comparison with ECMWF for validation is also shown. Besides the characterization of temperature and water vapour, from the analysis of the emitted radiance useful information can be gathered about cloud and aerosol contribution to radiation budget.

6745-31, Session 4
Influence of atmospheric profiles variations on airborne infrared limb observations
C. Malherbe, P. Chervet, C. Lavigne, ONERA (France) Interaction of light with atmospheric particles such as molecules, aerosols, cloud droplets or ice crystals is one of the major problems studied in the field of remote sensing and atmospheric propagation. The performance of an electro-optical sensor is affected by the atmospheric attenuation of the sourcetarget radiation , and by the atmospheric radiation that is part of the background/foreground radiation observed by the sensor. Generally, mean atmospheric profiles are used to compute radiative transfer. But to study the characteristics of an optical sensor, one is not only interested in the synthetic radiance for a single profile but rather in the sensitivity of radiances to changes in the profile. Temperature, pressure and density profiles have been generally obtained by use of ground lidar and radar, balloon, rocket or satellite experiments. Such data can be gathered in a climatological library, as data bank TIGR (Thermodynamic Initial Guess Retrieval) that is used for this study. TIGR water vapor and temperature profiles have been compared to satellite measurements provided by MLS instrument in order to apreciate their relevance. The sensitivity of radiance and transmission to the climatic conditions is tested. The goal of this study is to evaluate and quantify atmospheric component variations that influence radiance and transmittance for airborne limb viewing observations. In particular, radiance variations are analyzed according to the variations of temperature and of the different atmospheric molecules. This radiance variability is related to the water content and temperature modifications.

6745-32, Session 4
An atmospheric correction iterative method for very high resolution aerospace imaging spectrometers
A. Barducci, D. Guzzi, P. Marcoionni, I. Pippi, Istituto di Fisica Applicata Nello Carrara (Italy) Nowadays increasing the radiometric accuracy and spectral resolution of the new aerospace optical imagers for Earth observation could allow a better characterization of the environment. This is really true if accurate radiometric calibrations of the sensor are performed and atmospheric effects on the acquired data are carefully accounted for. To obtain spectral surface reflectance maps from the at-sensor radiance images, an improved atmospheric correction procedure have to be implemented. Based on the availability of data acquired at so high spectral resolution to allow the detection of different spectral features of many atmospheric constituents, an iterative estimation algorithm has been developed using the MODTRAN 4 radiative transfer code. The default atmospheric profiles available in that code have been firstly refined through at-ground level measurements of some parameters, like temperature, pressure, humidity, and

6745-34, Session 4
Extension of Chandrasekhars formula to a nonhomogeneous Lambertian surface and comparison with the 6S formulation
A. Sei, Northrop Grumman Space Technology (USA) The classical Chandrasekhars formula, which relates the surface albedo to the top of the atmosphere radiance, rigorously applies to a homogeneous Lambertian surface. For a nonhomogeneous Lambertian surface in a plane parallel atmosphere, an extension of this formula was proposed in the eighties and has been implemented recently in the 6S algorithm. To analyze this extension, this presentation derives the rigorous formula of the top of the atmosphere signal in a plane parallel atmosphere bounded by a non-homogeneous Lambertian surface. Then the 6S algorithm extension is compared to the exact formula and approximations and their validity are examined. The derivation of the exact formula is based on the separation of the radiation fields into direct and diffuse components, on the introduction of the Greens function of the problem and on integrations of boundary values of the radiation fields with the Greens function.

100

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6745: Remote Sensing of Clouds and the Atmosphere

6745-35, Session 4
Analysis of adjacency effects for two Lambertian half-spaces
A. Sei, Northrop Grumman Space Technology (USA) An analysis of one dimensional adjacency effects is presented. This classical problem is shown to be amenable to closed form error analysis when 1D adjacency correction schemes are used. In particular the error made when surface reflectance is retrieved using an infinite target assumption is given in closed form. This allows deductions for the behavior of the error as a function of wavelength and optical thickness. Typical lengthscales of the adjacency effects are deduced and the range and magnitude of the error are also given in closed form.

6745-38, Session 4
Exact analytical solution of 3D radiative transfer equation in problem of remote sensing
A. B. Gavrilovich, B.I. Stepanov Institute of Physics (Belarus) The urgent problems of photometry in remote sensing of atmosphere call a necessity of fundamental investigations directed to improvement of analytical methods solution of radiative transfer equation (RTE) for real scattering phase function. The problem of light scattering function by an aggregation of particles is one of the most complex problems of atmosphere optics. It is very difficult to mathematically describe the interaction of light with such a system when it is necessary to take into account the multiple scattering in 3D medium. However, this is required in the most important problems of optical diagnostics of clouds in real atmosphere. Because of multiparametricity and complexity of interaction of light with particles, till now the solution of such problems were based on the numerical methods. The numerical data obtained by approximate methods hampers to the analysis of the real geometrical and physical factors determining properties of scattered radiation. Because of this, at present, much attention is being given to the development of methods based on the use of analytical solution of RTE. The main difficulty of analytical solution of RTE, which is considered to be incurable, as is known, is taking into account the infinitely large number of spherical functions. For atmospheric aerosol, for example, a good approximation is provided by their number equal or larger than 400. The known analytical approximations are very idealized and, as a rule, do not provide the obtaining of physical right results. For example, the method of spherical harmonics leads to significant lasses of the information at small scattering angles. The method of small-angle approximation, by contrast, ignores the very informative region of large scattering angles. The new method developed by author and presented in this work is in free of the indicated drawbacks. We have realized a new of solution of 3D radiative transfer equation, which implies that the infinite system of spherical functions is replaced by a finite orthogonal basis of G-functions satisfying the complexity conditions. G-functions forms the class of functions belonging in the finite functional G-space. High sensitivity of the scattering effects to the properties of particles imposes very high requirements on the new method developed for solution of the transfer equation, namely, the solution should involve the whole information contained in real phase function. As object of investigation, we considered a 3D volume for aggregation of particles with an arbitrary scattering function. The exact solutions of RTE for intensity of scattered light by clouds are present in form of a finite series in the G-function space.

6745-36, Session 4
Retrieval of minor constituents in a cloudy atmosphere with remote sensing millimeter wave measurements
S. Del Bianco, G. Bianchini, M. Gai, L. Santurri, Istituto di Fisica Applicata Nello Carrara (Italy); C. Cecchi-Pestellini, Osservatorio Astronomico di Cagliari (Italy); B. M. Dinelli, Consiglio Nazionale delle Ricerche (Italy); B. Carli, Istituto di Fisica Applicata Nello Carrara (Italy) We know very little about the atmospheric chemistry in presence of clouds, because optical path through the cloud itself prevents remote sensing observations. The cloud effects on the radiative transfer are to increase the atmospheric optical depth and to introduce the scattering source function contributes. In the millimeter and submillimeter wave region the absorption effect is reduced for ice particle clouds and the scattering effect becomes important only for the larger particles composing the cloud. The millimeter and sub-millimeter wave region can be used to study the atmospheric chemistry in the presence of clouds. The effect of the clouds depends on particle phase, on radius and on numerical density of the particles composing the cloud. A theoretical retrieval analysis is made to assess conditions in which retrieval is possible. We find that in most cases an atmospheric continuum model is sufficient to describe the radiative transfer in cloud, however, in a few cases, that will be specified, a model that takes into account the scattering contribution is needed.

6745-37, Session 4
Analytical determination of the atmospherescattered radiation polarization degree for the sake of polarized remote sensing
V. P. Budak, S. V. Korkin, Moscow Power Engineering Institute Technical Univ. (Russia) In these paper we continue to develop our new method of solution of the vectorial radiative transfer equation (VRTE) boundary problem in the case of a strongly anisotropic scattering slab (clouds, ocean) illuminated by a plain unidirectional source of light with an arbitrary angle of irradiance and polarization degree. The efficiency of the proposed method lies in our way of calculation of the diffuse light field within the slab. We assume the complete VRTE solution to be the superposition of two parts: the most anisotropic small-angle part and smooth non-small angle regular part. The first has been computed in our previous papers using the vectorial small angle modification of spherical harmonics method (VMSH) based on the consideration of the smoothness of the spatial spectrum of the Stokes vector (SV) distribution within the slab. The VMSH as the source function allows us to formulate the VRTE boundary problem for non-small angle part only. This problem was solved by means of the matrix generalized Legendre polynomials (GLP) of the desired SV and scattering matrix expansions, the well known circular basis and the vectorial discrete ordinates method (VDOM). On account of the smoothness of the regular non-small angle part it needs not too much expansion GLP terms to be calculated. This together with quite simple matrix exponent expression for the VMSH determines the calculation efficiency of proposed method. We have compared our calculations with the standard VDOM, the Monte-Carlo simulation, the single scattering approximation and the scalar methods for the total intensity.

6745-39, Session 5
Anisotropic refractive index fluctuations spectrum in the stratosphere sensed from balloon-borne observations of stellar scintillation
C. Robert, J. Conan, V. Michau, ONERA (France); J. Renard, C. Robert, F. Dalaudier, Ctr. National de la Recherche Scientifique (France) Scintillation effects caused by the fluctuations of the refractive index of air are not negligible in the stratosphere. Recent experiments highlight the composite nature of optical turbulence in the stratosphere. We present an analytical model of scintillation based on a 3-D model of anisotropic and isotropic refractive index fluctuations spectrum that predicts scintillation rates inside the Rytov regime. This model uses a multi-layer decomposition of the turbulence profile. The effect of anisotropy leads to significant scintillation rates when a source is observed with a horizontal line of sight. Astronomical observations consisting in stellar scintillation made from balloon-borne spectrometer AMON-RA allow us to remotely probe statistical characteristics in the stratosphere, to validate the model of scintillation developed herein and to obtain refined values of its parameters. Data reduction from these observations brings out values of the inner scale of the anisotropic spectrum. We retrieve metric values of the inner scale that are compatible with space-based measurements. We find a major contribution of the anisotropic spectrum relatively to the isotropic contribution. This effect is particularly noticeable when the sight line plunges into the

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

101

Conference 6745: Remote Sensing of Clouds and the Atmosphere


atmosphere, leading to strong scintillation as well as coupled chromatic refraction effects. This makes clear the presence of gravity waves in the stratosphere. k algorithm will be included as an addition radiation transport option in a future release of SAMM-2. Unlike the MODTRAN correlated-k model, which is an LTE treatment of the problem, the correlated-k algorithm in SAMM-2 accounts for NLTE atmospheric conditions. In addition, MODTRAN uses band model parameters to approximate molecular response within a spectral interval while the SAMM-2 correlated-k algorithm processes molecular lines at runtime by reading line center information from the HITRAN 200 database and computing statistical cumulative probability distributions within a spectral interval under the presumption of a Voigt line shape profile. Atmospheric climatology is provided by the SHARC/SAMM Atmospheric Generator (SAG2), which includes MSISE-90/ NRLMSISE-00 and the NRL climatology databases. Since correlated-k algorithms provide a monochromatic treatment of radiation transport phenomena, this model is useful for treating any atmospheric phenomena requiring a Beers Law (spectrally monochromatic) treatment of the atmospheric transmission and/or radiance (such as multiple scattering or atmospheric structure). We will discuss the atmospheric background capabilities of this algorithm and illustrate its effectiveness by comparing radiance, transmission and structure output to the SAMM-2 QBL algorithm.

6745-41, Session 5
Empirical storm-time correction to the international reference ionosphere model Eregion electron and ion density parameterizations using observations from TIMED/SABER
C. J. Mertens, NASA Langley Research Ctr. (USA) Observations of thermospheric infrared emission from the TIMED/SABER instrument have fostered development of new data products, models, and analysis tools for the study of upper atmospheric and ionospheric response to solar-geomagnetic disturbances. The focus of this paper is on analyzing SABER 4.3 um limb emission measurements to quantify the E-region response to solar-geomagnetic storms, with the objective of developing an empirical E-region storm-time correction to the International Reference Ionosphere (IRI) model. The IRI model is a widely used empirical model for the specification of ionospheric parameters and is recommended for international use by the Committee on Space Research (COSPAR) and the International Union of Radio Science (URSI). However, the specification of the ionospheric response to solar-geomagnetic disturbances in IRI remains largely incomplete, and there is currently no storm-time correction to IRI parameters in the Eregion. We have initiated a study to use nighttime SABER 4.3 um limb emission measurements to develop an E-region stormtime correction to IRI. Enhancements in nighttime 4.3 um emission during storm periods are due to vibrational excitation of NO+ (i.e., NO+(v)), caused by auroral electron dosing and subsequent ion-neutral chemical reactions, followed by emission at 4.3 um. Since the E-region is largely inaccessible to observation, especially on a global scale, SABER measurements provide a suitable dataset from which to develop an empirical E-region storm model. The observation-based proxy used to develop the storm model is SABER-derived NO+(v) 4.3 um volume emission rates (VER). Furthermore, since NO+ is the terminal E-region ion, NO+(v) VER enhancements during solar-geomagnetic storms are indicative of the response of the electron density to solar-geomagnetic forcing, by charge neutrality. The NO+(v) VER is derived from SABER 4.3 um limb emission measurements by (1) removing the background contribution from CO2 infrared emission, and (2) by performing a standard Abel inversion on the residual radiance. All magnetically disturbed periods contained in the SABER database from 2002-2006 will be used to develop the storm model parameterization. An important benefit of our approach is that the empirical storm model is largely independent of NO+ or electron density profile shapes, and independent of chemistry, kinetics, or spectroscopic parameters.

6745-43, Session 5
Contributions of the OH airglow to space object irradiance
J. H. Gruninger, J. W. Duff, Spectral Sciences, Inc. (USA); J. H. Brown, Air Force Research Lab. (USA) We investigated the contributions of the hydroxyl (OH) airglow to the illumination of resident space objects. During nighttime, in a moonless sky, the airglow is the largest contributor to the sky brightness in the visible (vis), the near-infrared (NIR) and short-wave infrared (SWIR) spectral region. The dominant contributors to the airglow are vibrationally excited hydroxyl radicals, OH(n). The radicals are formed in vibrational states up to n=9 by the reaction of hydrogen atoms with ozone. The strong emissions, known as Meinel emissions, are sequences with delta n= 1-6. Emissions with n= 3, 4, 5 and 6 occur in the visible and NIR between .4 and 1.0 microns. From 1.0 to 2.5 microns there are very strong emissions from the delta n= 2 sequences. The delta n= 1 emissions extend into the thermal infrared to 4.5 m. In this work, we considered four band passes, a visNIR band pass, two SABER band passes centered at 1.6 and 2.0 microns, respectively, and a broad band pass around 2.7 microns. SAMM2 was utilized to compute spectra and line of sight radiances. We used line of sight (LOS) radiances to compute the irradiance on a space object that was taken as a flat plate with a Lambertian surface reflectance. Profiles of irradiance versus orientation were calculated. The OH airglow will illuminate a facet even if it is pointing somewhat upward. However, the irradiance in the 2.7 micron band pass comes almost entirely from the atmosphere in the low altitude and the earth emission.

6745-42, Session 5
A new model for calculating infrared background radiance at all altitudes including atmospheric clutter and clouds
J. W. Duff, R. Panfili, L. S. Bernstein, Spectral Sciences, Inc. (USA); J. H. Brown, Air Force Research Lab. (USA) This talk describes the development of a new radiation transport algorithm included in the latest version of the all-altitude AFRL background radiance code, SAMM-2. SAMM-2 combines the high-altitude (50 km and above) capabilities of SHARC with the low-altitude (50 km and below) capabilities of MODTRAN into a unified code which retains the capabilities of the older codes (clouds, aerosols, multiple scattering, etc.) and adds a high-resolution (0.001 cm-1) quasi line-by-line (QBL) radiation transport algorithm and the capability to simulate stochastic structure. Comprehensive coverage of the 0.4 to 40 m (250 to 25,000 cm-1) wavelength region is provided for arbitrary lineof-sight (LOS) in the 0-300 km altitude regime. Molecular emissions, valid under non-local thermodynamic equilibrium (NLTE) at high altitudes (50 km), as well as local thermodynamic equilibrium (LTE) at lower altitudes ( 50 km), are calculated. Using this as a foundation, a novel 1 cm-1 resolution correlatedk algorithm has been developed which provides an orders-ofmagnitude increase in computational efficiency when compared to the version 1.82 SAMM-2 QBL method. This new correlated-

6745-44, Session 5
TIMED/SABER limb measurements of OH Meinel emission at 1.6 and 2.0 micrometer: global behavior and interannual variability
J. R. Winick, R. H. Picard, Air Force Research Lab. (USA); P. P. Wintersteiner, ARCON Corp. (USA); D. Esplin, M. J. Taylor, Utah State Univ. (USA); I. Azeem, Embry-Riddle Aeronautical Univ. (USA); M. G. Mlynczak, NASA Langley Research Ctr. (USA); J. M. Russell III, Hampton Univ. (USA) We examine the global latitude and local-time behavior of the OH Meinel emission measured by the SABER limb-scanning radiometer on the TIMED spacecraft in the two SABER near-IR OH-Meinel bands, OH-A near 2.0 micrometer and OH-B near 1.6 micrometer, for the period 2002-2006. The behavior of the peak altitude of the derived volume emission rate and the intensity at the peak is remarkably consistent from year to year. Tidal influence is clearly present at low and mid-latitudes. However at northern high latitudes in the January-March period, distinct differences in the peak height and intensity are observed between 2004 and 2005. We investigate the possible cause of this interannual variation by relating it to the interannual variation of the SABER-derived atmospheric temperature/pressure profiles and by comparing to correlative data and model output.

102

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6745: Remote Sensing of Clouds and the Atmosphere

6745-45, Session 6
Budapest airport air quality long-term studies by remote sensing with DOAS and FTIR with focus upon runway emissions
K. Schfer, G. Schrmann, C. Jahn, C. Matuse, H. Hoffmann, Forschungszentrum Karlsruhe (Germany); S. Torok, V. Groma, KFKI Atomic Energy Research Institute (Hungary) Airport air quality is influenced by traffic mainly. Near runway the aircrafts are the main source. The quantification of these emission sources requires remote sensing methods because the airport operations should not be disturbed. DOAS is used in open-path mode to detect continuously NO2 cross the runway during more than 6 months. Those runway emission studies were performed for the first time. During a measurement campaign these findings were compared with aircraft taxi emissions. Aircraft emission indices of CO, NO and NO2 were determine by using open-path FTIR spectrometry also. The concentration measurements of CO2 which are necessary to calculate emission indices are provided by open-path FTIR spectrometry. These emission indices are compared with those listed for each engine in a data base of the International Civil Aviation Organisation (ICAO) for four different thrust levels (Idle, approach, cruise and take-off). The comparison of these indices to real in use measurements is a task to develop realistic emission inventory of airports. Open questions and required further developments will be discussed.

masses from Saharan regions follow two main paths from the sources to Southern Italy and the dust optical properties seem to be trajectory-dependent.

6745-48, Session 6
Evolution study of smoke backscattering coefficients in a cell by means of a compact mobile Nd:Yag lidar system
C. Bellecci, Univ. degli Studi di Roma/Tor Vergata (Italy); L. De Leo, CRATI s.c.r.l. (Italy); P. Gaudio, M. Gelfusa, Univ. degli Studi di Roma/Tor Vergata (Italy); T. Lo Feudo, CRATI s.c.r.l. (Italy); S. Martellucci, M. Richetta, Univ. degli Studi di Roma/Tor Vergata (Italy) Detection of smoke from forest fire was one of the first practical applications of lidar. As it is well known, smoke contains a large number of small particles of ash or soot, leading to a large backscattering efficiency and consequently favourable conditions for lidar application. We developed a compact mobile lidar system based on Nd:Yag Q-Switched laser source operating at three wavelengths: , and with repetition frequency equal to 10 Hz and laser pulse duration equal to 5 ns for the fundamental harmonic and 4 ns for the second and the third. This system has been tested with the experimental measurements of the smoke backscattering coefficients carried out in an ad hoc cell (length 8.60 m). Since the spatial resolution of our laser pulse is smaller than cell length it has been possible to evaluate the evolution of the smoke backscattering coefficients into cell itself. These results have been used to develop a computational model for simulating the smoke evolution within the cell. Subsequently we optimized the model to study the smoke evolution into the atmosphere, since our aim was to validate and so to carry out experimental measurements as soon as possible. Measurements of smoke backscattering coefficients and simulation are shown and discussed in the paper.

6745-46, Session 6
Mexico City airport air quality study by remote sensing with passive FTIR
K. Schfer, E. Flores-Jardines, C. Jahn, Forschungszentrum Karlsruhe (Germany) The scanning infrared gas imaging system (SIGIS-HR) and the quantitative gas analysis software MAPS (Multicomponent Air Pollution Software) are applied to investigate the spatial distribution of the temperature and gas concentrations (CO, NO) within the plume of aircraft turbines at Mexico City airport. The system integrates an infrared camera also. It is used for the localisation of the hot source that additionally suggests the best measurement position is the SIGIS-HR. The rough concept of the system will be presented and operational applications will be discussed. The results of the investigations of the temperature and gas concentrations (CO, NO) within the aircraft engine plumes will be shown. The limitations and of the systems will be discussed.

6745-49, Session 6
Properties of fire smoke in east Europe measured by remote sensing methods
A. Pietruczuk, Institute of Geophysics (Poland); A. P. Chaikovsky, B.I. Stepanov Institute of Physics (Belarus) Usually at spring and autumn seasons forest and peat-bog fires takes place at territory of western Russia, Ukraine and Belarus. Products of combustion are transported on long range distances. In a number of cases the level of air pollution by smoke exceeds sanitary norms in Central and Western Belarus and Poland. Information on the processes of smoke transportation with increasing number of fire seats in the East European region appears extremely important for environmental institutions to estimate ecological situation. Monitoring of smoke plumes were performed by lidar groups from B.I. Stepanov Institute of Physics and Institute of Geophysics PAS with cooperation with environmental services in the period of increased air pollution probability (spring, autumn). Field measurements were carried out by lidar satation at Minsk and Belsk which are part of EARLINET lidar network and collocated CIMEL Sun-photometer federated in AERONET. Remote sensing observations were supported by ground measurements based networks of instruments operated by local environmental services. The obtained results of smokes characteristics was compared to satellite retrievals, mathematical modeling and back trajectories as well. This work was financially supported by the European Commission under grant RICA-025991.

6745-47, Session 6
Atmospheric aerosol characterization during Saharan dust outbreaks at Napoli EARLINET station
G. Pisani, Univ. degli Studi di Napoli Federico II (Italy) and Consultant (Italy); M. Armenante, Istituto Nazionale di Fisica Nucleare (Italy); M. G. Frontoso, N. Spinelli, X. Wang, Univ. degli Studi di Napoli Federico II (Italy) and Consultant (Italy) In order to evaluate the tropospheric aerosol dynamic over Naples under Saharan dust outbreaks conditions a detailed analysis of lidar measurements performed between May 2000 and August 2003 in the frame of the EARLINET project have been carried out. Climatological analysis on sand plume has been done, with a comparison between normal and dust conditions. Results in terms of backscattering and extinction coefficient as well as of integrated quantities show that from the ground level up to 2 Km the aerosol load during Saharan dust transport events is almost the same of normal conditions. This is probably due to the relevant of widespread local aerosol sources, such as vehicular traffic, industrial activities, etc.. Nevertheless, when sand outbreaks occur, the extinction to backscattering ratio, the lidar ratio, clearly shows that the aerosol type in lowest atmospheric layer changes. Moreover an increase of extinction-to-backscatter ratio is observed in free troposphere as a function of the altitude as a consequence of gravitational settling. Trajectories analysis shows that air

6745-50, Session 6
Diurnal radiative forcing of biomass burning aerosols over Africa from merged GERB and SEVIRI data
C. P. Bertrand, A. Ipe, L. Gonzalez, G. Casanova, N.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

103

Conference 6745: Remote Sensing of Clouds and the Atmosphere


Clerbaux, D. Caprion, S. Dewitte, Royal Meteorological Institute of Belgium (Belgium) The synergy between the Geostationary Earth Radiation Budget (GERB) broadband radiometer and the Spinning Enhanced Visible and Infra Red Imager (SEVIRI) onboard the European meteorological satellite Meteosat-8 is exploited to estimate the diurnal variation of the direct short wave aerosols radiative forcing (SWARF) from biomass burning over tropical Africa at sub-GERB footprint scale. Biomass burning aerosols are first identified at the SEVIRI resolution (3 km at nadir) by applying a multispectral thresholding algorithm to the SEVIRI spectral measurements. Reflected SW fluxes at the topof-atmosphere for smoke aerosols are obtained by converting the measured GERB radiances at a 3x3 SEVIRI pixel window in term of flux using a theoretically derived smoke angular distribution model (ADM) based on the average scene identification from the 3x3 SEVIRI pixel box. The calculated smoke ADM is a function of aerosol optical depth, surface type and solar and satellite viewing geometry. The TOA SWARF for smoke aerosols is then estimated as the difference between radiative fluxes in the absence and presence of biomass burning aerosols. Finally, the calculated TOA fluxes for smoke aerosols are compared with those obtained without using a dedicated smoke ADM when performing the radiance to flux conversion to estimate the improvement in the near-real time processing of the GERB and SEVIRI data performed at the Royal meteorological Institute of Belgium due to the introduction of the smoke ADM.

6745-51, Session 6
Comparisons of satellite-derived aerosol optical depth over a variety of sites in the Southern Balkan region as an indicator of local air quality
M. E. Koukouli, S. Kazadzis, D. S. Balis, Aristotle Univ. of Thessaloniki (Greece); C. Ichoku, NASA Goddard Space Flight Ctr. (USA); V. Amiridis, National Observatory of Athens (Greece) This study describes the atmospheric aerosol load encountered over a number of sites from the Southern Balkan region with a relatively well-known air quality factor. Using the aerosol optical depth, AOD, retrieved from the two Moderate Resolution Imaging Spectroradiometers, MODIS, on board the Terra and Aqua NASA satellites, the aerosol content of numerous sites is investigated under the scope of local pollution sources, interregional transport and large scale dust and/or biomass burning events. The wide time range of six full years considered permits the discussion of possible climatological aspects as well. The MODIS AOD is further validated over a metropolis of Northern Greece using ground-based Brewer spectrophotometer measurements and co-located AOD and Aerosol Index values from the Ozone Monitoring Instrument, OMI, on board the Aura satellite. The metropolis is situated in a unique sea-side location which inflicts it with high humidity and sea-salt particles, and is furthermore frequently affected by biomass burning and desert dust aerosols arriving from surrounding sources. Local and regional pollution further influences the quality of the local air and the observed tropospheric optical depth.

6745-78, Session 6
Hazard related to anomalous emissions of endogenous gas in the Rome region: estimation of CO2 and H2S air concentration and soil flux by TDL and accumulation chambers
K. Weber, Fachhochschule Dsseldorf (Germany) No abstract available

104

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6746: SAR Image Analysis, Modeling, and Techniques


Tuesday 18 September 2007 Part of Proceedings of SPIE Vol. 6746 SAR Image Analysis, Modeling, and Techniques IX

6746-02, Session 1
Processing of TerraSAR-X payload data: first results
H. Breit, DLR Standort Oberpfaffenhofen (Germany); U. Balss, Technische Univ. Mnchen (Germany); R. Bamler, DLR Standort Oberpfaffenhofen (Germany) TerraSAR-X is a national German satellite providing a highresolution X-Band SAR instrument, which will be launched in April 2007. The mission is implemented in the framework of a public-private partnership between the German Aerospace Center (DLR) and EADS Astrium GmbH Germany and will provide high resolution SAR data products for commercial use and scientific exploitation. The key element of the instrument is an active phased array antenna nominally operated with a bandwidth of 150 MHz and an experimental 300 MHz capability. The instruments flexibility with respect to electronic beam steering and pulse-to-pulse polarization switching allows the acquisition of SAR data in stripmap, spotlight and ScanSAR imaging configurations in different polarization modes for a wide range of incidence angles. Processing of the payload data will be performed at DLRs Payload Ground Segment (PGS) for TerraSAR-X. The central part of PGS is the TerraSAR Multi-Mode SAR Processor (TMSP) focusing the SAR data in a unified way for the different imaging configurations. A wide range of processing options spanning from phase preserving complex products in slant range geometry to enhanced ellipsoid corrected intensity images lead to a comprehensive collection of SAR product types and variants. During the 5 months lasting commissioning phase the complete processing chain will be properly tuned and adjusted. The TMSP algorithms have to be configured, e.g. thresholds for calibration pulse analysis, estimation window sizes for SAR data analysis, parameterization of estimation algorithms. Also the configuration of product variants with respect to resolution and radiometric quality will be checked and refined. This paper gives a short review of the different imaging configurations and product variants and presents first example images. The first experiences with respect to product performance and quality are outlined.

degrees of the data complexity. To cope with the problem of estimation and model order selection Bayesian inference is used. The results are presented on single, dual, and quad PolSAR data.

6746-05, Session 2
Soil moisture maps of agricultural soils from ENVISAT/ASAR images: an attempt to correct the surface roughness effects
E. Santi, M. Brogioni, S. Paloscia, P. Pampaloni, S. Pettinato, Istituto di Fisica Applicata Nello Carrara (Italy) The availability of multi-temporal soil moisture maps is an important resource for the water management of a basin and the forecast of flood events. Since the local measurements of soil moisture content are strongly affected by spatial variability, besides being time-consuming and expensive, the use of microwave satellite sensors, with their complete and frequent coverage of the Earths surface, is extremely attractive. On the basis of the present knowledge the best frequency for measuring soil moisture seems to be L-band, whose data will be available soon from ALOS. The retrieval of soil moisture is also achievable at C-band, which is operational on ERS-2, RADARSAT, and ENVISAT satellites, but, in this case, the effects of soil surface roughness and vegetation cover on the backscattering coefficient should be taken into account by using adequate correcting procedures. In this paper, an approach to correct SAR data for the surface roughness effects, and generating multi-temporal soil moisture maps has been investigated. The ENVISAT/ASAR images used for this research were collected in the flat agricultural area of Alessandria in North-west Italy on November 2003, April and June 2004. Ground measurements of soil moisture (with a TDR probe) and fresh biomass of vegetation were carried out during the ENVISAT overpasses. Weather was generally cloudy and rainy in springtime and fall, and sunny and dry in summer. At each date, soil moisture was rather uniform on the whole area and its average value increased from November to April and then decreased again in June. In November most fields were bare soils, some of them smooth, since they were sowed with winter wheat, and some very rough, since recently ploughed. In April, the fields were or covered by already developed vegetation (wheat) grown on a relatively smooth surface (hstd 1 cm), or bare smooth soils prepared for the spring seeding. In June, all fields were completely vegetated with an underlying rather smooth surface. A preliminary test confirmed the sensitivity of the backscattering coefficient at C-band (measured in HH polarization and incidence angle of 23 degree at different dates) to the soil moisture measured on ground, although with some dispersion of experimental points due to the effect of surface roughness and vegetation biomass. The retrieval of soil moisture was obtained by using a feedforward neural network (ANN) having two hidden layers of neurons between the input and output, and trained by using the back-propagation (BP) learning rule. The training dataset was obtained from archive data collected on a completely different test site, the Montespertoli area in Italy during the SIRC/X-SAR project. The ANN inputs were the backscattering coefficient at C-band in HH polarization and 23, along with the surface roughness parameters (i.e. height standard deviation, Hstd, and correlation length). In order to better evaluate the effects of surface roughness, the Advanced Integral Equation Model (AIEM) was used for simulating the backscattering coefficient for different values of surface roughness and soil moisture. The comparison between measured and simulated data showed a good agreement, mainly for height standard deviation values between 0 cm and 3.5-4 cm and soil moisture between 10% and 25%. Therefore, the experimental training set was increased by simulating radar backscattering through the AIEM, which was driven with soil data taken in the range of ground measurements. The ANN was validated using experimental data collected on a restricted sample of fields, where ground measurements of soil moisture

6746-04, Session 1
Phase characterization of PolSAR images
M. M. S. Soccorsi, M. P. Datcu, DLR Standort Oberpfaffenhofen (Germany) High Resolution (HR) Single Look Complex (SLC) Synthetic Aperture Radar (SAR) observations, mainly of strong scattering scenes or objects show phase patterns. Phase patterns may occur due to the system behavior or they may be signatures of the imaged objects. Since state of the art stochastic models of SAR SLC describe mainly the information in the amplitude/intensity, now studies are needed to elaborate better models for the full information content. Thus, new statistical models of HR SLC SAR are proposed, they aim at the characterization of the spatial phase feature of SLC PolSAR data, i.e. they describe multi-band, complex valued textures. The definition of texture must be changed because it is not anymore characterizing the optical features but the electromagnetic properties of the illuminated targets. The content of the SAR image is characterized from its own geometry, which differs from the real one of the illuminated scene, and is dominated from strong scatterers. Nevertheless we are going to accept the classical texture definition, inherited from computer vision, in homogeneous areas and, furthermore, we are going to extend it for a characterization of isolated and structured objects The proposed models are in the class of Simultaneous AutoRegressive (AR) defined on a generalized set of cliques in the pixel vicinity. Models may have different orders, thus capturing different

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

105

Conference 6746: SAR Image Analysis, Modeling, and Techniques


and surface roughness were available. Subsequently, the test was extended to the whole area for generating soil moisture maps from the available ENVISAT images in HH polarization. The information on soil roughness of the whole area, needed as ANN input, was derived from the corresponding images in HV polarization. The obtained results showed a satisfactory agreement with ground truth data and meteorological conditions, and enabled us to generating maps with 4-5 levels of soil moisture. conditional probability is de-normalized in such a way that it holds one, and hence the change feature is identically zero, when the joint probability attains its maximum on either the row or the column, corresponding to the first or second image, respectively. Depending on such a scaling, some changes may be emphasized or concealed. All joint probabilities are calculated by digitizing the 2D histogram (scatter-plot) of locally windowed backscatters, and smoothing the outcome by means of a Gaussian filter. Experimental results carried out on two couples of multitemporal RADARSAT SAR images referring to the flooding event occurred on 25th - 26th November 2005 along the Tiber River demonstrate that the proposed IT feature outperform the logratio in terms of capability of discriminating changes and identifying flooded areas.

6746-06, Session 2
ICA decomposition of HR SAR images: application to urban structures recognition
H. Chaabouni, M. P. Datcu, DLR Standort Oberpfaffenhofen (Germany) Synthetic Aperture Radar (SAR) images are formed, by coherently adding the scattered radiations from the components of the illuminated scene objects. In the case of the High Resolution (HR) SAR, mainly for man made structures (from which the urban areas are mostly made), the resolution cell is comparable with the scale of the imaged objects. Thus, the random phase assumption does no longer, hold for typical high resolution urban SAR scenes, where prominent human-made scatterers, with near regular shape and sub-meter size, lead to correlated phase patterns. Indeed, when the pixel size shrinks to a critical threshold of about 1 meter, the reflectance of built-up urban scenes becomes dominated by typical metal reflectors, corner-like structures, and multiple scattering. The resulting backscattering becomes thus, hard to be modeled, but one can try to classify a scene based on the phase characteristics of the neighboring image pixels. Following this approach, one alternative for a better understanding of the HR SAR scenes, could be the Independent Components Analysis (ICA) decomposition. Indeed, the ICA is a de-mixing process whose goal is to express a set of random variables as linear combinations of statistically independent component variables. Such an approach could be useful for the recognition of urban structures, in HR SAR images, if we make the assumption that, the SAR image is a superposition of different sources. In this paper, the results on learning and decomposing the SAR image, on ICA generated bases, are presented and compared with eigenspace decomposition of the targets. The case of single, dual and quad PolSAR HR SLC images, are studied.

6746-08, Session 2
Inferring Titans surface features by means of Bayesian inversion algorithm applied to radar data
B. Ventura, Politecnico di Bari (Italy); D. Casarano, CNR-IRPI (Italy); C. Notarnicola, Politecnico di Bari, Dip. Interateneo di Fisica (Italy); M. A. Janssen, Jet Propulsion Lab. (USA); F. Posa, Politecnico di Bari (Italy) Titans surface, before Cassini reached the saturnian system, was essentially unexplored. Nowadays, after two years of the Cassini mission, a great amount of data dealing with Titans surface has been collected by means of a radar able to operate in four modes: radiometry, scatterometry, altimetry, and synthetic aperture radar (SAR) imaging, mounted on the Cassini spacecraft. In particular the analysis derived from the SAR imagery reveals a complex surface with peculiar features such as: dark and bright areas, periodic structures (sand dunes) and, above all, hydrocarbon lakes. A first analysis of the Radar Cross Section, RCS, of these features, by the means of the Integral Equation Method, IEM, has been carried out resulting in variations of the RCS in excess of 20 dB between the brightest and the darkest areas. In this paper, the investigation of Titans surface parameters (physical and morphological) has been carried out by the means of the Bayesian inversion technique. In order to extract surface parameters, hypotheses on dielectric constant values and surface parameters have been formulated. These hypotheses and the Titans surface morphology have been used in simulations with the IEM approach in order to build a training data set of expected RCS. Starting from these model simulations and the corresponding radar data, a family of probability density functions (pdfs) has been built through a comparison between real and simulated radar data; the sensor noise has been also taken into account. Then, each pdf can represent a surface single hypothesis. These pdfs have been used in an inversion procedure, based on Bayesian methodology. This approach has been initially applied to the area characterized by the presence of hydrocarbon lakes and then extended to other areas. Beside the estimation of dielectric constant and surface parameters, this inversion allows to identify the predominant scattering mechanism (surface or volume); in the case of the dunes, it allows also the estimation of the tilt angle and the average dune height. The roughness parameters for the hydrocarbons lakes were also matched with the findings of a gravity capillary wave model, implemented to study the dependence of radar-scale roughness of the liquid surfaces on the wind speed. This model uses a set of viscosity, surface tension and density parameters expected for the hypothesized liquid hydrocarbon mixture. On the other side, this methodology allows inferring information on the optical thickness of lakes. As a last step, the extracted parameters will be compared to radiometric data in order to verify that the hypotheses posed to surface parameters can be also supported by other kind of data.

6746-07, Session 2
An information-theoretic feature for identifying changes in multitemporal SAR images: an evaluation for the detection of flooded areas
B. Aiazzi, Istituto di Fisica Applicata Nello Carrara (Italy); L. Alparone, Univ. degli Studi di Firenze (Italy); S. Baronti, Istituto di Fisica Applicata Nello Carrara (Italy); T. Moramarco, C. Pandolfo, M. Stelluti, Istituto di Ricerca per la Protezione Idrogeologica (Italy) The role of the space-borne synthetic aperture radar imagery (SAR), for natural risk management, is increasing in recent years. Optical sensors have higher spatial resolution then SAR but can be used only during daytime and in good weather conditions. Conversely, SAR can acquire data at day or night indifferently and in particular can penetrate through clouds, thus being able to provide images useful for estimating the extension of inundated areas and delineate flood boundaries. Possible limitations induced by the presence of wind and vegetation over inundated areas can be tackled by analyzing multi-temporal overlying patterns and by photo analysis. The efficiency of the procedure for flood mapping can be further increased, especially if the remote sensing images are integrated with existent ground-based ancillary data. In this work, a novel pixel feature suitable for change analysis is derived from information-theoretic concepts. It does not require preliminary de-speckling and is capable of providing accurate change maps from a couple of SAR images. The rationale is that the negative of logarithm of the probability of an amplitude level in one image conditional to the level of the same pixel in the other image conveys an information on the amount of change occurred between the two passes. The

6746-09, Session 2
Multibaseline interferometric SAR at millimeterwaves
H. W. Essen, T. Brehm, FGAN-FHR (Germany); S. Boehmsdorff, WTD 52 (Germany)

106

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6746: SAR Image Analysis, Modeling, and Techniques


Inverse Synthetic Aperture Radar has the capability to provide the user with the 3-D-Information of land surfaces. To gather data with high height estimation accuracy it is necessary to use a wide interferometric baseline or a high radar frequency. However the problem of resolving the phase ambiguity at smaller wavelengths is more critical than at longer wavelengths, as the unambiguous height interval is inversely proportional to the radar wavelength. To solve this shortcoming a multiple baseline approach can be used with a number of neighbouring horns and an increasing baselength going from narrow to wide. The narrowest, corresponding to adjacent horns, is then assumed to be unambiguous in phase. This initial interferogram is used as a starting point for the algorithm, which in the next step unwraps the interferogram with the next wider baseline using the coarse height information to solve the phase ambiguities. This process is repeated consecutively until the interferogram with highest precision is unwrapped. On the expense of this multi-channel-approach the algorithm is simple and robust, and even the amount of processing time is reduced considerably, compared to traditional methods. The multiple baseline approach is especially adequate for millimeterwave radars as antenna horns with relatively small aperture can be used while a sufficient 3-dB beamwidth is maintained. The paper describes the multiple baseline algorithm and shows the results of tests on a synthetic area. Possibilities and limitations of this approach are discussed. Examples of digital elevation maps derived from measured data at millimeterwaves are shown. 1, the opportunity is unique because now there is the possibility to acquire data with different frequencies on the same area. In the past these occasions were related only to some short-term missions such as the SIRC-XSAR in 1994. As demonstrated in literature, different radar frequencies are sensible to different terrain features, and the combination of two or three frequencies can improve the retrieval process. The sensitivity to soil moisture of the backscattering coefficient measured at low microwave frequencies (P- to L-band) is a well-known phenomenon. In the past years, research activities carried out worldwide have demonstrated that sensors operating in this portion of the microwave spectrum are able to detect soil water content. Anyway for a long time L-band data have not been available. Thus, research for the retrieval of soil moisture has been focused on the potential of existing sensor characteristics, i.e. the Cband, which is operational on ERS-2, RADARSAT, ENVISAT. C-band backscattering is still sensitive to soil moisture, but it is significantly influenced by vegetation (especially at VV polarization) and surface roughness too, so that the estimation of spatial variations of moisture with the accuracy requested in many applications is still problematic and in many cases it requires complicate correction procedures to take into account the impact of vegetation and roughness. X-band data are greatly influenced by vegetation and in some cases, very dense vegetation can hinder the signal to reach soil and detect its features. On the other hand, this kind of data can be used to infer information about the vegetation impact. In this context, the availability of different frequencies can help the inversion process. The proposed inversion algorithm is based on Bayesian techniques which makes full use of measured data and simulated data. The data have been simulated by using the Integral Equation Model (IEM) for C and L band data and the GO (Geometric Optics) for X band. The main purpose of the algorithm is to retrieve information on soil moisture and roughness. The dataset used to verify the reliability of the algorithm is based on the SIRC-XSAR data acquired in 1994 on Southern Italy (Matera test site). In this experiment the sensor acquired contemporary images in the three bands over selected test sites. The Matera test site was also intensively sampled in terms of soil moisture and roughness measurements.

6746-10, Session 3
Velocity estimation of slow moving targets in ATInSAR systems
V. Pascazio, Univ. degli Studi di Napoli Parthenope (Italy); G. Schirinzi, Univ. degli Studi di Cassino (Italy); A. Budillon, Univ. degli Studi di Napoli Parthenope (Italy) Ground moving target indication (GMTI) is a very difficult problem, due to the difficulty of separating the signal returned from a moving target from the stationary background (clutter). In some methods, GMTI is accomplished by enhancing the target Doppler signature against the competing ground clutter returns. Recent clutter suppression techniques use space-time adaptive processing (STAP) and time-frequency processing. While these techniques are effective in improving the detect ability of fast targets, for slowly moving targets the signal from clutter separation is more critical. A method that can be used to detect slow ground moving target is Along Track Interferometry (ATI). Along Track Interferometric Synthetic Aperture Radar (AT-InSAR) systems use more than one SAR antennas (typically two), mounted on the same platform and displaced along the platform moving direction. The information about the radial velocity of the moving target is estimated from the interferometric phase of the images using statistical estimation techniques based on the statistical distribution of the measured phases. In this paper we analyze the radial velocity estimation with respect to ATI system parameters, such as velocity values, clutter and noise power levels considering a deterministic target whose velocity is estimated using a Gaussian model. This model allows to take into account the lack of knowledge of the target radar cross section (RCS) values and provides an analytical form for the interferometric phase probability density function, which in many cases well approximates the true distribution. Moreover, as it will be shown, adoption of a Gaussian model for the moving target RCS instead of the actual deterministic model will not impair significantly the GMTI performance.

6746-12, Session 3
Evaluating SRTM and ASTER DEM accuracy for the broader area of Sparti, Greece
K. G. Nikolakopoulos, P. I. Tsombos, A. Zervakou, Institute of Geology & Mineral Exploration (Greece) One of the major projects of the Institute of Geology & Mineral Exploration (IGME) is called Urban Geology. In the frame of that project there is need for a high accuracy DEM covering the whole country. The DEM should be used for the orthorectification of high resolution images and other applications such as slope map creation, environmental planning et.c. ASTER and SRTM are two possible sources for DEM covering the whole country. According to the specifications the ASTER vertical accuracy of DEM is about 20m with 95% confidence while the horizontal geolocation accuracy appears to be better than 50 m. More recent studies have shown that the use of GCPs resulted in a plannimetric accuracy of 15 m and in a near pixel size vertical accuracy. The Shuttle Radar Topography Mission (SRTM), used an Interferometric Synthetic Aperture Radar (IFSAR) instrument to produce a near-global digital elevation map of the earths land surface with 16 m absolute vertical height accuracy at 30 meter postings. An SRTM 3-arc-second product (90m resolution) is available for the entire world. In this paper we examine the accuracy of SRTM and ASTER DEMs in comparison to the accuracy of the 1/5.000 topographic maps. The area of study is the broader area of Sparti, Greece. The ASTER and the SRTM DEM were divided in four major categories (low, medium and high elevation areas, urban centre). After a first control for random or systematic errors a statistical analysis was done. A DEM from digitized contours from the 1:5.000 topographic maps was created and compared with ASTER and SRTM derived DEMs. Almost three hundreds points

6746-11, Session 3
Combination of X, C and L band SAR images for retrieval of surface parameters
C. Notarnicola, F. Posa, Politecnico di Bari, Dip. Interateneo di Fisica (Italy) In the last years many efforts have been made in order to build and launch satellites which will have on board SAR sensors working at different frequencies. This is the case of ALOSPALSAR that was launched in 2006 and of TerraSAR-X and COSMO-SkyMed that will be launched during this year. Along with C-band sensors such as ENVISAT-ASAR and RADARSAT

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

107

Conference 6746: SAR Image Analysis, Modeling, and Techniques


of known elevation have been used to estimate the accuracy of these three DEMs. Aspect maps were created and compared. The elevation difference between the three DEMs was calculated. 2D RMSE, correlation and the percentile value were also computed. The outliers were detected. The three DEMs were used for the orthorectification of very high resolution data and the final orthophotos were compared.

6746-15, Session 4
The case of PIMS: image information mining in an SAR data ground segment
G. Schwarz, A. De Miguel, M. P. Datcu, DLR Standort Oberpfaffenhofen (Germany) Conventional ground segments for spaceborne SAR instruments collect raw SAR data and convert them into standard image and interferometry products. Typical examples are the rou-tine generation of complex, detected, or geocoded image products. These data are archived and accessible via user interfaces linked to browsing tools that use catalogues with quick-looks, metadata, etc. On the other hand, during the last years considerable effort has been spent in the design of data mining systems and specific image information mining techniques that allow the retrieval of images from large archives based on their content. We will describe how the PIMS ground segment architecture combines product generation, archiving, and cataloguing with image information mining functions permitting automated feature extraction as well as interactive data analyses by users of various disciplines. The system implementation follows general guidelines to be obeyed by todays system de-signers: multi-mission capability with simple interfaces, use of standardized (web) services within well-defined environments, pluggable modules and component-based systems, and workflow managers for distributed computing. We will give first results of the performance of the PIMS system recently installed at the German Aerospace Center, where large size SAR data have to be ingested and analyzed in parallel. After feature extraction and clustering, interactive user interfaces allow for subjective scene labeling and image classification. We will discuss the basic system layout and its most important design options, notably the feature extraction parameterization, the system throughput and performance, the clustering options, and specific experiences gained with users. The existing system seems to provide reliable results for a variety of users.

6746-27, Session 3
Land subsidence monitoring using InSAR and GPS
S. H. Hosseini, M. J. Valadan Zoej, M. R. Mobasheri, K.N.Toosi Univ. of Technology (Iran); M. Dehghani, K.N. Toosi Univ. of Technology (Iran) Land subsidence is caused by human activities in many countries all over the world. Ground water withdrawal is one of the most important factors of land subsidence formation that can cause extremely expensive damages to buildings, road and pipelines. A necessary step to perform a proper analysis of the land subsidence is to obtain accurate measurements of the actual amount of subsidence at certain intervals. Nowadays the InSAR ability to detect the surface deformation has been demonstrated. In this study the InSAR technique is implemented in order to identify the surface deformation in a subsiding area located in north-east of Iran. The area is subsiding due to the excess water withdrawal for the agricultural purposes. The Vertical displacements due to the land subsidence were calculated using InSAR method and ENVISAT data spanning between 2003 and 2005. The subsidence rate was estimated as several centimetres per year using the stacking method. The deformation rate obtained by the InSAR method was then compared to the GPS measurements. The comparison shows the consistency of the InSAR results. An optimum GPS network was then designed using the displacement map obtained from the InSAR in order to achieve a precise time series analysis of the surface displacement.

6746-14, Session 4
Phase information contained in meter-scale SAR images
M. P. Datcu, G. Schwarz, M. M. S. Soccorsi, H. Chaabouni, DLR Standort Oberpfaffenhofen (Germany) The properties of single look complex SAR images have already been analyzed by many in-vestigators. A common belief is that no information can be gained from the phase of each pixel. This belief is based on the assumption that we have random phases when a sufficient number of small-scale scatterers generate uniformly distributed backscatter phases that are mixed in each image pixel. However, the random phase assumption does no longer hold for typical high resolution urban remote sensing scenes, where prominent human-made scatterers with near regular shape and sub-meter size lead to correlated phase patterns. If the pixel size shrinks to a critical threshold of about 1 meter, the reflectance of built-up urban scenes becomes dominated by typical metal reflectors, corner-like structures, and multiple scattering. The resulting phases are hard to model, but one can try to classify a scene based on the phase characteristics of neighboring image pixels. We applied two approaches to quantify the classification potential of these correlated phase patterns contained within SAR images taken from aircraft and satellites. On the one hand, we attempted a Gauss-Markov random field approach with complex numbers. As an alternative, we investigated the performance of a covariance and eigenspace method for small targets. Both techniques led to robust classifications of urban areas. We will compare the performance of the two methods, their pros and cons as well as pros-pects for ICA (independent component analysis) techniques when applied to complex numbers. The comparisons will be based on available aircraft and spacecraft data.

6746-16, Session 4
Improvement in SAR image maximum likelihood classification using adaptive stack filters
M. E. Buemi, M. E. Mejail, M. J. Gambini, J. C. Jacobo, Univ. de Buenos Aires (Argentina) Stack filters are a special case of non-linear filters. They have a good performance for filtering images with different types of noise while preserving edges and details. A stack filter decomposes an input image into several binary images according to a set of thresholds. Each binary image is filtered by using a boolean function. Adaptive stack filters are optimized filters that compute a boolean function by using a corrupted image and ideal image without noise at the training stage. In this work the behaviour of an adaptive stack filter is evaluated for the classification of synthetic aperture radar (SAR) images. These images are generated by a coherent illumination system and are affected by the coherent interference of the signal backscatter by the elements on the terrain. This interference causes fluctuations of the detected intensity which varies from pixel to pixel. This effect is called speckle noise, which unlike the one present in optical images, is neither Gaussian nor additive; it follows other distributions and is multiplicative. With this aim, a Monte Carlo experiment is carried out: simulated and real images are generated and then filtered with a stack filter trained with one of them. The classification results using maximum likelihood are evaluated and compared with the ones obtained by classifying the images without previous filtering. Our results show a better classification performance for filtered images. Improvements are noticed when feedback is added to the filtering process, that is, when previously output images are used as inputs for the filter.

108

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6746: SAR Image Analysis, Modeling, and Techniques

6746-17, Poster Session


SAR simulation system based windows HPC cluster
H. Guo, National Univ. of Defense Technology (China) SAR system simulation especally rawdata generation and imaging processing is so hard work with huge computation account that there must be high performance computation(HPC)ability. This paper report that a SAR sumulation system was built up based windows hpc cluster. The rawdata generation and imaging processing model is programed in MPICH-2 Distributed Coputation.

6746-22, Poster Session


A new parallel subaperture algorithm for highresolution SAR imaging
X. Liu, Y. Pi, Univ. of Electronic Science and Technology of China (China); C. Leng, China Academy of Launch Vehicle Technology (China); Z. Han, L. Fan, Univ. of Electronic Science and Technology of China (China) both real-time rate and resolution both are key indexes of Synthetic Aperture Radar(SAR) imaging, but there is a conflict between them. Real-time imaging becomes difficult because of the large computational requirement posed by high-resolution processing. Subaperture is widely used in high-resolution SAR. Compared with full-aperture processing, it can compensate the motion errors more accurately and get better images. Whereas, prevalent parallel imaging algorithms are either medium grained or large grained algorithms, which are all based on full-aperture processing. Subaperture, as a kind of small grained partition, cant be applied in these algorithm because of the large communication requirement, which restricts the application of existing algorithms in high-resolution SAR parallel imaging. This paper presents a new high-resolution and parallel imaging algorithm based on subaperture, through which we can obtain high-resolution SAR image while implementing parallel processing. It combines chirp-scaling algorithm with subaperture, which are both frequently used for high-resolution SAR processing. The new algorithm can highly effectively run on parallel computer, in which each note has the same load. It also reduces the large communication requirement posed by three transposes, for the CS procedure needs no communication any more. The experiments about point-target simulation and real echo data on PVM( Parallel Virtual Machine ) have proved that, compared with common CS and serial suaperture algorithm, the parallel subaperture algorithm obviously improves computing efficiency, and it can be applied in real-time parallel imaging. Furthermore it gets better image than CS algorithm based on full-aperture. The algorithm presented in this paper is suitable for high-resolution SAR parallel imaging.

6746-20, Poster Session


A spotlight SAR imaging algorithm based on fractional Fourier transform
M. Yin, Y. Pi, Z. Han, L. Fan, Univ. of Electronic Science and Technology of China (China) A new imaging algorithm based on fractional Fourier transform (FRFT) for Spotlight SAR is proposed in this paper. The algorithm performs the range processing by frequency scaling and the azimuth processing by FRFT. The azimuth processing of conventional SAR imaging algorithms is based on spectral analysis but in this paper we process the azimuth signal by FRFT based on the time-frequency analysis. FRFT is a generalized form of the well-known Fourier transform (FT). If the FT operator can be considered as an anticlockwise rotation in the time-frequency plane over an angle , the FRFT can be regarded as an anticlockwise rotation over arbitrary real angle. The FRFT has a good energy focusing ability to chirp signals because the chirp signal forms a line in an appropriate fractional Fourier domain, so we can separate different chirp signals by FRFT. As we known, the azimuth signal of spotlight SAR is approximately a set of chirp signals with the same time duration and chirp modulation rate and different Doppler frequency centers, so we can use FRFT to process the azimuth signal to form image. In this paper, we provide the detail analysis of the derivation of the new algorithm mathematically, and in the end of the paper some simulation experiments show that the new algorithm can offer high sidelobe suppression and high focused image compared to the conventional algorithm. Finally a spotlight SAR image formed by the new algorithm shows the validity of the algorithm.

6746-23, Poster Session


The target detection of SAR based on the corrected clutter position estimation
C. Leng, H. Chen, Y. Ning, China Academy of Launch Vehicle Technology (China) For SAR image target detector, one problem concerned is the influence of multi-look on detector. Because of serious coherent spots in SAR image, for the purpose of improving the quality of SAR image, usually non- coherent smoothing technology is used, videlicet, multi-look. Because distributing format of clutter will change with the look numbers changing after multi-look Processing, look number will affect clutter position estimation, in other words, look number will affect performance of the second type of detector. By comparing single-look position estimation with multi-look position estimation, this paper draws the conclusions that middle value is stability estimation, while mean value is instability estimation, and the availability of mean value estimation is obviously higher than that of middle value estimation when look number goes up. In order to make a compromise between the stability and availability, this paper proposes a detecting method based on the corrected clutter position estimation. In this paper we have deduced the stability and availability of clutter position estimation, the average value estimation and middle value estimation, which can construct the choice of estimation method, and we proposed a way of target detection base on the corrected clutter position estimation. In this way, it decayed the influence of clutter position estimation in single look map or in multi-look map.

6746-21, Poster Session


SAR image coregistration based on nonstationary scatterers
J. Wang, Y. Pi, Univ. of Electronic Science and Technology of China (China) The precise coresgistration can improve the quantity of interferogram. Over the last years, a variety of SAR image coregistration algorithms have been proposed in the literature of remote sensing field. Most of them are directly introduced into SAR complex data from optical real image discarding the phase information. As a result, the selection of passing pairs may be improperly reflects true backscattering property. Nonstationary scatterers show varying electromagnetic behaviors as they are illuminated from different positions and at different frequency components during SAR integration. Nonstationary scatterers have never been developed for application to SAR image coregistration. In this paper, the potential of introducing nonstationary scatterers to SAR complex image coregistration is evaluated to increase the quality of interferogram. First, Using deconvolution, synthesized SAR images are decomposed into subaperture datasets, which correspond to the scene responses under different azimuthal look angles. A statistical analysis of subaperture datasets permits to clearly discriminate nonstationary scatters that showing varying behaviors during the SAR integration. After some common nonstationary scatterers have been selected in images(master and slave), the passing function can be obtained through coarse and precise coregistration. Airborne L-band repeat-pass interferometric data of the German Aerospace Center(DLR) experimental airborne SAR is used to validate the method.

6746-24, Poster Session


Resolution improvement in both azimuthal and range directions with repeat-pass spaceborne SAR
Y. Zhang, Ctr. for Space Science and Applied Research (China); X. Shi, Ctr. for Space Science and Applied Research (China) and Graduate Univ. of the Chinese Academy of Sciences (China); J. Jiang, Ctr. for Space Science and

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

109

Conference 6746: SAR Image Analysis, Modeling, and Techniques


Applied Research (China) High resolution Spaceborne SAR images are increasingly demanded in a lot of applications both in military and civilian fields. However, higher resolution of SAR image usually means larger transmitted signal bandwidth and larger synthesized aperture. For spaceborne SAR, this means the complexities of satellite and radar may increases in a geometric series correspondingly because of higher power, larger data storage and higher date rate for transmission are required. This paper introduced a simple method to obtain a much higher resolution than the designed resolution in both azimuthal and rang directions by synthesizing the data taken by repeat passes without increasing any complexity in SAR hardware and satellite platform. The basic idea of the method was to firstly establish the equivalence between the signal models of repeat pass SAR signals in both azithumal and range directions and the signal model of stepped frequency chirp signals (SFCSs) when some conditions were presumed, i.e. for range direction, a interferometric condition was required and for azimuthal direction, a small squint angle increase between repeat passes was required, and then using the already proposed method for SFCSs compression to process the data of repeat passes. In the course of processing, each observation in range direction or in azimuthal direction was treated as a subchirp in SFCSs. The major facts affecting the final resolution one could get were investigated and found they were the relative range measurement accuracy and the absolute squint angle measurement accuracy between repeat passes. Detailed derivations and simulations were presented to show the effectiveness of the method.

6746-25, Poster Session


A novel edge detection operator in SAR image
W. Li, Northwestern Polytechnical Univ. (China) Most traditional approaches, such as Canny operator, Sobel operator, are effective to detect edge in images, even which disturbed by additive noise. Unfortunately, they all do not wok well when they are employed to detect edges of SAR image for the reason that the SAR image is polluted by the multiplicative noise which satisfied the Gamma distribution. In this paper, a new approach based on Canny operator is proposed here, in which the kernel function is modified according to the noise, and preprocessing is performed to change noises distribution function. The mathematical theory and simulation all show its superior to Canny operator and others.

110

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6747: Optics in Atmospheric Propagation and Adaptive Systems


Monday-Tuesday 17-18 September 2007 Part of Proceedings of SPIE Vol. 6747 Optics in Atmospheric Propagation and Adaptive Systems X

6747-01, Session 1
Millimeterwave propagation over sea in tropical regions
H. W. Essen, H. Fuchs, FGAN-FHR (Germany); J. Foerster, Forschungsanstalt der Bundeswehr fur Wasserschall und Geophysik (Germany) A wide field of applications ranging from traffic control in coastal regions over protection against terrorist attacks and naval applications, demands sensors with a stable performance under any environmental conditions. Electro-optical sensors are competing with radar. The performance of radar sensors operating within the marine boundary layer is influenced by the actual atmospheric conditions, the sea surface and the geometry between radar and reflection point. To assess the propagation within defined layers, experiments were performed using a radar operating against reference reflectors carried on naval vessels, which moved on outbound and inbound courses. Based upon measurements of atmospheric properties and sea surface parameters, refractivity profiles were calculated and using the parabolic equation model TERPEM, propagation factors were determined. In the framework of a cooperative program between Singapore and Germany new radar measurements have been done in the sea area around Singapore, which is a typical tropical environment. The data have been analyzed and propagation models have been tested using the relevant environmental information. It turned out that atmospheric conditions exist, which are considerably different from the known European Sea situations. The paper describes the experimental approach and discusses results for tropical conditions comparing those with results from European coastal environments.

A meteorological bulk model (such as PIRAM developed in France or LWKD developed in Canada) can easily be linked with a ray tracing algorithm to predict the refraction effects for the optical transmission bands. For radar studies in maritime environments, the refractivity vertical profiles are generally used as input data of propagation codes based upon the parabolic equation method. Bulk models apply the Monin-Obukhov similarity (MOS) theory. In this theory, the choice of the stability functions is one of the major keystones regarding the accuracy of the obtained results. Many measurements campaigns conducted in the past years and concerning radar or EO propagation showed that, under unstable atmospheric conditions (i.e. sea temperature greater than the air temperature), the bulk models results are truly reliable. It is well known that the validity of the MOS theory is limited under stables cases ((i.e. air temperature greater than the sea temperature). Moreover, experimental data related to stable cases are not so frequent and generally their poor accuracy does not allow any quantitative conclusion. Fortunately, during the VAMPIRA trial (conducted during the spring 2004 by the NATO AC/323 SET-56/RTG32), some very good data have been obtained during stable cases. During the last year conference, Dion et al. showed that the choice of the stability functions proposed by Kondo in 1975 and implemented in PIRAM and LWKD could lead to some discrepancies between the modelling and the measurements of apparent elevation angles of fixed targets. In this paper, we will investigate the effects of alternative formulations. These effects will be discussed for EO applications as well as for radar applications.

6747-04, Session 1 6747-02, Session 1


Prediction of IR transmission in a coastal Baltic environment: comparison of model predictions and measurements
D. Dion, Jr., Defense Research Establishment Valcartier Canada (Canada); L. Gardenal, Nurun, Inc. (Canada); H. H. Vogel, Danish Defense Research Establishment (Denmark) During VAMPIRA, conducted in Eckernfrde, Germany, in the Baltic, observations were performed in the mid-wave and longwave IR of blackbody-targets mounted on a ship that sailed away from the coast. Day and night sessions of varying visibility and wind speed were carried out and analysed. In this paper, atmosphere losses in the 2 IR bands are inferred from the VAMPIRA tracking sessions and compare with model predictions. We use MODTRAN for the calculation of gaseous attenuation in conjunction with several modal aerosol extinction models currently available, namely: NAM (as in MODTRAN), WKD (as in IRBLEM), ANAM3 and MEDEX. The various models are compared and put in their historical context. We found that for VAMPIRA, the 3-mode models, NAM and WKD, describe better the effective atmospheric losses than the 4-mode models ANAM3 and MEDEX.

The SAPPHIRE trial: investigations on angular deviation caused by refraction


K. Stein, D. P. Seiffer, FGAN-FOM (Germany) The NATO Panel SET-088 TG-51 has the charter to investigate infrared research topics relating to Littoral Ship Self-Defence. The two main research areas for TG-51 are low-altitude maritime IR propagation phenomenology and ship signature properties. Atmospheric scintillation and refraction prediction models were validated in several trials conducted by different NATO groups. So far most trials were conducted in cold waters. In June 2006, TG 51 performed the SAPPHIRE trial (Ship and Atmospheric Propagation PHenomenon InfraRed Experiment) to collect data in littoral areas under conditions of warm sea temperatures. The location of the trial was the US Naval Research Laboratorys Chesapeake Bay Detachment (CBD) field site on Chesapeake Bay. The objectives of the trial were to validate ship signature models and scintillation/refraction models. This paper gives a description of the SAPPHIRE trial. The purpose of FGAN-FOM was to investigate the influence of changing weather conditions on the apparent elevation of a target. Therefore, we setup IRcameras and high speed cameras at CBD overlooking Chesapeake Bay. The cameras observed a set of lights installed on an Island in 16 km distance. In this paper we discuss and analyse the measured elevations and compare them to the propagation model IRBLEM (IR Boundary Layer Effects Model by DRDC, Canada.

6747-03, Session 1
Refraction effects under atmospheric stable conditions in coastal environments
J. Claverie, Ecoles de Cotquidan (France); D. Dion, Jr., Defense Research Establishment Valcartier Canada (Canada); K. Stein, Forschungsgesellschaft fr Angewandte Naturwissenschaften e.V. (Germany) The performances of Electro-Optical (EO) systems such as visible or infrared cameras, lasers, operating within the Marine Surface Boundary Layer (MBSL), i.e. at heights up to a few tens of meters above the sea surface, are disturbed by various propagation mechanisms: molecular attenuation, aerosol extinction, refraction and turbulence. Refraction is responsible for focusing and defocusing of rays, detection range limitations, mirage formation and angular deviation.

6747-05, Session 1
Measurements of IR propagation in the marine boundary layer in warm and humid atmospheric conditions
L. T. Heen, E. B. Madsen, P. Steenfeldt-Foss, K. Wikan, H. Fonnum, A. D. van Rheenen, E. Brendhagen, B. M. Almklov, Norwegian Defence Research Establishment (Norway) A multinational field trial (SAPPHIRE) was performed at the Chesapeake Bay, USA, during June 2006 to study infrared ship signature and atmospheric propagation effects close to the sea surface in a warm and humid environment. In this paper analysis is performed on land and ship mounted infrared sources studied

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

111

Conference 6747: Optics in Atmospheric Propagation and Adaptive Systems


using infrared (MWIR and LWIR FPA) cameras mounted at about 4 m above mean sea level. Several meteorology stations mounted on land, on a pier and on a buoy - were used to characterize the propagation environment, while sensor heights were logged continuously. Both sub- and superrefractive conditions were studied. Measurements are compared to results from earlier field trials performed in Norway during typical NorthAtlantic atmospheric conditions (cool air with little water content), and dissimilarities between medium wave and long wave infrared are emphasized. The ship mounted source - a calibrated blackbody source - was used to study contrast intensity and intensity fluctuations as a function of distance. The distance to the apparent horizon is also determined. In addition, normalized variance of intensity for land based sources has been calculated for a number of cases and these values can easily be converted to refractive index structure constant Cn^2 values. Measurement results are compared to results from the IR Boundary Layer Effects Model (IRBLEM)*. * IRBLEM is proprietory to the Department for National Defence of Canada as represented by DRDC-Valcartier.

6747-08, Session 1
Measurements of the relative intensity of ship exhaust gas as a function of distance to infrared sensors
A. D. van Rheenen, E. Brendhagen, L. T. Heen, E. B. Madsen, H. Fonnum, P. Steenfeldt-Foss, K. Wikan, B. M. Almklov, Norwegian Defense Research Establishment (Norway) We present results from infrared imaging experiments, performed under hot and humid conditions at Chesapeake Bay, Maryland, USA in the summer of 2006. Specifically, the objective was to study the intensity of the exhaust gases from a ship at different distances. In particular there is an interest to quantify the intensity decrease of the plume with distance and correlate this with simulations of atmospheric transmission. For this purpose the ship ran a predetermined course making broadside passes at predetermined distances from the shore-based IR camera as part of the course. The distances were 1.6, 2.4, 3.2, 4, 6, and 8 km. The camera is sensitive in the 3 - 5 m wavelength range. Digital recordings were made during the ship broad-side passes. It is challenging to identify gas cloud pixels against a background because the pixels are not necessarily clustered. We present a statistical method to identify the gas cloud pixels and use their averaged intensity as a measure for the estimated temperature. The resulting apparent temperature versus distance data are then compared with simulations using standard atmospheric transmission software.

6747-06, Session 1
Refraction measurements and modeling over the Chesapeake Bay during the NATO (TG51) SAPPHIRE trials, June 2006
A. N. de Jong, P. J. Fritz, TNO (Netherlands) Presently used propagation models predict optical refraction effects due to vertical temperature gradients for paths over water reasonably well when the air temperature is lower than the water temperature (negative ASTD: Air-to Sea Temperature Difference). Due to a lack in experimental data, predictions for conditions of positive ASTD have only been verified on a limited scale so far. Therefore the SAPPHIRE trial, organized in June 2006 including a 16.2 km optical path over the Chesapeake Bay near Washington DC, was a unique opportunity to investigate refraction for positive ASTD conditions. TNO installed a mid-path buoy with a number of accurate temperature sensors, providing the temperature gradient at a height of 3.7 m. In addition a theodolite provided series of absolute Angle of Arrival data (AOA), using a set of mastmounted lamps on the opposite side of the Bay. Images were continuously recorded by means of a time-lapse video recorder. During the analysis of the data a number of hypothetical temperature profiles were fitted to the measured air temperature gradient and the air temperature at zero height. With these data AOAs were calculated with a precision ray-tracing scheme. It was demonstrated, that some of hypothetical profiles provided much better agreement between predicted and measured AOAs than the profiles associated with the so-called bulk model, used in the common propagation models.

6747-09, Session 2
Atmospheric models to analyze and predict the optical turbulence
S. Cheinet, Institut Franco-Allemand de Recherches de Saint-Louis (France); P. Siebesma, Koninklijk Nederlands Meteorologisch Instituut (Netherlands) The performance of an optical sensing system may be strongly affected by the atmospheric turbulence, through signal phase alterations, scintillations, etc. Close to the surface, the turbulence characteristics considerably vary with the local weather. The presentation discusses some current techniques and issues in documenting and predicting the optical turbulence. First, we briefly introduce some standard results on atmospheric turbulence and wave propagation in the atmosphere, emphasizing the role of various scales in the eddy sizes spectrum. Second, we discuss the existing techniques to document the atmospheric turbulence at the scales of interest. It is illustrated that current micro-meteorological models provide an idealized 3D+time view of the large-eddies turbulence, allowing original investigations on its impact on optical propagation. Finally, we discuss the possibility to predict the optical turbulence. It is argued that, whereas present-day weather forecasts may be used for surface-layer applications, this approach faces some scientific challenges when applied to the boundary layer and troposphere.

6747-07, Session 1
Measurements of the vertical radiance profile using infrared sensors
A. D. van Rheenen, Norwegian Defense Research Establishment (Norway); E. Brendhagen, L. T. Heen, E. B. Madsen, Norwegian Defence Research Establishment (Norway); H. Fonnum, P. Steenfeldt-Foss, K. Wikan, B. M. Almklov, Norwegian Defense Research Establishment (Norway) Measurements of the spectral radiance of the sky and the sea, taken near Halifax during the September 2001 SIMVEX trial, indicated that the use of user defined atmospheric profiles, i.e. high altitude atmospheric contributions, were necessary in order to obtain agreement between measurements and results from simulations using atmospheric radiance codes. This paper analyzes data obtained under hot and humid conditions during the SAPPHIRE trial at Chesapeake Bay, Maryland, USA, in the summer of 2006. Digital recordings of the sea and sky background were made using cameras sensitive in both the 3 5 m and 8 - 12 m wavelength range. The center of the field of view of the cameras was pitched from -15o to +15o. In parallel with the imaging experiments, spectrometric data was collected at the same time. In addition, many different types of meteorological data were collected. Measurements of the vertical radiance profile near the horizon will be compared with simulation results from MODTRAN using various meteorological input parameters and meteorological models.

6747-10, Session 2
Scintillation index analysis of optical plane wave propagating through non-Kolmogorov strong turbulence
I. Toselli, Politecnico di Torino (Italy); L. C. Andrews, R. L. Phillips, Univ. of Central Florida (USA); V. Ferrero, Politecnico di Torino (Italy) Optical plane wave propagating through atmospheric turbulence is affected by irradiance fluctuations known as scintillation. Scintillation index of an optical wave in strong turbulence can be analyzed by extended Rytov theory, which uses filter functions to eliminate the effect of cell turbulence sizes that do not contribute to scintillation, and it already has been calculated by Kolmogorovs power spectral density model. However several experiments showed that Kolmogorov theory is sometimes incomplete to describe atmospheric turbulence properly. In this paper, for horizontal path, we use extended Rytov theory to carry out plane wave scintillation index analysis in non Kolmogorov strong turbulence. We do it using a non Kolmogorov power spectrum which uses a generalized

112

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6747: Optics in Atmospheric Propagation and Adaptive Systems


exponent factor and a generalized amplitude factor. Although our final expressions for the scintillation have been obtained by extended Rytov theory, which is necessary to adopt in strong turbulence conditions, they can be used for every turbulence conditions also in weak turbulence. function on the basis of microphysics description of the cloud thanks to a ray-tracing code, seems to be the best way to fulfill the requirement. An optimization of the mixing ratio of four pristine ice crystals shapes is studied to best describe the diversity of shapes encountered in natural cirrus with minimum computation time. The bulk scattering phase functions of seven size distributions are computed from the optimized ice crystals mixing obtained. They are compared to the corresponding phase functions used in remote sensing process of MODIS airborne simulator bands. Finally, their impact on PSF computed thanks to a propagation model based on a Monte Carlo method are discussed in the spectral domain from visible to near infrared.

6747-11, Session 2
Statistical turbulence vertical profiles at the Roque de los Muchachos Observatory and Teide Observatory
B. M. Garca-Lorenzo, Instituto de Astrofsica de Canarias (Spain) A suitable characterization of the vertical turbulence distribution on a site should be based in statistical behaviour such as it is required for other parameters in site testing. We present the statistical results of the optical-turbulence profiles at the Roque de los Muchachos observatory and Teide observatory over annual periods. The data were obtained using the generalized SCIDAR technique at the 1m Jacobus Kapteyn Telescope and 1.5m Carlos Sanchez Telescope under unbiased statistical sample criteria because the campaigns were done monthly in the new moon nights. Statistically, most of the turbulence is concentrated close the observatory level (2400 m above sea level) with no more than two turbulent layers at higher altitudes. The temporal evolution of monthly statistical turbulence profiles indicates that the turbulence is concentrated at lower altitude layers during winter. We find a seasonal behaviour of turbulence structure at the ORM.

6747-15, Session 3
Shift estimation and non-uniformity in sequences of scene images
D. C. Dayton, Applied Technology Associates (USA); J. D. Gonglewski, Air Force Research Lab. (USA) Most non-conventional approaches to image restoration of scenes observed over long atmospheric slant paths require multiple frames of short exposure images taken with low noise focal plane arrays. The individual pixels in these arrays often exhibit spatial non-uniformity in their response. In addition base motion jitter in the observing platform introduces a frame-toframe linear shift that must be compensated for in order for the multi-frame restoration to be successful. In this paper we describe a maximum a-posteriori parameter estimation approach to the simultaneous estimation of the frame-to-frame shifts and the array non-uniformity. This approach can be incorporated into an iterative algorithm and implemented in real time as the image data is being collected. We present a brief derivation of the algorithm as well as its application to actual image data collected from an airborne platform.

6747-13, Session 2
Impact of turbulence on the measurement of CC&D materials in desert area
K. R. Weiss-Wrana, A. Malaplate, Forschungsgesellschaft fr Angewandte Naturwissenschaften e.V. (Germany) Within the scope of the investigation of CC&D materials in desert environment, it is necessary to evaluate the impact of the turbulence on thermal imagers. Turbulence decreases the image quality and so the effectiveness of electro-optical systems. It causes a reduction of the spatial resolution of thermal imagers, which is characterized by the turbulence MTF. The total MTF is the product of MTFsensor and MTFturb. As a figure of merit for the spatial resolution we used the area under the total MTF (MTFA). The turbulence MTF depends on system parameters like spectral range, and optics and on the atmospheric parameters, the structure parameter of the index of reflection, Cn2, and the cross wind. In the past FGAN-FOM has carried out a long-term experiment to measure the diurnal run of Cn2 in arid climate, Negev desert in Israel. Based on our turbulence database we calculated the MTFA for our thermal imagers with diverse lenses under diverse turbulence conditions. The selected Cn2 values are representative for the diurnal run of Cn2 in arid summer, different time of the day, respectively. Results will be presented.

6747-16, Session 3
Design of a Shack Hartmann wavefront sensor using extended source in the infrared
C. Robert, B. Fleury, ONERA (France); S. Magli, L. Vial, Sofradir (France); J. Conan, V. Michau, ONERA (France) Adaptive optics provides a real time compensation for atmospheric turbulence which severely limits the resolution of ground-based observation systems. The correction quality relies on a key component: the wave front sensor (WFS) that analyses the perturbation. ShackHartmann WFS is widely used for measurement of phase perturbations from turbulence. An adaptive optics system in the mid IR providing high spatial resolution for tactical applications is currently designed at ONERA. The IR Shack Hartmann wavefront sensor using an extended source is a sensitive device for atmospheric turbulence and distorted wavefront measurements. This paper discusses the issue of endoatmospheric wawefront sensing. An analytical analysis of the various error terms is presented including anisoplanatism differential scintillation, wawefront sensing using correlation, multi-directional wawefront sensing on an extended source. We then explain and justify the design of the IR wavefront sensor using an extended source. First images with the Shack-Hartmann WFS camera to be delivered in June 2007 will be presented.

6747-14, Session 2
Optimization of the mixing ratio of ice crystal shapes in cirrus clouds for atmospheric point spread function modeling
I. Muguet, ONERA (France) and Univ. de Rouen (France); P. Chervet, ONERA (France); C. Roz, Univ. de Rouen (France) The image of a target observed by an airborne detector is blurred by atmospheric particles, and particularly by complex ice crystals present in cirrus clouds. During last decades, a better characterization of the microphysical and optical properties of cirrus clouds has been achieved through improvement of both in situ measurement and calculation codes. In the case of image transmission through cirrus clouds, a critical issue is to obtain a detailed description of the angular distribution of the scattered radiation in the forward direction. Direct measurement of the scattering phase function provided by a nephelometer can not be used in this case, because it is not sampled enough. Moreover, a previous study has shown that analytical formulation of the phase function based on the asymmetry parameter was not suitable for the viewing configurations studied. Computation of the scattering phase

6747-17, Session 3
Evaluation and progress in the development of an adaptive optics system for ground object observation
G. Marchi, R. Weiss, Forschungsgesellschaft fr Angewandte Naturwissenschaften e.V. (Germany) In the past years one of the compelling problems emerged in the recent conflicts has been the ability to distinguish between threats and friendly forces from a safe distance. In environmental conditions where strong turbulence is present the visible bandwidth of the electromagnetic spectrum is badly affected and no clear image can be distinguished, in some cases even at short distances. It is seldom possible to rely on other systems

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

113

Conference 6747: Optics in Atmospheric Propagation and Adaptive Systems


than direct vision to achieve a clear distinction of the target characteristics. Hereby we present the idea of a new system based on the applications of the already successful concept of adaptive optics based on the use of wavefront sensors and deformable correcting mirrors to compensate the distortion. The research develops on two planes: the first is the characterization of the influence of the atmospheric turbulence on the image when the line of sight lies parallel to the ground. The second is the development of the adaptive optics sensor and of the programs to achieve the requested performances. The achievements the work in progress and the future plans are presented.

6747-21, Session 4
New LGS for large aperture telescope
L. A. Bolbasova, V. P. Lukin, Institute of Atmospheric Optics (Russia) We propose a modified approach [1, 2] to from a laser guide star (LGS) with a wide, collimated laser beam launched through the full aperture of a telescope. The scattering volume is observed from the center of a telescope as an extended source. Using a special form of field stop (diaphragm) uniquely positioned for each subapertures, the wavefront sensor sees only a small part of the source corresponding to the area on the sky, which is cut out by the field stop. We have done the analytical and numerical calculations of variances of the differences angular of arrival between a plane wave (signal from a natural star) and a set of identical spherical waves, formed in the plane of the LGS. Each of these spherical waves subtends the area of a separate subaperture at the telescope pupil. Since the linear size of the subaperture is approximately equal to the coherence radius, the measured wavefront can be restored as a smooth phase function. We also present calculated variances for high modal components of the phase fluctuation. The angular resolution and Strehl parameter were analyzed for the telescope working with the signal of the LGS. It is shown that reduction of residual phase distortions increases with the number of subapertures for a given telescope size. 1. D.F. Buscher, G. Love, R. Myers, Laser beacon wave-front sensing without focal anisoplanatism, Optics Letters, 27, No.3, 2002. 2. D. Bonaccini, V. Lukin, Laser guide star with collimated laser beam for large aperture telescope, Frontiers in Optics 2006. Abstracts. Rochester. USA, p.129, 2006.

6747-18, Session 3
On the problem of beam focusing in the turbulent atmosphere
F. V. Shugaev, E. N. Terentiev, L. S. Shtemenko, O. A. Nikolaeva, T. A. Pavlova, O. I. Dokukina, M.V. Lomonosov Moscow State Univ. (Russia) Two parts of the problem were analyzed. The first one is the adequate description of turbulence. The result is the simulation of the evolution of the refractive index due to turbulence. The second one is the beam focusing on condition that the refractive index is subject to spatial and temporal variations. The turbulence was simulated with the aid of the solution to the Navier-Stokes equations. Two kinds of initial conditions were used: (i) the vortical field was given, the velocity divergence (dilatation) being zero; (ii) the velocity divergence was given, the vorticity being zero. In all cases, the initial values of the density and temperature were constant. The problem is set in the infinite space, the initial data being random functions. The solution of the Navier-Stokes equations was reduced to the solution of integral equations of the Volterra type. The iterative procedure was used. The comparison of the subsequent iterations allows to conclude that the convergence takes place. The problem of compensation for turbulent distortions of a laser beam was solved. The resolving function determines the necessary deformation of the mirror. The knowledge of the resolving function indicates the way to the beam focusing in the turbulent atmosphere.

6747-22, Session 4
Singular phase dynamics in vortical optical beam
O. V. Tikhomirova, V. P. Aksenov, Institute of Atmospheric Optics (Russia) The spatial dynamics of vortex dipole nested in a Gaussian optical beam is analyzed with scalar diffraction theory. It is shown that the generation and annihilation of vortices are accompanied by extreme wave front distortions. Process of nucleation and annihilation of vortex dipole is considered to confirm the vortex aftereffect in the process of transformation of an aberration wave front into a singular one. After the vortex dipole annihilation the absolute values of the average and Gaussian curvatures increase greatly in the local areas of wave front. The phase reconstruction error increases in these areas in the process of wave front sensing. To investigate the features of the vortex aftereffect for simple field the light rays which take part in the combined translational and rotary motion of energy around the field zero-lines are constructed. The phase is computed as a potential of the optical field using the light rays. Integration of the phase gradient along the ray or energystream line allows a unique value of the phase to be connected with a point of the ray. Wave front of the beam is constructed from the computed phase. As in the vicinity of the vortex core the energy-stream line takes a spiral shape the phase incursion along the line increases. After the vortex annihilation and vanishing of the spiral shape the incursion remains and creates the vortex aftereffect in the form of the extreme wave front distortions. The revealed effect should be taken into account when constructing the systems of adaptive optics aimed at functioning in strong turbulence conditions.

6747-19, Session 3
Cross-wind profiling based on the scattered wave scintillations in a telescope focus
V. A. Banakh, D. A. Marakasov, Institute of Atmospheric Optics (Russia); M. A. Vorontsov, Army Research Lab. (USA) Reconstruction of wind profile based on the turbulent spatiotemporal statistics of reflected optical wave focused by the receiving telescope is considered. Both the expression for the spatial temporal correlation function and spectrum and the algorithm of wind profiling based on the spatio-temporal spectrum of intensity of a wave scattered off an diffuse target are presented. Computer simulations performed under conditions of weak optical turbulence shown wind profiles reconstruction by the developed algorithm.

6747-20, Session 4
Effect of phase fluctuations on propagation of vortex beams
P. A. Konyaev, V. P. Lukin, V. A. Sennikov, Institute of Atmospheric Optics (Russia) We present some results obtained by numerical modeling of the propagation of vortex beams with a spiral phase through a randomly inhomogeneous medium being presented by a phase screen placed in the beginning of the propagation path. Such beams, if propagated under conditions of weak turbulence, also experience distortions, like Gaussian beams. The vortex beams being the Laguerre-Gaussian modes are found to have the same broadening properties while propagated through a randomly inhomogeneous medium as the Gaussian beams. The broadening of averaged vortex beams does not depend on the vortex charge and coincides with the broadening of a Gaussian beam.

6747-23, Session 4
Modelling of powerful light pulse propagation in air under backscattering conditions
O. K. Khasanov, T. V. Smirnova, O. M. Fedotova, Institute of Solid State and Semiconductor Physics (Belarus); A. P. Sukhorukov, M.V. Lomonosov Moscow State Univ. (Russia) Powerful light pulse self-action in Kerr medium is known to result in self focusing originated from refractive index transverse gradients. Its longitudinal gradient, which occurs under beam propagation as well, causes partial backward-reflection. Study of this effect is important for potential applications.

114

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6747: Optics in Atmospheric Propagation and Adaptive Systems


In this work the nonlinear dynamics of ultrashort intense light pulse in atmosphere is investigated analytically and numerically. The focusing and defocusing processes are taking into account. Application of the eikonal related to the nonlinear phase increment allows to reduce Maxwell equations to the set of equations of non-linear Schroedinger type for forward and backward (BW) waves. We take into account the photo-induced free-electron plasma generation (FEPG) at femtosecond time scale. Among another processes promoting BW generation we consider higher-order effects such as self-steepening (SS) and space-time self-focusing (STSF). In order to solve the equation system numerically, we introduced a mesh in spatio-temporal domain and approximated the initial task by the system of difference equations realized via iterative alternating direction method involving intermediate layer in z variable at propagation direction. Obtained equation set was solved by sweep technique combined with iterations, while the electron density equation was integrated by Runge-Kutta method. Simulation codes were developed in FORTRAN. Backscattering process is investigated in dependence on the ratio of input pulse power to critical one, pulse duration and its wavelength. The intensity of BW is shown to reach more than 0.1 of forward wave when above ratio exceeds 30:1. The beam exhibits a multifoci behaviour along the propagation axis. The spectrum changes of the propagating beams are investigated. The ways of an identification of BW contribution to pulse propagation caused by each of above-mentioned processes such as the reflection from nonlinear focus, FEPG, SS and STSF are analysed. characterized mainly by air-sea temperature difference, humidity and air pressure. An approach was initiated to retrieve the vertical profile of the refractive index from sea clutter data. The method is based on the LS-SVM (Least-Squares Support Vector Machines) theory and has already been validated on simulated data. Here an inversion method to determine propagation factors is presented based upon data measured during the Vampira campaign conducted as a multinational approach over a transmission path across the Baltic Sea. As the propagation factor has been measured on two reference reflectors at mounted onboard a naval vessel at different heights, the results can be combined in order to increase the accuracy of the inversion system. The paper describes the experimental approach and discusses results achieved with the inversion method.

6747-25, Poster Session


Forecast of the optical turbulence in the marine surface layer based on the products of the numerical weather prediction model
Y. Li, F. Dai, Institute of Applied Physics and Computational Mathematics (China) The forecasts of the optical turbulence in the marine surface layer were made in different seasons based on the numerical products of the numerical weather prediction model. It was found that the seasonal variation of the optical turbulence was small over the tropical oceans, but much larger over the higher latitude oceans, particularly over the eastern coast oceans in the north hemisphere. Generally, the optical turbulence is weaker in the summer hemisphere, but stronger in the winter hemisphere. If the propagation path was the horizontal path 10m above the sea level, the seasonal variations of the Rytov variance, atmospheric coherence length, and the isoplanatic angle are small over the tropical oceans, but are large over the higher latitude oceans. The Rytov variance is generally relatively smaller, and the atmospheric coherence length and the isoplanatic angle are generally relatively larger in the summer hemisphere, but the Rytov variance is relatively larger, and the atmospheric coherence and the isoplanatic angle are relatively smaller in the winter hemisphere. The horizontal patterns of the forecasted surface optical turbulence are very similar to each other based on the two different products respectively from the two different numerical weather prediction models at the same time, but the horizontal pattern of the strength of the forecasted surface optical turbulence is much sharper based on the numerical weather prediction model with higher horizontal resolution.

6747-32, Session 4
Wavefront sensing of an optical vortex and its correction with the help of a bimorph mirror
F. A. Starikov, RFNC-VNIIEF (Russia); V. P. Aksenov, Adopt, Ltd. (Russia); I. V. Izmailov, F. Y. Kanev, Institute of Atmospheric Optics (Russia); G. G. Kochemasov, BIOFIL Ltd. (Russia); A. V. Kudryashov, Adopt Ltd. (Russia); S. M. Kulikov, RFNC-VNIIEF (Russia); Y. I. Malakhov, International Science & Technology Ctr. (Russia); N. V. Maslov, A. N. Manachinsky, A. V. Ogorodnikov, S. A. Sukharev, RFNC-VNIIEF (Russia) One of key trends in the development of modern adaptive optical systems is connected with the correction of scintillation effects arising in laser beams at propagation through inhomogeneous medium. They result in decreasing the efficiency of light energy transportation and the distortion of information carried by laser beam. It is urgent to create the wavefront sensors with high measurement accuracy as well as the adaptive systems for correction of wavefront with screw dislocations acquired in the regime of strong scintillations. The sensing of phase front of the vortex laser beam has been carried out with the help of a Hartmann-Shack sensor. The vortex beam is generated in the form of a Laguerre-Gaussian beam (LG01 mode) with the help of the special helicoidal phase plates manufactured by the kinoform technology. The measured shifts of focal spots on the hartmannogram are compared with the calculated shifts. From the measured wave front tilts the reconstruction of singular phase surface has been performed with using the novel reconstruction technique. The removing of phase singularity from an optical vortex is demonstrated in the close-loop adaptive system including the bimorph deformable piezoceramics-based mirror.

6747-26, Poster Session


Determination of the velocity vector of turbulence layers from G-SCIDAR observations using an algorithm based on wavelet transforms
B. M. Garca-Lorenzo, Instituto de Astrofsica de Canarias (Spain) The G-SCIDAR data massive processing for the determination of the magnitude and direction of the wind require an efficient and contrasted code. We have developed a fully automated algorithm involving wavelet transforms to derive the wind velocity of atmospheric turbulent layers from Generalized SCIDAR measurements. The algorithm makes use of five crosscorrelations of a series of scintillation patterns separated by lapses of t, 2 t, 3 t, 4 t and 5 t. The analysis is performed using wavelet and provides the position, direction and altitude of the different turbulent layers detected in each of the five crosscorrelations. The comparison and consistency of results in consecutive cross-correlations allow the determination of the velocity of turbulence layers and avoid misidentifications associated with noise and/or overlapping layers. The developed software takes into account the projection effects on the observing direction of the actual velocity vector of turbulence layers. We have applied the algorithm to simulated data with excellent results. Velocities derived from real G-SCIDAR observations are compared to the velocities provided by balloon measurements. Our software has been design to analyse huge amounts of G-SCIDAR measurements.

6747-24, Poster Session


Determination of evaporation duct heights by an inverse method
H. W. Essen, H. Fuchs, FGAN-FHR (Germany); J. Foerster, Forschungsanstalt der Bundeswehr fur Wasserschall und Geophysik (Germany); V. Fabbro, R. Duvenois, ONERA (France); C. Bourlier, J. Saillard, Univ. de Nantes (France) The detection and tracking of naval targets, including low RCS objects like inflatable boats or sea skimming missiles requires a thorough knowledge of the propagation properties of the maritime boundary layer. Models are in existence, which allow a prediction of the propagation factor using the parabolic equation algorithm. As a necessary input the refractive index of the atmosphere has to be known. This parameter, however, is strongly influenced by the actual atmospheric conditions,

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

115

Conference 6747: Optics in Atmospheric Propagation and Adaptive Systems

6747-27, Poster Session


Atmospheric turbulence profiling at the Teide Observatory: comparison and calibration of SODAR and SCIDAR measurements
B. M. Garca-Lorenzo, Instituto de Astrofsica de Canarias (Spain) The atmospheric optics is of relevant importance for the modern astronomy due to the excellent image quality requirements of current projects. The spatial and temporal behavior of the turbulence structure is crucial for optimizing the efficiency of adaptive optic (AO) systems whose compensate from atmospheric distortions in the light from astronomical objects. In particular, the input parameters for the design of AO systems working with several deformable mirrors (Multi-conjugated AO) are determined through the statistical behavior of turbulence layers. SCIDAR and SODAR are two independent remotesensing techniques to characterize the turbulent atmospheric structure. We have performed the comparison and crosscalibration of data from these techniques obtained at the Teide Observatory.

turbulence structure. The hybrid profiler has been installed at the Carlos Sanchez telescope at the Teide observatory. The GSCIDAR arm is already working properly and we are still testing the Shack-Hartmann arm.

6747-31, Poster Session


Measurements of transmission in the visible and the IR in the coastal Baltic environment
D. Dion, Defense Research Establishment Valcartier Canada (Canada); L. Gardenal, Nurun, Inc. (Canada); H. H. Vogel, Danish Defense Research Establishment (Denmark); Y. Hurtaud, Ctr. dExpertise Parisien (France); B. Pezery, CTSN (France); K. Stein, Forschungsgesellschaft fr Angewandte Naturwissenschaften e.V. (Germany); A. N. de Jong, TNO (Netherlands); D. Tsintikidis, Space and Naval Warfare Systems Ctr., San Diego (USA); J. L. Forand, Defence Research and Development Canada (Canada) A multinational campaign was organized by the NATO Technical Group 32 to assess transmission in coastal environments: the Validation Measurements of propagation in IR and Radar (VAMPIRA) experiment. VAMPIRA was conducted in the Baltic, near Surendorf, Germany, from 27 March to 4 April 2004. Transmision was measured in the IR and the visible using a diversity of techniques. Transmissometers were installed across the Bay, while aerosol measurements were made on the pier using Particle Measurement Systems (PMS), and visibilitymeters and particle volume monitors (PVM) were deployed on the coast and on a ship. Furthermore, VAMPIRA comprised sessions where blackbodies mounted on a ship were tracked in the midwave and long-wave IR as the ship sailed away. Some VAMPIRA measurements have already been presented at SPIE. The purpose of this paper is to compare transmission measurements and make comparisons with transmissions that can be deduced from the blackbody tracking sessions.

6747-28, Poster Session


Dome-seeing subtraction from G-SCIDAR measurements
B. M. Garca-Lorenzo, Instituto de Astrofsica de Canarias (Spain) We have developed an algorithm to eliminate the dome seeing contribution to turbulence profiles derived from G-SCIDAR data. The algorithm, based on the parity of functions, is completely automated and it takes only a few seconds to process a full night of G-SCIDAR measurements. Seeing measurements obtained from turbulence profiles derived from G-SCIDAR observations and removing the dome contribution with our algorithm are in good agreement with seeing data obtained using DIMMs.

6747-34, Session 2:
Simulation of high-intensive tubular pulsed beam propagation in air,
O. Khasanov, O. Fedotova, T. Smirnova, Y. Petukh, Institute of Solid State and Semiconductor Physics NASB (Belarus): A. Sukhorukov, Moscow State Univ. (Russia) We study high- power femtosecond vortex pulsed beam dynamics in atmosphere under large distance propagation at photoionization conditions. We base on the numerical solution of the system of modified nonlinear (3+1)D Schrdinger equation for an electric field envelope and kinetic equation for free electron density. The competition of Kerr nonlinearity and plasma defocusing is taken into account. The variational approach predicts the vortex pulsed beam multifoci behavior when its input power is close to critical one and the loss is negligible. The input beam power increase results in growth of defocusing factors related to the plasma domination. The interaction of a singular pulsed beam with nonlinear dielectric medium stipulates higher plasma densities in comparison with gaussian pulsed beams, at this the free electron number increases with increasing the beam topological charge. The process of the electron plasma formation due to multiphoton ionisation in this case is responsible for a nonlocal character of the light- matter interaction. This may be the main factor for the attainment of vortex beam stability condition, favouring the establishment of a quasi-soliton regime of radiation propagation over a distance exceeding several diffraction lengths. The effect of delayed (Raman) contribution is considered.

6747-29, Poster Session


Adaptive optics system prototype for the automatic control of geometrical fluctuations in a laser beam in air
S. Grasso, Univ. degli Studi di Roma Tre (Italy); F. Acernese, R. Romano, F. Barone, Univ. degli Studi di Salerno (Italy) In this paper we present a prototype of an adaptive optics system for the control of geometrical fluctuations in a laser beam in air based on the interferometric detection of the beam phase front. We show that this technique is of particular interest when high sensitivity and high band-pass are required for correction of small perturbations. The main technique adopted for the control is based on the interferometric detection of phase-front fluctuations: the beam is made to interfere with a reference beam, properly obtained, and the phase difference is read by means of a pixellated photodiode. The signals are then sampled by a fast automatic control system, which computes also the correction signals, and sent to the deformable mirror. In the paper we show and discuss some experimental results of the technique, discussing the most important limits and estimating the possibility of future developments and improvements.

6747-30, Poster Session


The hybrid Shack-Hartmann/G-SCIDAR instrument
M. d. l. A. Rodriguez Hernandez, J. M. Delgado, Instituto de Astrofisica de Canarias (Spain); J. Jimenez Fuensalida, instituto de Astrofisica de Canarias (Spain); B. M. GarcaLorenzo, Instituto de Astrofsica de Canarias (Spain); E. Hernandez, Instituto de Astrofisica de Canarias (USA); C. K. Hoegemann, H. Vazquez Ramio, Instituto de Astrofisica de Canarias (Spain) We have built a hybrid turbulence profiler measuring simultaneously the atmospheric turbulence structure with a Shack-Hartmann wave front sensor and a G-SCIDAR (scintillation sensor). This is the first instrument combining two different techniques and measuring simultaneously the

116

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6748: Image and Signal Processing for Remote Sensing


Tuesday-Thursday 18-20 September 2007 Part of Proceedings of SPIE Vol. 6748 Image and Signal Processing for Remote Sensing XIII

6748-01, Session 1
An adaptive PCA-based approach to pansharpening
V. P. Shah, N. H. Younan, R. L. King, Mississippi State Univ. (USA) A pixel in multispectral images is highly correlated with the neighboring pixels both spatially and spectrally. Hence, data transformation is performed before performing pan-sharpening. Principal component analysis (PCA) has been a popular choice for spectral transformation of low-resolution multispectral images. Current PCA-based pan-sharpening methods make an assumption that the first principal component (PC) of high variance is an ideal choice for replacing or injecting it with high spatial details from the high-resolution histogram-matched panchromatic (Pan) image. However, experiments show that the low-resolution first PC is not always an ideal choice. This paper proposes a new method to improve the quality of the resultant images that are obtained using the PCA-based pansharpening methods. This approach is based on adaptively selecting the PC required to be replaced or injected with high spatial details. The pan-sharpened image obtained by the proposed method is evaluated using well-known quality indexes. Results show that the proposed algorithm increases the quality of the resultant fused images when compared to the standard approach.

to the horizontal confusion. c) Segment selection Segments possessing a disparity with high confidence are easily classified as building or non building thanks to the corresponding height. Non-paired linear segments are possibly associated to a building supported by a disparity segment based on proximity and parallelism or perpendicularity. APPLICATION The approach was applied to the case of building detection in a stereo pair of Ikonos images of the Bagdad region in the context of GMOSS European network of excellence. The disparity of an area of 1Mpixel was processed in about 1 second, showing the rapidity of the approach. Building segments are clearly highlighted thanks to their height attribute while building shadow cast on the ground, roads and fields are rejected. All buildings of interest were detected by the approach. CONCLUSION We presented an efficient building detection based on the disparity of corresponding linear segments of a Ikonos stereo pair. Candidate segments are filtered based on length, linearity and disparity. The implementation is simple, fast while allowing subpixel accuracy. It was applied to the detection of buildings for change detection or damage assessment.

6748-03, Session 1
Superresolution techniques for remote sensing applications
P. B. Kempeneers, Vlaamse Instelling voor Technologisch Onderzoek (Belgium); J. C. Chan, Vrije Univ. Brussel (Belgium); S. Debacker, Univ. Antwerpen (Belgium); L. Bertels, Vlaamse Instelling voor Technologisch Onderzoek (Belgium); F. Canters, Vrije Univ. Brussel (Belgium); P. Scheunders, Univ. Antwerpen (Belgium) Most remote sensing applications have to deal with tradeoffs that exists in satellite or airborne sensors. Due to a limited number of pixels accross track and a maximum amount of storage onboard, a satellite sensor can only acquire data with high spatial resolution at the cost of a small coverage (swath). The problem with airborne acquisition is merely its high cost. This limits the number of tracks and thus the area that can be covered. Another tradeoff that is typically encountered in airborne sensors, concerns spectral and spatial resolution. The limited integration time does not allow to acquire data with a maximum number of spectral bands at the highest spatial resolution. Several techniques have been proposed to improve the image data quality. Pan-sharpening is an image fusion technique that is used when a high spatial resolution image with low spectral information is available (typically a panchromatic image). Spectral information is then added from a co-registered coarse spatial resolution multispectral image. On the other hand, superresolution techniques increase the spatial resolution if a number of overlapping images is available. If the sub-pixel shift can be estimated accurately enough, a single image with improved spatial resolution can be created. However, several restrictions apply. The objective of this study is to reveal most of these restrictions. We tested on both synthetic and real data, increasing the complexity at different stages. A first image data set was acquired with a digital camera of 120 frames per second, mounted on an airborne platform (helicopter). Several hundreds of frames (images) were obtained over the same scene, while observation angles were only slightly different. This allowed for a considerable increase in spatial resolution while maintaining good image quality. We tested this technique for urban applications such as the detection of house fronts. However, in remote sensing, we not used to such high frame rates. In particular spaceborne sensors acquire imagery in lines (pushbroom or whiskbroom). In this case, overlapping images are only available from multiple acquisitions during different overpasses or using multi-angular viewing satellites such as CHRIS (Compact High Resolution

6748-02, Session 1
Building detection from disparity of edges
C. Beumier, Royal Belgian Military Academy (Belgium) INTRODUCTION The availability of very high resolution satellite images has enabled the automatic detection of man-made structures of large areas for applications such as damage assessment or change detection. In particular, stereo pairs of Ikonos or Quickbird images allow the estimation of the third dimension so distinctive for buildings and responsible for artifacts complicating traditional 2D approaches. Since the areas to be studied are quite large we propose a simple, fast and accurate approach for building detection which computes the disparity at pixels of sufficient gradient and highlights linear segments so representative of building outlines. APPROACH Important cues for buildings in 1-m satellite images are linear edges corresponding to roof limits or ridges and shadow casts. Other important linear edges relate to ground objects like roads or parking lots. We perform building detection or verification thanks to the localization of linear edge segments in stereo pairs. The disparity of corresponding linear segments offers a robust discrimination of building candidates relative to other objects. Unmatched segments in the vicinity of elevated segments are potentially used for additional assessment of building presence or building completion. Disparity estimation along linear edges saves unnecessary computations. a) Linear segments Both images of a stereo pair are first preprocessed to obtain the gradient magnitude and orientation at each pixel. An edge following procedure based on orientation and magnitude groups edge pixels into thin segments. Short segments are filtered out. A linearity criteria based on the minimal inertia moment is applied to keep strongly linear segments. The remaining segments are precisely approximated by a line with sub-pixel localization. b) Segment disparity Linear segments of the left and right images are paired according to their (vertical) mutual coverage, their orientation and the (horizontal) disparity range specified as parameter and roughly known from the image set. A matching pair is given a confidence equal to the average of the left and right segment mean edge magnitudes. The sub-pixel precision of linear segments allows for the fine disparity estimation. Pairing segments to estimate disparity has the additional advantage of tackling nearly horizontal segments which are poorly addressed when the disparity is estimated by single pixels due

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

117

Conference 6748: Image and Signal Processing for Remote Sensing


Imaging Spectrometer) or MISR (Multi-angle Imaging SpectroRadiometer). In both cases, the effect of different observation angles increases the complexity of image coregistration. Moreover, the number of overlapping images is restricted to only a fraction of high speed frame cameras. Another aspect of our study was to investigate the reliability of the spectral information in the reconstructed image. One of objectives was to apply the superresolution techniques to hyperspectral imagery, where spectral signatures must be conserved. Finally, the performance of a classifier was tested on both the original and reconstructed image. Both accuracy and the delineation of class boundaries were assessed. Different superresolution algorithms were implemented for this purpose, including POCS (Projection onto convex sets) and IBP (Iterative Back Projection). We will show the results on the available image data and present the requirements on coregistration, noise level, and different observation angles that apply for the different methodologies. the other hand, data at coarser scales present less precise geometrical details, but also a greater immunity to noise. From this viewpoint, a multiscale approach, exploiting the information at coarser scales to globally identify change areas and the information at finer scales to improve the accuracy of detection of the details, may represent an effective choice. In this paper, a multiscale contextual unsupervised changedetection method is proposed, in which the wavelet and Markov random field (MRF) approaches are combined to integrate the information conveyed by different scales. The method adopts an image-differencing strategy, which is a usual choice for unsupervised change-detection problems involving remotesensing optical imagery. Wavelet transforms are applied to the difference image in order to generate multiscale observations of the image itself. Wavelets provide a two-dimensional multiresolution signal decomposition that allows the input (difference) image to be expanded as the superposition of several components, each highlighting the image content at a given observation scale (Daubechies, 1993). An MRF approach is adopted to fuse the information conveyed by the multiscale data. MRFs permit the introduction of spatial context into pixel labelling problems, such as segmentation and restoration (Dubes et al., 1992). An MRF provides an effective way for blending information about local spatial interaction into a global framework. Here, MRFs are adopted since they allow both the contextual and the multiscale information to be integrated in the detection process by using suitably defined energy functions. Specifically, the MRF-based approach proposed in (Storvik et al., 2005) for multiresolution image classification is adopted and combined with the above-mentioned wavelet and imagedifferencing processing steps in order to extend it to multiscale change-detection. This MRF model is characterized by several internal parameters, that are to be estimated before detection. To this purpose, the iterative expectation-maximization algorithm is adopted (Storvik et al., 2005) that addresses estimation problems characterized by data incompleteness and that converges, under mild assumptions, to maximum-likelihood estimates of the unknown parameters (Moon, 1996). The proposed change-detection method is validated by experiments on real optical satellite images. Specifically, a detailed experimental analysis is carried out by analyzing the impact on the detection accuracy of the choice of the specific adopted wavelet transform, of the number of extracted observation scales, and of the spatial contextual constraints related to Markov modelling. Relevant accuracy improvements are also obtained as compared to the results provided by a previously proposed single-scale approach (Moser et al., 2003). References: Daubechies, I., 1992, Ten lectures on wavelets, SIAM. Dubes, R. C., Jain, A. K.: 1993, Random field model in image analysis, J. Appl. Stat., 16-131-163. Moon, T. K.: 1996, The Expectation-Maximization Algorithm, IEEE Sig. Proc. Mag., 13:47-60. Moser, G., Melgani, F., Serpico S. B., 2003, Advances in unsupervised change detection, in Frontiers of Remote Sensing Information Processing, editor: C. H. Chang, World Scientific Publishing, pp. 405-426. Storvik, G., Fjortoft, R., Solberg, A. H. S.: 2005, A Bayessian approach to classification of multiresolution remote sensing data, IEEE Trans. Geosci. Remote Sensing, 43:539-547.

6748-04, Session 1
A multiscale joint segmentation technique for multitemporal and multisource remote sensing images
L. Galli, D. Passaro, S. Avolio, Advanced Computer Systems S.p.A. (Italy) We developed a new hierarchical joint segmentation technique, which provides an effective fusion of a sequence of multitemporal single-channel SAR images of a given area with multispectral optical images over the same target area. The proposed segmentation method is totally unsupervised, and it allows identifying regions that are homogeneous with respect to the whole data set (both optical and multitemporal SAR images). It is based on a fast multi-scale iterated weighted aggregation method, which has been proved to be effective for the segmentation of data, like remote sensing images, which present heterogeneous multiscale properties. In this paper, we generalize the weighted aggregation method to multitemporal SAR and multisource optical remote sensing data, considering two different joint statistical models: 1) multivariate Gaussian model for the multiband optical images and gamma distribution for the SAR images, 2) again multivariate Gaussian model for the multiband optical images and multivariate log-normal distribution for the SAR images. For gamma distribution, the generalized-likelihood (GL) distance is used, while, for multivariate normal and lognormal distribution both the GL distance and Bhattacharyya distance are analyzed. A quantitative analysis of the proposed joint segmentation technique for the fusion of multitemporal SAR and multispectral optical images is carried out using real images. To this purpose, any desired classification schema can be applied after the segmentation step on the identified homogeneous regions, which allows the full exploitation of the spatial-temporal information available in the multitemporal and multisource data. Results show that the proposed joint segmentation technique, combined with even simple classification methods, greatly improves the discrimination capability of the classifier.

6748-05, Session 2
Multiscale unsupervised change detection by Markov random fields and wavelet transforms
S. B. Serpico, G. Moser, E. Angiati, Univ. degli Studi di Genova (Italy) Multitemporal remote-sensing images represent a powerful source of information for monitoring the evolution of the state of the Earth surface. They convey a great potential for applications like environmental monitoring, urban studies, forest monitoring, or environmental disaster prevention and management. In this context, a relevant task is the identification of the changes that occur in a given monitored area between two distint observation dates. In order to generate accurate change maps, a multiscale approach can be adopted, in which several images at different scales are computed and used for change-detection purposes. The images at the finest scales are likely to highlight many geometrical details, but also to be more affected by noise. On

6748-06, Session 2
An adaptive parcel-based technique robust to registration noise for change detection in multitemporal VHR images
F. Bovolo, L. Bruzzone, S. Marchesi, Univ. degli Studi di Trento (Italy) see attached file

118

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6748: Image and Signal Processing for Remote Sensing

6748-07, Session 2
Radiometric normalization of high spatial resolution multitemporal imagery: a comparison between relative and absolute correction methods
M. El Hajj, A. Bgu, Maison de la teledetection (France) Time series of satellite images acquired with high spatial resolution are an important source of information for change detection and trend analysis in the agricultural domain. For a quantitative use of these data, radiometric normalization is necessary. Our work concerns the sugarcane monitoring on Reunion Island using multi-temporal imagery. In this study, relative and absolute radiometric normalization methods are compared. The aim is to examine whether a relative normalisation method can replace an absolute radiometric correction. The methods were tested using a Spot 4&5 time series belonging to the KALIDEOS-ISLE database, built up by the CNES. The relative correction method consists in isolating points of the image with constant reflectance through time and in establishing calibration equations between a reference date and the other dates. This method doesnt require exogenic data. The absolute correction method is performed by using 6S model. It permits to retrieve the Top Of Canopy reflectance from the Top Of Atmosphere reflectance. 6S needs atmosphere characterization at satellite acquisition times. Normalization results are compared on sugarcane fields. The comparison showed a good correlation between methods: the coefficient of determination values are greater than 0.91 for all the bands, and is 0.93 for the NDVI. However, differences between the normalized temporal profiles of the NDVI were still noticed. These differences can lead to different interpretation on crop development and can have an impact of the biomass estimation. The next step is to test a third method that processes an absolute atmospheric correction followed by a relative correction.

change from no-change observations. In particular, for clustering on the basis of a probability mixture model, the nochange probabilities can be used directly for iterative reweighting. The parameters of the mixture model may be estimated with the EM algorithm [5]. In the present contribution these different re-weighting schemes are compared quantitatively on the basis of partially simulated change/no-change observations as well as with real multispectral imagery. The imagery (ASTER and LANDSAT TM) is obtained under ground cover conditions exhibiting wide variations in the amount of change present. In the case of the clustering alternative, since the EM algorithm is slow to converge for large datasets, a number of strategies are investigated to improve computation efficiency: sub-sampling of the MAD variates, dimensionality reduction with the MAF transformation, multi-resolution clustering, simulated annealing and GibbsMarkov spatial-spectral clustering. Results of the comparison will be presented, and the most effective iteration scheme will be recommended. [1] A. A. Nielsen, K. Conradsen, and J. J. Simpson. Multivariate alteration detection (MAD) and (MAF) post-processing in multispectral, bitemporal image data: New approaches to change detection studies. Remote Sensing of Environment, 64:119, 1998. [2] A. A. Nielsen. The regularized iteratively reweighted MAD method for change detection in multi- and hyperspectral data. IEEE Transactions on Image Processing, 16(2):463478, 2007 [3] M. J. Canty and A. A. Nielsen. Improved automatic radiometric normalization of multitemporal satellite imagery. submitted for publication to IEEE Transactions on Image Processing. [4] M. J. Canty. Image Analysis, Classification, and Change Detection in Remote Sensing, With Algorithms for ENVI/IDL. Taylor and Francis, 2006. [5] C. M. Bishop. Pattern Recognition and Machine Learning. Springer, 2006.

6748-08, Session 2
Investigation of alternative iteration schemes for the IR-MAD algorithm
M. J. Canty, Forschungszentrum Juelich GmbH (Germany); A. A. Nielsen, Danmarks Tekniske Univ. (Denmark) The Multivariate Alteration Detection (MAD) algorithm consists of a canonical correlation analysis of two co-registered, multior hyper-spectral images, followed by pairwise subtraction of the canonical variates [1]. The differences, referred to as the MAD variates, have the desirable property of ordering the change information in the bitemporal data according to similarity, i.e., correlation, of the two images. In addition, the MAD variates have nice statistical properties. For example they are mutually uncorrelated and invariant under affine transformations of the original data. Moreover, the sum of the squares of the standardized MAD variates is approximately chisquare distributed with N degrees of freedom, where N is the dimensionality of the data (number of multispectral bands). In [2] advantage was taken of the latter property to propose an iterative re-weighting scheme (IR-MAD) to further enhance the change information. In this scheme, the image observations are weighted according to their probability of no change as determined by the chi-square distribution, and the MAD transformation is repeated. The procedure is iterated until no further change is observed in the canonical correlations. This provides a better background of no change against which to measure change [2]. An ambiguity arises in the method to be used to determine nochange probabilities on each iteration. In fact, three strategies can be identified: (1) In [2] the MAD variates are standardized at each iteration step in the usual way by dividing them by their standard deviations. (2) In [3] and [4] it is suggested to standardize with respect to the standard deviations determined from the eigenvalues of the canonical correlation for the reweighted data, as these should more closely reflect the statistics of the no-change observations. This procedure is found to give improved results when the MAD transformation is used for radiometric normalization [3]. (3) As an alternative to assigning weights on the basis of chi-square values, one can apply, again after each iteration, a clustering algorithm to discriminate

6748-09, Session 2
An unsupervised method for maximum margin change vector analysis
F. Bovolo, Univ. degli Studi di Trento (Italy); G. Camps-Valls, Univ. de Valncia (Spain); L. Bruzzone, Univ. degli Studi di Trento (Italy) This paper proposes a maximum-margin support vector method for change detection based on change vector analysis (CVA). Specifically, we propose to apply the support vector domain description (SVDD) classifier to deal with high dimensional spectral change vectors (SCVs). The method is automatic and unsupervised, and provides stable results with regard the Bayesian threshold selection method. Good results are observed both with linear and radial basis function (RBF) kernels in two complex change detection real multispectral problems.

6748-38, Poster Session


Segmentation of multi-look fully polarimetric SAR images based on Wishart distribution and MRF
Y. Wu, K. Ji, W. Yu, Y. Su, National Univ. of Defense Technology (China) There is richer information contained in fully polarimetric SAR data, compared with single polarization. Exploiting the information to improve accuracy is important in segmentation of fully polarimetric SAR images. Several frequently used feature vectors and segmentation methods are investigated, and a novel method is proposed for segmenting multi-look fully polarimetric SAR images. In order to fully use the statistical a priori knowledge of the polarimetric SAR data and the spatial relation between neighboring pixels, the Wishart distribution of the covariance matrix is integrated with the Markov random field model, and then the iterative conditional modes (ICM) algorithm is taken to implement the maximum a posteriori (MAP) estimation of pixel labels. Although the ICM has good robustness and fast convergence, it is easily affected by initial conditions, so a Wishart-based ML is used to obtain the initial segmentation map, consequently the statistical a priori

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

119

Conference 6748: Image and Signal Processing for Remote Sensing


knowledge can be exploited more completely in the initial segmentation procedure. Two famous multi-look fully polarimetric SAR images, acquired by the NASA/JPL AIRSAR sensor, are used in experiments. One is the L-band image of Flevoland, The Netherlands, another is the L-band image of San Francisco. The proposed approach is compared with other five ones, such as C-mean method and that based on the multivariate Gaussian distribution. As expected, higher segmentation accuracy is obtained using the novel method, as well as less isolated pixels and better connectivity, due to full consideration of the statistical a priori knowledge of the polarimetric data and the spatial relation between adjacent pixels. flat areas in the DEM and hence all cells can be easily designated a flow direction. This new method notably improved both filling depressions algorithm and determining flow directions algorithm, as well as integrated filling depressions with determining flow directions. Furthermore, unlike the traditional method, which had to scan all the cells in the DEM matrix one by one, the new approach just scanned the remaining data needed to be further processed, by employing two stacks to store them. As a result, the total cells and scanning times were dramatically decreased due to stacks nature of first in last out, and the filled data were all stored in one of the stacks, which was very useful for computing terrain slope and in turn for calculating topographical index. The proposed algorithm, evaluated on 10~100 random surfaces which were generated using the fractal theory and the two-dimensional inverse Fourier method, was many times, even thousands of times faster than the existing ones, which depends on the algorithm used to compare and the size of DEMs. Moreover, the topographical indices of Hanjiang watershed and Jinghe basin were computed using 90m~90m USGS DEMs, which were based on the new method and conventional methods respectively. The result also suggested its superiority over the existing ones. The new algorithm is fully detailed and a pseudo-code was provided in the present paper.

6748-42, Poster Session


A new algorithm for filling the depressions in massive DEM data
J. Xu, Institute of Atmospheric Physics (China) Abstract: Previous research has almost universally recognized that depressions in DEM data are the major obstacle of determining hydrologic flow directions. Numerous methods were proposed to tackle this problem. However, with the advent of remote sensing projects such as SRTM, which makes massive DEM data available, the conventional approaches become time-consuming and inefficient in dealing with the massive data, and it is especially true for a hydrologic model for emergency decision. A new method is presented here. It is innovative because, unlike all existing methods, which process DEM data straightforward without utilizing the topographical features implied in the data, it first categorizes the DEM data into eight groups by an imagery category technique. It maybe, however, cause a new problem since it cost additional time. To eliminate the new problem, another novel algorithm is proposed in this paper so as to categorize the data as quickly as possible. Furthermore, instead of scanning all the cells in the matrix of DEM data, the new method employs two stacks to store the remaining data needed to be further processed. As a result, the total cells and scanning times are dramatically decreased due to stacks good nature: first in last out. The new methods practical computation time, evaluated on 10~100 random surfaces which were generated using the fractal theory and the two-dimensional inverse Fourier method, is many times, even thousands of times faster than the exiting algorithms, which depends on the algorithm used to compare and the size of DEMs. Also, ten 90m~90m USGS DEMs, ranging from 1201~1201cells to 12001~12001 cells in size, were employed to test this new method, which again showed its superiority over prior ones. The new algorithm is fully detailed and a pseudo-code is provided.

6748-44, Poster Session


Unsupervised SAR images change detection with hidden Markov chains on a sliding window
Z. Bouyahia, The Univ. of Manouba (Tunisia); S. Derrode, Ecole Nationale Superieure (France); F. Ghorbel, The Univ. of Manouba (Tunisia) To limit the impact of stationarity of Markov models on change detection results, we propose a new Hidden Markov Chain (HMC) model on a sliding window. This model is applied to detect change between two SAR images (Nyiragongo Volcano, 2002) and compared to other Markov models (with the Kullback-Leibler divergence as detector).

6748-45, Poster Session


Fuzzy pairwise Markov chain and copulas for non-Gaussian image segmentation
S. Derrode, Institut Fresnel (France); C. Carincotte, Multitel (Belgium); S. Bourennane, Institut Fresnel (France) In this paper we propose to segment images with a new model called Fuzzy Pairwise Markov Chain (Fuzzy-PMC) with is the combination of previously published Fuzzy HMC and PMC models. The interest of such an algorithm is to better model transitions between segmented classes in the case of fuzzy gradations between areas in images.

6748-43, Poster Session


A new algorithm to determine flow direction using grid digital elevation models and its application in computing topographical index
J. Xu, Institute of Atmospheric Physics (China) Previous research has almost universally recognized that depressions in DEM data are the major obstacle of determining hydrologic flow directions. Numerous methods were proposed to tackle this problem. However, the conventional approaches failed to process filling depressions and determining flow directions as a whole so that they are too time-consuming and inefficient in calculating topographical index, and it was especially true for massive DEM data. A new innovative approach was proposed and presented in this study. This newly proposed approach, first of all, categorizes the initial DEM data into eight groups according to the elevation values in each cell of the DEM with the quartile classification method, and then sets the category value to each cell of a category as its quartile except boundary ones, and stores the new values in a transient matrix. Secondly, scan the transient matrix from the outside to inside of the matrix, i.e. from the category of the minimum quartile to the one of the maximum, restore its initial value if its initial value is larger than or equal to the sum of its neighbors and H(where H is a tiny number), else sets its value to the minimum of its neighbor value if its quartile value is larger than its neighbors and continue this process until the depressions are filled in the original DEM. So far, there were no

6748-46, Poster Session


Reduced false alarm automatic detection of clouds and shadows on SPOT images using simultaneous estimation
S. Le Hegarat-Mascle, Institut dElectronique Fondamentale (France) and Univ. Paris Sud (France); C. Andre, Institut Pierre Simon Laplace (France) In this study, we propose an automatic approach for detecting and masking clouds, cloud shadows and mist present on optical remote sensing images such as SPOT/HRVIR ones. This detection and masking is necessary for land studies from remote sensing data, such as land cover / land use classification, vegetation and soil moisture monitoring. For studies of the land cover changes, the accurate masking of cloud and their shadows is even more critical since the actual changes generally represent only a small percentage of the whole scene, and their overestimation (due to cloud or shadow presence) is at once important relatively to the actual percentage of changes. The adopted approach is based on image processing techniques rather than on cloud micro-physical knowledge. The algorithm is parameterized by only three parameters that are rather robust since their value was kept identical for the processing of all the 39 SPOT/HRVIR images considered that

120

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6748: Image and Signal Processing for Remote Sensing


corresponds to various acquisition conditions, seasons, and landscapes. Relatively to the classical threshold techniques, our method makes use of three principal cloud/shadow features that allow the reduction of the false alarm rate: (i) Clouds (or shadows) can be viewed as connex objects (once for detection it is no matter to consider as different clouds the different connex parts of a cloud), (ii) Each cloud generates a shadow with similar shape and area, (iii) Knowing the acquisition geometry, the relative position of a cloud and its shadow in the image can be known but one parameter (the cloud altitude) The first feature allows the use of hysteresis threshold techniques or growing region. The two last features allow the mutual validation of cloud and shadow detections. The global algorithm includes then three main steps: (i) First, only clouds are detected. Now, we observed that, on soil/vegetation surfaces, MIR and green band measurements are approximately linearly linked. Here, we propose to estimate this linear relationship, automatically and robustly relative to the presence of cloud pixels, defining an adapted cost function and using dynamic programming. Then, we perform cloud detection using a hysteresis threshold operation on the difference image Xdiff, whose pixels correspond to the difference between green channel observation and its estimation from MIR observation and the estimated linear approximation. (ii) Second shadows are estimated from MIR image. First, from previous cloud detection, the direction d between cloud and shadow locations on the image can be computed (in a determinist way knowing the sun location and sensor acquisition geometry). Then, for each previously detected cloud, its shadow location is obtained minimizing the average value of the MIR values of the pixels corresponding to the translation of the considered cloud pixel along d direction. The shadow is then estimated using a growing region technique from a germ pixel (those having the minimum value) within the estimation of the translated cloud Ct, and using a threshold computed from Ct. (iii) Finally, mutual validation is performed from cloud and shadow respective shapes. Having applied the proposed algorithm to the 39 AMMA database images, its performance was evaluated, showing an effective reduction of the false alarm rate without increasing the misdetection rate. Spectrum are cared in the traditional method. The value influence of image Fourier transform Spectrum upon image texture orientation is ignored, so sometime there is a lager warp between the image texture main orientation by traditional method and the real image texture main orientation. To solve the problem, we proposed the weight covariance matrix (that is, obtaining the covariance matrix by using the Fourier transform Spectrum to weight the coordinate, then getting the texture main orientation by PCA). It is discovered that the frequency spectrum distribution obtained by traditional algorithm of Fourier circular spectrum cannot really reflect its frequency behavior. So a new algorithm with rectangle loops Fourier spectral energy percentage is proposed by analyzing the conjugate symmetry of Fourier transform. In the algorithm, the image power spectrum is divided into 10 equidistant concentric rectangle loops of the frequency domain, and each rectangle loops of the frequency domain is divided into four segments by the image texture main orientation. The ratio of the power spectrum energy within each rectangle loop segments making up the total energy is calculated, and the obtained value is used as the feature of image frequency distribution. Because the symmetry of Fourier transform spectrum, 20 features are obtained by the proposed algorithm. The experiments demonstrate that the new algorithm can reflect the texture features of the image with different frequencies better. The SVM is employed as our Classifier, it is considered a good candidate, because of its good generalization performance, especially for dataset with small number of samples in high dimensional feature space.

6748-49, Poster Session


Hyperpectral image compression using KLT and adaptive directional wavelet transform
J. Wu, Xidian Univ. (China) This paper presents a novel 3D transform zeroblock coding scheme for hyperspectral images. An efficient KLT and adaptive directional lifting-based transform are employed to reduce redundancies in both the spectral and spatial dimensions. A modified context-based 3D zeroblock algorithm is proposed and is used to encode the transformed coefficients.Compared to the above conventional 3D coding methods, for example 3DSPECK and AT-3DSPIHT , the new method can obtain a SNR gain of up to 3dB.

6748-48, Poster Session


Water wake extraction from air photos based on orientation Fourier spectrum
H. Wang, D. Pan, Y. Ding, X. He, State Oceanic Administration (China) It is difficult to directly detect the object in the water with sunglint or cloud shadows. To solve the problem, this paper proposes a water wake recognition method based on orientation fourier power spectrum and Support Vector Machines (SVM). The whole algorithm is composed of two steps. First, we select sub-image samples of the same size with wake texture and without wake texture, then, improve the traditional method which use the Principal Components Analysis and the Fourier transform Spectrum to obtain the orientations of every sumimages from their fourier power spectrum, partition the power spectrum into 20 regions, get texture features by calculating the sum of every region, get the input feature vectors which are translation invariant and rotation invariant, and train the SVM network; Second, we divide the whole image into sub-images with the same size as the first step, calculate feature vectors using the algorithm developed in the first step, and judge whether the sub-image contains a wake texture by the trained SVM network. We will get a binary image by classified result of the whole image, detect lines using Hough transform and judge whether there is a wake in the whole image. From a lot of experiment results, it is proved that the proposed algorithm can detect the wake texture precisely. Because the sub-image sample of wake texture is of orientation in evidence, it is very important to detect the texture orientation features for classifying sub-image samples with wake texture and without wake texture. To detect the image texture main orientation, it is a popular method to use Principal Component Analysis (PCA) and Fourier transform Spectrum. However, only the positions of larger value in image Fourier transform

6748-50, Poster Session


A new target association algorithm based on invariant features in remote sensing images
L. Lei, Y. Su, Z. Li, National Univ. of Defense Technology (China) Target identification fusion based on multi-source remote sensing images can make full use of the redundancy and complementary information from all sensors, acquiring more accurate result of target recognition. One of the absolutely necessary preconditions of identification fusion is target association, which is to determine if the information from two or more images are related to the same target and should be fused together. There are two problems when associating multiple targets in remote sensing images: Firstly, with low temporal resolution observation, the targets kinematic states cannot be estimated accurately and the classic Kalman filtering algorithms are no more applicable. We must seek for other time-independent information as the associating parameters, which can be image invariant features. Secondly, image feature-based single target association algorithms cannot deal with the illegibility arising from multiple targets association, which dont take into account the uncertainty of feature extraction and correspondence. A novel multiple targets association method based on Multiscale Autoconvolution (MSA) feature matching and Association Cost Matrix (ACM) optimization is proposed. At first, the MSA features of each target are computed based on affine invariant feature extracting method, which can overcome changes in targets pose and imaging viewpoint. Secondly, the ACM is constructed based on the dissimilarities of MSA feature matching between any target pairs from two images respectively. Finally, the optimal ACM is estimated using

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

121

Conference 6748: Image and Signal Processing for Remote Sensing


simulated annealing (SA) algorithm on the condition that the whole association cost is minimized. Simulation experiments show that the applicability of the method for multiple targets association. will open new possibilities. Of particular interest herein is the detection and identification of a variety of man-made structures. Meter to sub-meter satellite images have generated new interests in extracting man-made structures such as roads/ highways and buildings in the urban area. The difficulty of using VHR imagery such as IKONOS and Quickbird is that the classification accuracy for such purposes is far from satisfactory. Spectral characteristics of urban land cover classes such as road surfaces, parking lots, roofs and open areas are so similar that they could not be separated using only spectral information. As a result, there are increased interests in incorporating spatial information in image classification of urban land covers using VHR imagery. In this paper we use, what we call Geometric Activity (GA) features, to include spatial information in the classification process. These features are per-pixel indices, which tell something about shapes and sizes of the object the pixel belongs to. For the creation of the features we do not use a segmentation process, since reliably segmenting a satellite image as a very difficult process. Our features are based on a multi-scale investigation of the image. Two different kind of features are used. The first ones are based on ridge features with increasing window sizes. The second ones are based on morphological operations with increasing sizes of the structuring elements (morphological profiles). These multi-scale features lead to a high dimensional feature vector for each pixel. We call this feature vector the signature of the pixel. As a result, a feature extraction process is necessary before including the features in a classification process. Many feature extraction algorithms exist in literature. A good feature extraction algorithm tries to maximize the discriminative power of a feature. These algorithms will use training data containing class information. However, in this paper we show that it is not always optimal to use the same classes and or training data for the classification process and the feature extraction process. Each kind of signature (ridge features, morphological profile, ...) contains its own typical information. In order to extract this information, we should use for each signature its own classes. These classes might be subclasses, superclasses or even completely different classes than the classes used for the classification. An expert should decide on the classes and training data to use for a particular signature. For this purpose, we designed a software tool in which an expert can define the classes and click the image to select training data. After each click, the software performs a feature extraction based on multiple discriminant analysis and the resulting summarised features are presented to the user. To generate useful features even with very little training data, we use a leave one out estimation of covariance matrices. This software tool allows an easy interactive way of generating features based on an experts knowledge of the information content of the signatures.

6748-51, Poster Session


A hybrid classification method using spectral, spatial and textural features for remotely sensed images based on morphological filtering
H. Okumura, M. Yamaura, K. Arai, Saga Univ. (Japan) An new hybrid classification method spectral, spatial and textural information for remotely sensed images is developed. Morphological filtering for a category edge obtained with the maximum likelihood method is employed to improve the misclassification results in the conventional supervised texture classification method. This method consists of the following procedures: 1) Decision of classification category 2) Selection of training areas for each classification categories 3) Classification using maximum likelihood method 4) Edge extraction from classification results 5) Estimation of texture edges using iterative morphological filtering 6) Supervised texture classification based on normalized Zernike moment vector using training areas of procedure 2. 7) Texture assignment using the texture edges of procedure 5 and classification results of procedure 6. The processing results shows higher classification accuracy in comparison with the conventional method.

6748-52, Poster Session


Data fusion of multisensor spaceborne imageries based on 3D discrete wavelet scheme
H. Yoo, Seoul National Univ. (South Korea); K. Lee, Hansung Univ. (South Korea); B. Kwon, Seoul National Univ. (South Korea) The discrete wavelet transform has been widely used in image processing techniques: compressing large sized image sets, reducing the noise of imagery, and fusing and analyzing multiscale images. The wavelet scheme in traditional image processing was based on 2D DWT. 2D DWT focuses on treating only planes rather than analyzing color or multi-spectral imagery. In contrast, the 3D DWT expands planes of 2D DWT into volume data. Remotely sensed imagery also can be considered as a kind of 3D data because it is composed of several bands in order frequency. Nonetheless, few researchers have attempted to analyze imagery using the 3D DWT. This study aims to apply 3D DWT to the fusion of space-borne imagery. To integrate images, we use IKONOS, KOMPSAT and LANDSAT images. LANDSAT image has more bands or spectral information but less spatial information. In contrast, KOMPSAT and IKONOS imagery have more spatial information than LANDSAT imagery. We integrate the spectral information of LANDSAT imagery and the spatial information of other high-resolution imagery. This study provides new information including changes of spatial information and spectral information simultaneously as distinct from previous studies provides mainly fusion techniques to obtain high resolution color imagery.

6748-54, Poster Session


Contribution of Landsat ETM+ thermal band for land cover classification using SMAP and ML algorithms: case study, Eastern Carpathians
A. Ehsani, F. Quiel, Kungliga Tekniska Hgskolan (Sweden) Landsat Thermal band measures the emitted radiation of the earth surface. In many studies the ETM+ thermal band with 60 meter resolution is excluded from processing and classification despite the valuable information content. Two different methods of Bayesian segmentation algorithm were used with different band combinations. Sequential Maximum a Posteriori (SMAP) is a Bayesian image segmentation algorithm which unlike the traditional Maximum likelihood (ML) classification attempts to improve the accuracy by taking contextual information into account, rather than classify pixels separately. Landsat 7 ETM+ data with Path/Row 186-26, dated 30 September 2000 were used. In order to study the role of thermal band with these methods, two data sets with and without the thermal band were used. Nine band combinations were selected based on the highest value of Optimum Index Factor and Principal Component analysis (PCA). Using visual and digital analysis, field observation data and auxiliary map data like CORINE land cover, 14 land cover classes are identified. Spectral signatures were derived for every land cover. Spectral

6748-53, Poster Session


Feature extraction on multiscale geometric features for classification of remote sensing images from urban areas
R. Bellens, L. Martinez-Fonte, Univ. Gent (Belgium); J. Cheung-Wai Chan, Vrije Univ. Brussel (Belgium); S. Gautama, Univ. Gent (Belgium) Recent advances in Earth observation technology have led to an increased availability of data products at very high spatial and spectral resolutions that may open up new areas in the application of satellite imagery. Land-use mapping in complex settings such as urban and suburban environments is one of the domains for which the new Very High Resolution (VHR) data

122

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6748: Image and Signal Processing for Remote Sensing


signatures as well as feature space analysis were used for detailed analysis of effectiveness of the input bands. The result shows that SMAP as the superior method can improve Kappa values for all band combinations with on average 17% compared to ML algorithm. Using all 7 bands both SMAP and ML classifications algorithm achieved the highest Kappa accuracy of 80.37 % and 64.36 % respectively. Eliminating the thermal band decreased the Kappa values by about 8% for both algorithms. The band combination including PC1, 2, 3, and 4 (PCA calculated for all 7 bands) produced the same Kappa as bands 3, 4, 5 and 6. The Kappa value for band combination 3, 4, 5 and 6 was also about 4% higher than using 6 bands without the thermal band for both algorithms. Contextual classification algorithm like SMAP can significantly improve classification results. The thermal band despite the lower spatial resolution can bear information complementary to other spectral bands and can improve classification accuracy. morphometric parameters are strongly related to topographic features and geomorphological processes. Such data allow us to enumerate topographic features in a way meaningful for landscape analysis. Kohonen Self Organizing Map (SOM) as an unsupervised neural network algorithm is used for classification of these morphometric parameters into 10 classes representing landforms such as ridge, channel, crest line, planar and valley bottom. These classes were analyzed and interpreted based on spectral signature, feature space, and 3D presentations of the area. The number of occurrence of each landform class within a processing window of 9*9 was calculated and provided 10 new input bands. Again SOM was trained over these new bands and a map with four classes of dominant landscape features was produced. Both landform maps were evaluated by superimposing contour lines. Result showed that Self Organizing Map is a very promising and efficient tool for such studies. There is a very good agreement between identified landscape forms and contour lines. The procedure is encouraging and can be used for landform or landscape studies.

6748-55, Poster Session


Topographic studies of archaeological areas by remotely sensed images
P. P. M. Merola, D. Guglietta, S. Sampieri, A. Allegrini, Consiglio Nazionale delle Ricerche (Italy) In the recent years there has been a growing interest to use of remote sensing in archaeology. Since 1994 the CNR Atmospheric Pollution Institute has carried out research activities involving both remote sensing and archaeology by MIVIS (Multispectral Infrared and Visible Imaging Spectrometer) airborne hyperspectral scanner to recognize the soil discontinuities connected to buried anthropic elements. The MIVIS spectrometer is a passive remote sensing instrument which collect the earth surface radiation from an airborne platform. The MIVIS data, that can be enhanced, rectified and reclassified using several algorithms and specific software, permit the identification of buried archaeological structures because that any buried ruins, either of human or natural origin, affects over time, soil surface characteristics creating anomalies. These anomalies are due to different factors, such as physical and chemical soil features and vegetation cover status. The above factors are strictly connected and are responsible of surface spectral responses, in particular they characterize complex mixture of vegetation, soil brightness, environmental effects, shadow, soil colour and moisture. MIVIS remote sensing permits to analyse simultaneously a wide range of different wavelengths: these research have shown that the use of the Visible and Infrared channels of the MIVIS sensor are particularly well useful for the study of cover vegetation and the soil moisture. The aim of this work is topographic studies of Salapia, roman centre located in Daunia areas, south of Italy. On this site the major archaeological information were driven by the analysis of the VIS-NIR spectral bands and the Apparent Thermal Inertia image.

6748-57, Poster Session


Improved minimal inter-quantile distance method for blind estimation of noise variance in images
V. V. Lukin, S. K. Abramov, National Aerospace Univ. (Ukraine); K. O. Egiazarian, J. T. Astola, Tampere Univ. of Technology (Finland); B. Vozel, K. Chehdi, Univ. de Rennes I (France) Multichannel (multi and hyperspectral, multitemporal, multipolarization) remote sensing is widely used in different applications. Although considerable efforts are spent on perfection of remote sensing apparatus, noise is still one of the basic factors that deteriorates original image quality and prevents reliable retrieval of useful information from registered data. Because of this, image filtering is a typical stage of multichannel image pre-processing. The most efficient known filters require a priori information on noise type and its statistical characteristics like variance. Such a priori information is also valuable in solving such image processing tasks as edge detection, image classification, etc. Since noise characteristics can vary from one observation mission to another and this holds for each component image of multichannel data, there is a great need in automatic (blind) methods for determination of noise type and its characteristics. Techniques for determination of noise type have been already designed [1]. The task is to provide their appropriate accuracy for images with different content. This is especially problematic if an image contains a lot of texture. Recently we have proposed a method based on minimal interquantile distance [2]. It consists of three basic stages. First, a set of local variance estimates formed in square shape tessellating blocks is sorted with obtaining an array and a preliminary estimate of noise variance Vprel is obtained by finding a minimum of inter-quantile values. For a given number of blocks Nb, differences between the t+s-th and t-th order statistics are calculated and analyzed where s/Nb is set fixed (see plots in Fig. 1). Fig. 1. The plots of vs This estimate Vprel (or, more exactly, the corresponding quantile) also serves as an estimate of a percentage of blocks that correspond to image homogeneous blocks or, in other words, do not relate to edge/detail/texture fragments. Then, a set of normal estimates of noise variance is determined and, at the last stage, they are averaged to get the final estimate Vfin. Since the estimate Vprel is used at the next stages and it can be characterized by inappropriate accuracy, the task is to improve both the accuracy of the estimates Vprel and, respectively, Vfin. Several ways to do this are possible. One way for improving the estimate Vprel is to use interpolation (low pass filtering) of the curve that is rather noisy (see Fig. 1) . It is shown that depending upon image and noise properties the variance of the estimate Vprel can be considerably decreased. In turn, this leads to better determination of the interval of quantiles corresponding to normal estimates of noise variance. Another

6748-56, Poster Session


Landform identification using neural network self-organizing map and SRTM data: case study, Eastern Carpathians
A. Ehsani, F. Quiel, Kungliga Tekniska Hgskolan (Sweden) During an 11 day mission in February 2000 the Shuttle Radar Topography Mission (SRTM) collected data over 80% of the Earths land surface, for all area between 60 degrees N and 56 degrees S latitude. Since SRTM data became available, many studies utilized them for application in topography and morphometric landscape analysis. Exploiting SRTM data for recognition and extraction of topographic features is a challenging task and could provide useful information for landscape studies at different scales. In this study the 3 arc second SRTM digital elevation model was projected on a UTM grid with 90 meter spacing for a mountainous terrain at the Polish- Ukrainian border. Terrain parameters (morphometric parameters) such as slope, maximum curvature, minimum curvature and cross-sectional curvature are derived by fitting a bivaraite quadratic surface with a window size 5*5 or 450 meters on the ground. These

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

123

Conference 6748: Image and Signal Processing for Remote Sensing


modification is a methodology of processing such normal estimates. Instead of averaging some robust methods can be applied. They are investigated and the corresponding practical recommendations are drawn. Finally, there is also an opportunity to partly remove abnormal local variance estimates from original set. This can be done by using the results of image preliminary segmentation [1, 3]. If an estimate is obtained in a block that belongs to a segmented region of an image it is remained for further consideration. Otherwise, an estimate is rejected. In this manner, a trimmed set is obtained. It is demonstrated that analysis (processing) of estimates in this trimmed set is able to provide improved accuracy of final estimates. The efficiency of the proposed method is studied for a wide set of model distribution parameters. Numerical simulation results that confirm applicability of the proposed approach will be presented in the final paper. They also allow evaluating the designed method accuracy. Recommendations on method parameter selection are given. 1. B. Vozel, K. Chehdi, L. Klaine, V.V. Lukin, S.K. Abramov, Noise identification and estimation of its statistical parameters by using unsupervized variational classification, Proceedings of ICASSP, 2006, Toulouse, France, vol II, pp 841-844. 2. V.V. Lukin, S.K. Abramov, A.A. Zelensky, J.T. Astola, Use of minimal inter-quantile distance estimation in image processing, Proceedings of SPIE Conference on Mathematics of Data/Image Pattern Recognition, Compression, and Encryption with Applications IX, SPIE Vol. 6315, San Diego, USA, 2006, 12 p. 3. Lukin V.V., Abramov S.K., Ponomarenko N.N., Vozel B., Chehdi K., Methods for blind evaluation of noise variance in multichannel optical and radar images, Telecommunications and Radioengineering, Vol. 65 (6), 2006, pp. 509-537. strength-of-turbulence parameter, Cn2, in the weak turbulence regime. We will present examples of the typical temporal scales of interest extracted from the datasets using both Empirical Mode Decomposition (EMD) and wavelet analysis. Wavelet techniques search for a specific predefined structure within each scale length; EMD extracts the intrinsic structure for each of the frequency bands in the data. The results of the two processing techniques are compared to illustrate the differences and to emphasise how these algorithms can complement each other.

6748-61, Poster Session


Classification of motion-blurred images using circular moments
C. Toxqui-Quitl, Instituto Nacional de Astrofsica, ptica y Electrnica (Mexico); A. Padilla-Vivanco, Univ. Politcnica de Tulancingo (Mexico); F. S. Granados-Agustin, Instituto Nacional de Astrofsica, ptica y Electrnica (Mexico) In this work, we consider the use of circular moments for invariant classification of images which have been blurred by motion. The test images of the objects under consideration have been acquired when they are vibrating or rotating at different frequencies. A comparative analysis is presented using several circular moment sets as Zernike, Pseudo-JacobiFourier, Orthogonal Mellin-Fourier, shifted Chebyshev-Fourier, radial harmonic Fourier, and Wavelet. An intensity normalization of the input images is done to homogenize them due to inhomogeneous illumination produced by the acquisition. The classification method is tested using images from objects which have intrinsically little differences between them. Experimental results show that, the proposed classification method based in Zernike and Wavelet circular moments can be well addressed to grade images smeared by motion, using test objects under high vibrations or rotations.

6748-59, Poster Session


Time frequency analysis for radar interferogram denoising
H. Fattahi, M. J. Valadan Zouj, M. R. Mobasheri, M. Dehghani, K.N. Toosi Univ. of Technology (Iran) Interferometry SAR is a powerful technique to produce topographic information and deformation maps using interferometric phase. The Interferometric SAR is restricted by many decorrelation factors such as temporal and baseline decorrelation and atmospheric effects that increase the noise of interferogram.However, the interferometric phase should be unwrapped in order to obtain the desired information. The number of residues in the interferogram will be increased due to the presence of noise. As a result, the phase unwrapping will be nearly impossible. Hence, the noisy interferogram should be filtered before phase unwrapping. Interferometric noise is a nonstationary process and need to reduce with time-frequency analythis with local processing ability. In this paper a filter based on Windowed Fourier Transform (WFT) or Short Time Fourier Transform has been developed for interferogram noise reduction. The presented approach does not need to divide the interferogram into sub-windows nor an initial input as a threshold. The proposed method has been validated using simulated noisy interferograms as well as real interferograms and the results have been compared with Goldstein and Werner algorithms applied in DORIS and Gamma softwares. The achieved results demonstrate the superiority of the proposed WFT-based filter in noise reduction compared to Goldstein and Werner filter. Furthermore, the proposed approach reduces the number of residues in the denoised interferogram and improves the phase unwrapping results.

6748-62, Poster Session


Practical lessons learned in deploying quick reaction remote sensing technologies
J. A. Orson, National Reconnaissance Office (USA); T. Hague, U.S. Air Force Academy (USA) Evolving threats encountered by coalition forces during contingency operations drive the need for new and improved airborne intelligence, surveillance, and reconnaissance capabilities. In many cases, these items are prototype, one-ofa-kind aircraft systems that have not completed a traditional course of testing. Operation Horned Owl is one such experimental quick reaction capability, in which experimental ground-penetrating radar and an off-the-shelf electro-optical and infrared sensor suite were rapidly integrated on a C-12 airframe and deployed to support Operation IRAQI FREEDOM. Utilizing sensor fusion and change detection, 21 Improvised Explosive Devices (IEDs) and 24 weapons caches totaling over 93,000 lbs of explosives were detected and disarmed, saving the lives of countless Iraqi citizens and coalition soldiers. This paper discusses the practical application of deploying unique sensor technology in a true operational environment - forward deployed combat arenas. Due to accelerated system employment, the aircraft received only basic stateside testing to characterize sensor performance. Remaining testing was conducted concurrent with combat operations to support ground forces. This nontraditional test approach and application of sensor technology highlights several areas that challenged successful system development. These areas along with lessons learned are provided to aid decision makers of future programs when balancing system maturation with meeting operational demand.

6748-60, Poster Session


Using empirical mode decomposition and wavelet analysis to study the optical turbulence record over 0.6 km horizontal paths
M. P. Chang, J. Rosado, Univ. de Puerto Rico Mayagez (USA) We report on work done towards the understanding of optical beam propagation characteristics in the lower atmosphere. We have recorded over 2 years worth of optical scintillometer data in both coastal and urban environments in the Caribbean sector, over 0.6-km pathlengths. The scintillation data represent the

6748-63, Poster Session


Study on the spectral transmission characteristics of MWIR through the atmosphere
J. Choi, T. Kim, Chung-Ang Univ. (South Korea) This paper is a part of developing a program that predicts surface emitted radiance from ground objects by considering solar irradiation and atmospheric convection. The radiance emitted from a surface can be calculated by using the

124

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6748: Image and Signal Processing for Remote Sensing


temperature and optical characteristics of the surface together with the spectral atmospheric transmittance. The thermal modeling is essential for identifying objects on the scenes obtained from the satellites. And the temperature distribution on the object is necessary to obtain their infrared images in contrast to the background. We considered the composite heat transfer modes including conduction, convection and spectral solar radiation for objects within a scene to calculate the surface temperature distribution. This software could be used to model the thermal energy to obtain the temperature distribution over the object by considering the direct and diffuse solar irradiances and by assuming the conduction within the object as onedimensional heat transfer into the depth. LOWTRAN7 and MODTRAN4 are used to model the spectral solar radiation including the direct and diffuse solar energy components. The object considered is assumed to be consisted of several different materials with different properties, such as conductivity, absorptivity, density, and specific heat etc. Results illustrate the radiance characteristics considering spectral atmospheric transmission of the midwave infrared (MWIR) radiation.

6748-66, Poster Session


The georeferencing errors of satellite data in remote sensing applications
E. Alecu, National Meteorological Administration (Romania) The change detection and analysis, as remote sensing applications, are based on the multi-temporal and multi-sensors approaches. The procedures of geometric correction and georeferencing of satellite data are extremely important in order to obtain reliable information. It is a known fact that georeferencing and geometric correction could lead to errors in the results of satellite image processing and the multi-temporal and multi-sensors applications are the most affected by these procedures. Errors due to georeferencing and geometric correction of satellite images are reflected on the correlation factor for the images analyzed. A simple mean to estimate these errors is to check this correlation by degrading the geometric position of one single band or of two different bands of the same sensor, using the same resolution or different resolutions. The method to obtain this artificial degradation was to apply a translation on three directions, rotation and deformation of one image comparing to the other. There are used different sizes of images, different covers and different complexity to evaluate the most influencing conditions. This approach allows quantification of the inappropriate image georeferencing and to know the size of distortion on the final results in case of comparison of images from different dates and different sensors.

6748-64, Poster Session


KOMPSAT-2 direct sensor modeling and geometric calibration/validation
D. C. Seo, Korea Aerospace Research Institute (South Korea) The horizontal geo-location accuracy of KOMPSAT-2, without GCPs (Ground Control Points) is 80 meters CE90 for monoscopic image of up to 26 degrees off-nadir angle, after processing including POD (Precise Orbit Determination), PAD(Precise Attitude Determination) and AOCS (Attitude and Orbit Control Subsystem) sensor calibration. In case of multiple stereo images, without GCPs, the vertical geometric accuracy is less than 22.4 meters LE 90 and the horizontal geometric accuracy is less than 25.4 meters. There are two types of sensor model for KOMPSAT-2, direct sensor model and Rational Function Model (RFM). In general, a sensor model relates object coordinates to image coordinates The major objective of this investigation is to check and verify the geometrical performance when initial KOMPSAT-2 images are employed and briefly introduce the sensor model of KOMPSAT-2.

6748-10, Session 3
Recent developments and future directions in hyperspectral data classification
A. J. Plaza, Univ. de Extremadura (Spain) No abstract available

6748-11, Session 3
Introducing training and parameter tuning for KOSP classification of hyperspectral images
L. Capobianco, L. Carli, F. Nencini, A. Garzelli, Univ. degli Studi di Siena (Italy) We introduce in this paper an optimization phase which jointly performs equivalent training and parameter tuning specifically designed for KOSP labeling of hyperspectral images. The KOSP algorithm is a sort of kernel-based generalized matched filter providing the class abundances of the sample under test. It does not present an explicit training phase, but requires the definition of a prototype spectral signature for each class (target) and a number of orthogonal signatures for the background (other classes or unlabeled data). Given a labeled training dataset, we propose to adopt a k-folder cross validation strategy maximizing the (average) classification accuracy which selects the p optimal spectral signature prototypes, being p the number of classes, and the gamma parameter of the RBF kernel. Moving the prototype signatures from the starting positions of the class means allows to modify the separating functions in the kernel feature space in a similar way to the SVM training phase. The proposed method is applied on the feature space obtained through canonical correlation analysis of the original HS space. Preliminary results on true HS data demonstrate that the classifier performances are improved with acceptable computing complexity of the training/ tuning phase which may be performed in a semi-automatic way.

6748-65, Poster Session


NUC algorithm of KOMPSAT-2 in on-board relative radiometric calibration
J. Song, Korea Aerospace Research Institute (South Korea) The KOMPSAT-2 satellite is a push-broom system with MSC (Multi Spectral Camera) which contains a panchromatic band and four multi-spectral bands covering the spectral range from 450nm to 900nm. The PAN band is composed of six CCD array with 2528 pixels. And the MS band has one CCD array with 3792 pixels. Raw imagery generated from a push-broom sensor contains vertical streaks caused by variability in detector response, variability in lens falloff, pixel area, output amplifiers and especially electrical gain and offset. Relative radiometric calibration is necessary to account for the detector-to-detector non-uniformity in this raw imagery. Nonuniformity correction (NUC) is that the process of performing a on-board relative correction of gain and offset for each pixel to improve data compressibility and to reduce banding and streaking from aggregation or re-sampling in the imagery. A relative gain and offset are calculated for each detector using scenes from uniform target area such as a large desert, forest, sea. In the NUC of KOMPSAT-2, The NUC table for each pixel are divided as HF NUC (high frequency NUC) and LF NUC (low frequency NUC) to apply to few restricted facts in the operating system of KOMPSAT-2. This work presents the algorithm and process of NUC table generation and shows the imagery to compare with and without calibration.

6748-12, Session 3
On the role of spectral resolution and classifier complexity in the analysis of hyperspectral images of forest areas
L. Bruzzone, Univ. degli Studi di Trento (Italy); M. Dalponte, Univ. degli Studi di Trento (Italy) and Ctr. di Ecologia Alpina (Italy); D. Gianelle, Ctr. di Ecologia Alpina (Italy) See the attached pdf file

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

125

Conference 6748: Image and Signal Processing for Remote Sensing

6748-13, Session 3
An assessment on the effectiveness of presegmentation of hyperspectral imagery to support various system trade analyses
S. T. Kacenjar, P. North, M. Hoffberg, Lockheed Martin Corp. (USA) Use of a Mean Class Propagation Model (MCPM) has been shown to be an effective approach in the expedient propagation of hyperspectral data scenes through the atmosphere1. In this approach, real scene data are spatially sub-divided into regions of common spectral properties. These sub-regions which we call classes are propagated in-toto through the atmosphere onto a collection platform such as an aircraft or satellite. This eliminates the need in many cases for detailed and timely pixelby-pixel calculations across the scene. This is done by computing a mean and spectral covariance for each class member of the scene and propagating these spectral measures through the atmosphere. However, this approach does introduce artifacts into the spatially segmented scene that must be addressed and quantified. Specifically, higher order spatial effects are not captured during the statistical assessments and subsequent propagation of each class member through the atmosphere. Such approximations, coined as second order spectral clipping (SOSC), may impact the fidelity of certain hyperspectral analyses such as operations involving target detection and identification. This paper will (1) examine class statistical measures that quantify the level of statistical deviation between real scene data and modeled representations of that data generated by the MCPM process. It will (2) examine how these statistical deviations are mitigated with the choice of pre-clustering methods, number of apriori classes, and cost functions used during the clustering operation. Lastly, it will (3) explore the impacts of the level of SOSC upon receiver operating characteristic curve (ROC) detection performance of lowcontrast targets that are be imbedded within the scene. To aid in this investigation, several hyperspectral data cubes will be examined using Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) data spanning a variety of scene background complexities.

Support Vector based methods, which rely on a nonparametric background model. Despite of the variety of works carried out on such algorithms, it is difficult to find in the literature a comparative analysis of these methodologies performed on a unique data set and therefore in the same operative conditions. In this work we present a comparative experimental analysis of different ADs carried out on a data set acquired ad-hoc by the prototypal hyperspectral sensor SIM-GA, manufactured by Galileo Avionica in Florence. This sensor is furnished with two cameras, operating in VNIR and SWIR, characterised by a very high spectral and spatial resolutions (about 2 nm and 1 mrad, respectively). In the measurement campaign, which was held in Pisa on 15 December 2005, the sensor was mounted on board a middle-size airplane. In the experiment, synthetic targets of various dimensions were embedded in a natural background together with calibration panels. The acquisition geometry was carefully characterised and the targets and background classes spectra were accurately measured by the FieldSpec handheld spectroradiometer. The performances attained on the investigated scenario are evaluated in terms of Receiver Operating Characteristics (ROC) or of typical merit figures derived from the ROCs, like the false alarm rate at a given level of the detection probability. The experimental results are finally compared and discussed also on the basis of the problems related to the practical implementation of the strategies.

6748-15, Session 4
Blind hyperspectral unmixing
J. M. P. Nascimento, J. M. Bioucas-Dias, Instituto Superior Tcnico (Portugal) Hyperspectral remote sensing exploits the electromagnetic scattering patterns of the different materials at specific wavelengths. Hyperspectral sensors have been developed to sample the scattered portion of the electromagnetic spectrum extending from the visible region through the near-infrared and mid-infrared in hundreds of narrow contiguous bands (10 nm). Spaceborn and airborne hyperspectral sensors acquire images of ground surface radiances. These radiances collected in spectral vectors, are mixtures of spectra from the substances present in the respective pixel coverage. Linear spectral unmixing considers that a mixed pixel is a linear combination of endmember signatures weighted by the correspondent abundance fractions. Under this model, the observations from a scene are in a simplex whose vertices correspond to the endmembers. Several approaches such as vertex component analysis (VCA), pixel purity index (PPI), and N-FINDR have exploited geometric features of hyperspectral mixtures to determine the smallest simplex containing the data. Those methods assume the presence in the data of at least one pure pixel of each endmember. This is a strong requisite that may not hold in some data sets. Independent Component Analysis (ICA) has recently been proposed as a tool to blindly unmix hyperspectral data. However, ICA applicability is compromised by the statistical dependence existing among abundances. This dependence results from the constant sum constraint imposed on the abundance fractions by the acquisition process. In ICA jargon, sources are not independent. Thus, the central assumption of ICA is not satisfied. This paper proposes a new method to blindly unmix hyperspectral data, termed dependent component analysis (DECA), where abundance fractions are modelled by a mixture of Dirichlet densities, thus automatically enforcing source nonnegativity and constant sum constraints. The mixing matrix is inferred by a generalized expectation-maximization (GEM) type algorithm. Compared with the geometric based approaches, the advantage of DECA is that there is no need to have pure pixels in the observations. The effectiveness of the proposed method is illustrated using simulated data based on U.S.G.S. laboratory spectra and real hyperspectral data collected by the AVIRIS sensor over Cuprite, Nevada.

6748-14, Session 4
Comparative analysis of hyperspectral anomaly detection strategies on a new high spatial and spectral resolution data set
S. Matteoli, F. Carnesecchi, M. Diani, G. Corsini, Univ. of Pisa (Italy); L. Chiarantini, Galileo Avionica SpA (Italy) In the last decade the use of hyperspectral sensors for target detection and recognition has been widely investigated in the literature. Hyperspectral sensors have very high spectral resolution and can potentially discriminate different objects on the basis of their unique spectral signature. Anomaly Detectors (AD) assume no knowledge about the spectrum of the potential target and simply explore the data cube to find those pixels whose spectrum significantly differs from the background. The success of anomaly detection arises from the fact that it does not require any form of atmospheric compensation and can be applied to the raw radiance image. Of course, a complex scene may contain a number of anomalies which do not necessarily represent a target of interest, especially in the case of very high spatial resolution data. Therefore, anomaly detection is often a first step in the analysis of the scene, providing the Regions of Interest (ROI), which may contain potential targets (cueing). In the literature, several ADs have been presented. The basic anomaly detection algorithm is the RX (Reed-Xiaoli). Its main limitation is the assumption of the Gaussian model to describe the local background. In fact, typically a scene is spatially non stationary and the Gaussian hypothesis is often violated even if the algorithm is applied locally. To avoid this problem additional techniques have been proposed. An approach is to reduce the spatial variation of the background by a preliminary step called background suppression. Another approach, the so-called Cluster-Based Anomaly Detection (CBAD), assumes a global Gaussian mixture model for the background. More recently, non-linear approaches have been proposed, such as the kernelRX, which employs a Gaussian model in the feature space, or

126

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6748: Image and Signal Processing for Remote Sensing

6748-16, Session 4
Hyperspectral clutter, phenomenology, and detection algorithms
M. Bernhardt, C. A. Steer, Waterfall Solutions Ltd. (United Kingdom) In this paper we report on improved hyperspectral detection algorithms by appealing to the natural processes that govern the competition between the different vegetation species that provide the clutter against which the detection problem is posed. Vegetation growth is restrained by inter- and intra-species competition for finite resources, such as light and water. Insight into these processes can be given by simple models using cellular automata. Two specific models have been chosen based on this physical motivation and developed further. Cellular automata are grid-based models that have simple interaction rules that act in the local neighbourhood of each member of the species. Despite the local and conceptually simple nature of the interaction rules, complex, long-range spatial correlations of the vegetation population can be generated. Abundance maps for each species are generated from these simulations. These maps measure the relative spectral contribution of each species to the simulated observed spectra. By invoking the linear mixing model, these models are then validated against real clutter distributions and used to develop improved hyperspectral detection processing. The validation method uses abundance maps together with a radiometrically accurate Monte-Carlo ray-tracing scene simulator (Cameosim) to produce hyperspectral cubes. Discussion is directed towards a comparison of the statistical properties of the simulated and observed background distributions. The paper also describes the statistical properties that have been used to characterise the phenomenology of real hyperspectral clutter. Examples include the nature of the tails of the clutter distribution, whether the distributions are elliptically contoured, and the structure of the spatial correlations present in vegetation backgrounds. A number of cellular automata models are then developed whose rules are motivated by toplevel biological competition principles. These cellular automata are used to generate abundance maps for various vegetation species. The statistical properties derived have a direct importance in automatic anomaly detection methods, as many of these methods assume an underlying background distribution. For instance, the RX anomaly detection assumes an elliptical contoured background distribution. Insight gained from the vegetation clutter models provides evidence of the local probability distribution around each pixel. Having gained some understanding of this link between hyperspectral vegetation clutter and the natural processes that give rise to the spatial distributions of vegetation, candidate hyperspectral detection algorithms were developed to attempt to exploit this phenomenology. The performance of these algorithms was compared on real hyperspectral image data and benchmarked against existing techniques, and example results are reported in this paper.

sensor-specific fast radiative transfer formalism for UV-visible to LWIR wavelengths that is derived from MODTRANs correlated-k parameters. Scattering parameters, blackbody and solar functions are cast as a few sensor-specific and bandpassspecific k-dependent source terms for radiance computations. Preliminary transmittance results are within 2% of MODTRAN with a two-orders-of-magnitude computational savings. Preliminary radiance computations in the visible spectrum are within a few percent of MODTRAN results, but with orders of magnitude speed up over comparable MODTRAN runs. This new RT capability has potential applications for remote sensing applications such as hyperspectral scene simulation and lookup table generation for atmospheric compensation analysis as well as target acquisition algorithms for near earth scenarios

6748-18, Session 4
Non-negative factorization of non-negative matrices
J. H. Gruninger, Spectral Sciences, Inc. (USA) A new non-negative factorization method has been developed. The method is based on the concept of non-negative rank (NNR). Bounds for the NNR of a non-negative matrix are determined relative to the rank of the matrix, with the lower bound being equal to the rank. The method requires that the data matrix be non-negative. Unlike other non-negative factorization methods, the approach does not assume or require that the factors be linearly independent and no assumption of statistical independence is required. There is no initial guess to a solution, no minimization of a cost function and hence no local minima that leading to multiple non-unique solutions. The factorization is unique. The method uses the rank or pseudorank of the non-negative matrix based on its singular value decomposition (SVD). An analytical approach generates both left and right non-negative factors that exactly reproduce the SVD expansion. The rank or pseudo-rank of the matrix needs to be pre-determined, but the number of non-negative factors, the NNR, does not have to be known in advance. Once determined for an estimate of the pseudo-rank, the nonnegative factors do not have to be re-determined for a new estimate of the pseudo-rank. The non-negative factors are not necessarily data vectors, but the set of factors can be shrunk to identify extreme points in the data. The concept of NNR is illustrated with examples and the method is applied to synthetic and measured spectral imaging data. The method finds both end-member spectra and their abundances, as the left and right non-negative factor matrices. Examples include instances where the NNR exceeds the rank. This situation occurs in hyperspectral data whenever the number of unique materials, exceeds the number of linearly independent bands. Standard un-mixing and factorization methods that are based on least squares or non-negative least squares techniques on not applicable to this situation as the implicit inversion is singular. The method offers particular advantage in multi-spectral data sets where instances of the number of materials present in a scene exceeding the number of linearly independent bands are ubiquitous. The approach utilizes the work of Cohen and Rothblum (1993) who introduce the concept of NNR. The bounds on the NNR quoted by Cohen and Rothblum are that the NNR is greater than or equal to the rank and is lest than or equal to the smaller of the number of rows or columns of the data matrix. In the present work the upper bound is tightened to the NNR being less than or equal to twice the rank minus two. In the present work, the dual nature of the factorization problem, is column basis, row basis is utilized to determine the analytical solution. References J. E. Cohen and U. G. Rothblum Non-negative Ranks, Decompositions, and Factorization of Non-negative Matrices Linear algebra Appl. 190, 149-168, (1993).

6748-17, Session 4
Correlated-k based fast, accurate bandpass radiance and transmittance calculations for hyperspectral and multispectral scenes
P. K. Acharya, R. Panfili, A. Berk, S. M. Adler-Golden, Spectral Sciences, Inc. (USA); A. Wetmore, R. C. Shirkey, Army Research Lab. (USA) The ability to rapidly calculate at-sensor radiance over a large number of lines of sight (LOSs) is critical for scene simulation and look-up table generation, both of which are increasingly used for sensor design, performance evaluation and data analysis. We have recently demonstrated a new radiation transport (RT) capability that combines an efficient multipleLOS multiple scattering (MS) algorithm with a broad-bandpass correlated-k methodology called kURT-MS, where kURT stands for correlated-k based Ultra-fast Radiative Transfer. The multiple-LOS capability is based on DISORT and exploits the existing MODTRAN-DISORT interface. kURT-MS is a new

6748-19, Session 5
Automatic land and sea surface temperature estimation from remote sensing data
S. B. Serpico, G. Moser, Univ. degli Studi di Genova (Italy) Satellite remote sensing systems provide a repetitive and consistent view of the Earths surface and measurements taken by on-board sensors are currently the only way to derive Earth

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

127

Conference 6748: Image and Signal Processing for Remote Sensing


surface parameters from surface-leaving radiation on a regional and global scale. Land surface temperature (LST) and sea surface temperature (SST) are important quantities for many environmental models, such as energy and mass exchange between atmosphere and surface (Caparrini et al., 2003), numerical weather prediction, climatic variability, and global ocean circulation (Dash et al., 2002). In order to estimate LST and SST from satellite data many algorithms have been devised, most of which require a-priori information about the surface and the atmosphere (Dash et al., 2002). However, the high spatial variability of surface and atmospheric parameters causes these traditional methods to produce significant estimation errors, thus making their application on a global scale critical. A recently proposed approach involves the use of support vector machines (SVMs) (Zortea et al., 2006). The SVM generates an approximation of the relation between satellite data and matching in-situ measurements (adopted for training purposes). This supervised learning strategy can be theoretically proven to exhibit very good generalization and robustness properties (Vapnik, 1995) and has resulted in accurate LST estimates (Zortea et al., 2006). However, such a strategy requires the user to set several internal parameters. In this paper, the SVM approach to LST estimation is adopted and extended to the SST case, and a method is proposed for automatically setting the parameters required by the SVM to values that minimize estimation errors. The proposed method consists of minimizing the spanbound functional, which represents an upper bound on the leave-one-out estimation error and which can be computed using only the training set (i.e., without involving test samples or time-expensive crossvalidation strategies) (Chang et al., 2005). A drawback of this functional may lie in the fact that it is a nondifferentiable function of the unknown parameters; this prevents the use of gradientbased or Hessian-based numerical minimization algorithms. The proposed method overcomes this drawback by adopting the Powells minimization method, which does not require derivative computations, i.e., it is applicable to non-differentiable functions, as well (Press et al., 2002). The method essentially aims at emulating (although without using derivatives) the behavior of the Hessian-based conjugate gradient algorithm by iteratively generating and updating a set of directions which approximates a set of conjugate directions (Press et al., 2002). The proposed method is validated by experiments on AVHRR data, endowed with in-situ LST and SST measurements. The experimental results are compared to those obtained with a typical strategy for parameter setting, i.e., a grid search for the parameter configuration yielding the highest test-set accuracy. Experimental results generated by the proposed method turns out to be very similar to those obtained by the grid search, with negligible differences in the mean absolute errors and in the root mean square errors. SVM-generated maps of LST and SST estimates created using parameters set by the proposed method and by grid searches visually confirm this similarity. On the other hand, the computational time of Powells method turns out to be considerably shorter than the execution time of grid searches, which also exponentially increases with the number of parameters being evaluated and thus may become large. References: Caparrini F., Castelli F. and Entekhabi D.: 2003, Mapping of land-atmosphere heat fluxes and surface parameters with remote sensing data, Boundary-Layer Meteorology, 107:605633. Chang M.-W., and Lin C.-J.: 2005, Leave-one-out bounds for support vector regression model selection, Neural Comp., 17:1188-1222. Dash P., Gottsche F.-M., Olesen F.-S., and Fischer H.: 2002, Land surface temperature and emissivity estimation from passive sensor data: theory and practice - current trends, Int. J. Remote Sens., 23:2563-2594. Zortea M., De Martino M., Moser G., Serpico S. B.: 2006, Land surface temperature estimation from infrared satellite data using support vector machines, Proc. of IGARSS-2006, Denver, USA, 31 July 31 - 4 August 2006, pp. 2109-2112. Press W. H., Teukolsky S. A., Wetterling W. T., and Flannery B. P.: 2002, Numerical recipes in C, Cambridge University Press. Vapnik V.: 1995, The nature of statistical learning theory, York: Springer-Verlag.

6748-20, Session 5
Moisture effects removal on soil spectra
F. Tavin, A. Minghelli-Roman, Univ. de Bourgogne (France); S. Mathieu, Alcatel Alenia Space (France); W. Liu, Institute of Urban Meteorology (China); F. Baret, Institut National de la Recherche Agronomique (France); P. Gouton, Univ. de Bourgogne (France) Soil classification from multi/hyperspectral images is a wide field of investigation. A major difficulty of these classifications is linked to the influence of the soil moisture on the soil reflectance spectrum. Indeed, the variations induced by this influence can be higher than variations from on soil spectrum to another and that can leads to misclassifications. That is the reason why many articles propose moisture models to understand this phenomenon and to try to cancel its effects [15]. Our approach considers that the spectrum of a given soil with a given moisture level is the product of the value of a moisture function and the value of the spectrum of the same dry soil. The moisture function takes two parameters that are the wavelength and the moisture level and does not depend on the soil itself. As we aim to find the corresponding dry soil from any moist soil, we need to determinate from the spectrum itself the moisture level and to invert the moisture function. As methods estimating the moisture level from the soil spectrum already exist [6], we focused our research on the moisture function. We studied for each wavelength the relation between the spectra of the moist soil and the same dry soil. This function has been estimated from 92 different soils measured with 5 different moisture levels. This spectral library has been acquired in laboratory conditions. The best function fitting the real values is polynomial (2nd order) with a correlation coefficient R(c) around 0.9. Once all spectra of the library have been corrected from the moisture effect, we compared the corrected spectra with the corresponding dry soils spectra. The RMSE is then around 0.05 between corrected spectra and dry soil spectra when it was around 0.3 without moisture correction. References: [1] Etienne Muller, Henri Dcamps, Modeling soil moisture reflectance, Remote Sensing of Environment, vol. 76, n2, May 2001, pp. 173-180. [2] David B. Lobell, Gregory P. Asner, Moisture Effects on Soil Reflectance, Soil Science Society of America Journal, vol. 66, 2002, pp. 722-727. [3] Liu W., F. Baret, X. Gu, Q. Tong, L. Zheng, B. Zhang, Relating soil surface moisture to reflectance, Remote Sensing of Environment, vol. 81, 2002, pp.238-246. [4] Liu W., F. Baret, X. Gu, B. Zhang, Q. Tong, L. Zheng, Evaluation of methods for soil surface moisture estimation from reflectance data, International Journal of Remote Sensing, vol. 24, n10, 2002, pp.2069-2083. [5] A. L. Kaleita., L. F. Tian, M. C. Hirschi, Relationship Between Soil Moisture Content and Soil Surface Reflectance, Transactions of the American Society of Agricultural Engineers, vol. 48, n5, 2005, pp.1979-1986. [6] Michael L. Whiting., Lin Li, susan L. Ustin, Predicting water content using Gaussian model on soil spectra, Remote Sensing of Environment, vol. 89, 2004, pp.535-552.

6748-22, Session 5
Image-based method for estimating signalnoise to ratio (SNR) in remotely-sensed data
A. Asmat, Univ. of Southampton (United Kingdom); G. M. Foody, The Univ. of Nottingham (United Kingdom); P. Atkinson, Univ. of Southampton (United Kingdom) Remotely sensed imagery has been used for developing and validating various studies regarding land cover dynamics. Accuracy in extracting land surface information features from satellite remotely sensed data relies heavily on the spectral data quality of the imagery. Unfortunately, in many instances the quality of the image is influenced by atmospheric conditions during data acquisition. Noise in remote sensing imagery degrades the interpretability of the data. In addition, noise added to the data in an optical remote sensing system is typically

128

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6748: Image and Signal Processing for Remote Sensing


independent of the data, and is generally additive in nature. Thus, to utilize the imagery data, the random variation or noise associated with the sensors signal should be measured. Estimation of noise contained within a remote sensing image is essential in order to counter the effects of noise contamination. This is important, as the noise determines the accuracy with which absorption features can be distinguished in the waveband and object can be detected on the ground. Several methods have been developed to estimate the signalto-noise ratio of remotely sensed imagery. In this study, demonstrate the use of spatial statistic method to estimate additive signal noise on remotely sensed imagery. The importance to carry out this study by noting classification accuracy on the basis that an increase in classification accuracy reflects in improved image quality. advanced DPCM method (MAD-constrained) is far preferable to an RMSE-constrained method, like spectral decorrelation followed by JPEG 2000. Hyperspectral data are usually compressed in radiance format. Specific analyses, such as detection of materials and spectral unmixing, are better accomplished from data in reflectance format, i.e., after atmospheric absorption effects have been removed from radiance data. Preliminary results of simulations performed on reflectance data obtained from compressed radiance data show that, for a given compression ratio, by following the virtually lossless approach both BAO-SAM and SAM are minimized and discrimination of spectrally similar materials, e.g., clays, is significantly expedited. [1] N. Keshava, Distance metrics and band selection in hyperspectral processing with applications to material identification and spectral libraries, IEEE Trans. Geosci. Remote Sensing, Vol. 42, No. 7, pp. 1552-1565, 2004. [2] B. Aiazzi, L. Alparone, S. Baronti, Near-lossless compression of 3-D optical data, IEEE Trans. Geosci. Remote Sensing, Vol. 39, No. 11, pp. 2547-2557, 2001. [3] C. Lastri, B. Aiazzi, L. Alparone, S. Baronti, Virtually lossless compression of astrophysical images, EURASIP J. Applied Signal Processing, Vol. 2005, No.15, pp. 2521-2535.

6748-23, Session 6
Interband distortion allocation in loss compression of hyperpectral imagery
S. Baronti, C. Lastri, B. Aiazzi, Istituto di Fisica Applicata Nello Carrara (Italy); L. Alparone, Univ. degli Studi di Firenze (Italy) Technological advances in imaging spectrometry have recently led to acquisition of data that exhibit extremely high spectral resolution. The increment in number of spectral bands has motivated the extension of vector signal/image processing techniques to hyperspectral data and the development of distortion models aimed at quantifying the performance of analysis algorithms that are usually employed. Typical distance metrics compare two pixel spectra and return a scalar value. The Euclidean minimum distance (EMD) and the spectral angle mapper (SAM) are well-known examples. In a recent paper [1], the characteristics of such metrics are examined and a band add-on (BAO) technique is derived. Spectral bands are iteratively selected in order to increase the angular separation between two pixel spectra by exploiting a mathematical decomposition of SAM. As a consequence, only a subset of the original hyperspectral bands contributes to the new distance metrics, referred to as BAO-SAM, whose operational definition guarantees that it is monotonic as the number of bands increases. BAO-SAM is potentially more useful than SAM for classification, since SAM between two pixel spectra exhibits lower values as the spectral resolution increases. In order to meet the quality issues related to the transmission and archiving of hyperspectral data, differential pulse code modulation (DPCM) is usually employed for lossless/nearlossless compression, i.e., the decompressed data have a userdefined maximum absolute error, being zero in the lossless case. DPCM basically consists of a prediction followed by entropy coding of the quantized differences between original and predicted values. A unit quantization step size allows reversible compression as a limit case. Several variants exist in a DPCM prediction scheme, the most sophisticated being adaptive [2]. Such distortion metrics as root mean square error (RMSE), maximum absolute distortion (MAD), average and maximum SAM, are usually adopted to assess lossy compression algorithms. However, it is desirable that a hyperspectral data compression method of general validity should not be specialized to a specific application but optimized in terms of a distortion metrics, which exhibits favorable characteristics when such applications as detection/classification/unmixing are carried out from compressed data. In this work, we investigate the problem of distortion allocation along wavelength, either MAD for near-lossless methods (DPCM) or RMSE for lossy methods (spectral decorrelation + JPEG 2000) of AVIRIS data. Two strategies are compared: given a target average bit rate, distortion (either MAD or RMSE) may be set to be constant varying with wavelength. Otherwise it may be allocated proportionally to the noise level on each band, according to the virtually lossless protocol [3]. Thus, a different quantization step size depending on the estimated standard deviation of the noise, is used to quantize either prediction residuals (DPCM) or wavelet coefficients (JPEG 2000) of each spectral band, thereby determining band-varying MAD/RMSE values. Comparisons with the uncompressed originals show that the average SAM is minimized by constant distortion allocation. Conversely, the average BAO-SAM is minimized by the noise-adjusted variable spectral distortion allocation according to the virtually lossless protocol. Furthermore an

6748-24, Session 6
Interactive decoding for the CCSDS recommendation for image data compression
F. Garcia-Vilchez, J. Serra-Sagrist, A. Zabala, X. Pons, Univ. Autnoma de Barcelona (Spain) In 2005, the Consultative Committee for Space Data Systems (CCSDS) approved a new Recommendation (CCSDS 122.0-B1) for Image Data Compression (IDC). This Recommended Standard defines an image-data compression algorithm applicable to digital data from payload instruments and specifies means to control compression rate and how these compressed data shall be inserted into source packets for retrieval and decoding. Our research group has followed the design and the approval of the Recommendation since 2003. Since then we have developed an implementation of the Recommended Standard and tested it in several scenarios. In addition, we extended the algorithm to any number of wavelet transform levels, which is limited to three in the Recommendation. The following step in our research is to support multi-channel images. We are currently working to add several progression orders to the Recommended stream (analogous to the ones defined in JPEG2000). In this way, we can also encode once and decode in many ways. Besides, we would be able to transmit images using the JPIP protocol. In order to achieve these new features, some additional headers should be embedded in the encoded stream. Such headers identify what is represented by each part of the stream. With this setting, the Recommendation will be improved by adding quality scalability, resolution level scalability, and component scalability, without the need to decode the whole encoded stream. The additional embedded headers place the encoded images out of the Recommendation. However, we obtain a considerable set of new features. In addition, if we receive an image encoded following the Recommendation, we can add the headers without inverting all the decoding process. This is important because this Recommendation is addressed to payload instruments, which usually have computational and memory limitations. Once the image is received in the ground, it can be partially decoded and re-formatted by adding embedded headers, which allows the interactive decoding and also allows interactive transmission. Apart from the design and the implementation of an interactive decoder, we are also working in a graphical interface that allows the end user to easily manage the features that we are adding to the Recommendation.

6748-26, Session 7
Robust classification of hyperspectral images
A. S. Solberg, Univ. I Oslo (Norway) No abstract available

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

129

Conference 6748: Image and Signal Processing for Remote Sensing

6748-27, Session 7
Investigation of an ensemble framework for classification of hyperspectral remote sensing data with nearly equal spectral response classes
M. Zortea, S. B. Serpico, Univ. of Genoa (Italy) The current availability of high-dimensional image data remotely sensed by hyperspectral sensors, carried onboard satellite and aerial platforms, provides a new class of Earth observation data for improved surface characterization. Successful examples of such typology of instruments are the Hyperion and the Airborne Visible/Infrared Imaging Spectrometer (AVIRIS). Typically, surface properties is acquired on more than 200 channels, opening new possibilities in remote sensing digital image classification, particularly when dealing with classes that are spectrally very similar. Hyperspectral imaging has wide ranging applications, among others, a detailed classification of land resources through hyperspectral imagery is expected to allow better environmental management, better predictions of crop yield, more accurate remote mineral exploration, better forest mapping, geological studies, etc. The main challenge on classifying hyperspectral data with statistical approaches is dealing with the problem of curse of dimensionality. Typically, the amount and quality of ground truth labelled data (training samples) is limited to the accurate estimation of a large number of classifiers parameters. The resulting classifiers are often unstable and have poor generalization [Ham et al., 2005]. Different approaches in solving the small sample sizes and high dimensionality problem, which are inherently coupled issues in hyperspectral data analysis, have been proposed in literature. Approaches based on regularization of the covariance matrix, feature reduction (selection or extraction), and the use of semilabeled data are example of methods expected to mitigate loss of classification accuracy as the data dimensionality increases [Landgrebe, 2003]. Recent approaches in the context of hyperspectral image classification include the use of kernel-based methods. In particular, Support Vector Machines [Vapnik, 1995] seems a promising technique due its well-suited properties for the problem of hyperspectral image classification [Melgani and Buzzone, 2004]. SVMs can efficiently deal with large input spaces, are quite robust to noisy samples, and produce decision boundary expressed as a function of a subset of training samples (sparse solution) [Camps-Valls and Bruzzone, 2005]. A mostly common to such previous approaches for classifying hyperspectral remote sensing data is that the classification map is generated by a single classifier (model). The theory and practice of classifiers ensembles also provide tools for accurate classification of remote sensing data. A well know example of ensemble method is the Random Forests. It was proposed as a machine learning technique in [Breiman, 2001] and has been widely and rapidly improved for use in pattern recognition. In remote sensing data classification, recent studies include the use of the Random Forests approach for classification of multisource remote sensing and geographic data [Gislason et al., 2006]. In a different perspective, the Random Forests approach using random feature based subset selection methods has been investigated for classification of hyperspectral data [Ham et al., 2005]. Results suggest that the Random Forest framework help improved generalization of the classifier in analysis of hyperspectral data. The effectiveness of ensemble methods critically depends on accuracy, diversity and learning characteristics of base learners [Valentini, 2004]. As originally proposed, the Random Forests approach uses the CART decision tree classifier, which may provide poor classification when classes are characterized by nearly equal spectral response in remote sensing hyperspectral data. We propose an experimental analysis aimed at understanding and assessing the potentialities of an ensemble of SVM classifiers as a supervised tool for the accurate classification of such kind of data. We investigate the use of bagging [Breiman, 1996] and random selection of input features for promoting diversity in the ensemble. When using SVM, a few internal SVM model parameters should be estimated [Vapnik, 1995]. An optimization through a classic

grid search procedure for parameters selection could render the SVM processing time very long, depending on grid resolution chosen. In order to reduce the processing time to set the SVM parameters, but without an expected significant loss of the accuracy, a K-fold cross-validation strategy is used for selecting, among the limited set of resulting candidates, the SVM parameters that maximizes the performance in terms of the classification accuracy of each single classifier of the ensemble. Additionally, it is used to estimate the weight (contribute) of each single classifier in the voting process. The final map is generated by means of a weighed combination of n classification maps (weighted majority voting, [Kuncheva, 2004]). The weighted voting process is expected to allow a feedback from the quality of the resulting random choices of input features. It means, we attempt to give the more competent classifiers more power in making the final decision. The real data examined in the paper is the well-known AVIRIS Indian Pines image [Landgrebe, 2003], an agricultural scene containing some difficult to separate classes. The two and multiclass classification problems are addressed in detail. Accuracy results are calculated for a range of (small) training set sizes, using feature reduction in the original feature set available. The ensemble size is another parameter considered. Additionally, results are presented when a previous feature reduction step is applied to the dataset. As benchmark, we compare the ensemble accuracies with those obtained using well tuned single SVMs that uses linear and non linear kernels. Encouraging results suggest the proposed method as promising, in addition of being computationally easy to implement. References: Breiman, L., Random forests, Machine Learning, vol. 45, no. 1, 2001. Breiman. L., Bagging predictors, Machine Learning, vol. 24, no. 2, 1996. Camps-Valls, G. Bruzzone, L. Kernel-based methods for hyperspectral image classification, IEEE Trans. Geosci. Remote Sensing, vol. 43, no. 6, 2005. Gislason, P. O., Benediktsson, J. A., Sveinsson, J. R., Random Forests for land cover classification, Pattern Recognition Letters 27, 2006, pp. 294-300. Ham, J., Chen, Y., Crawford, M. M., Ghosh, J., Investigation of the Random Forest Framework for Classification of Hyperspectral Data, IEEE Trans. Geosci. Remote Sensing, vol. 43, no. 3, 2005. Kuncheva, L., Combining Pattern Classifiers: Methods and Algorithms. Wiley-Interscience, 2004. Landgrebe. D. A., Signal Theory Methods in Multispectral Remote Sensing. Wiley-InterScience, 2003. Melgani, F., Bruzzone, L., Classification of hyperspectral remote sensing images with support vector machines, IEEE Trans. Geosci. Remote Sensing, vol. 42, no. 8, 2004. Valentini G., Dietterich T. G., Bias-Variance Analysis of Support Vector Machines for the Development of SVM-Based Ensemble Methods, Journal of Machine Learning Research, vol. 5, 2004. Vapnik, V., The Nature of Statistical Learning Theory, New York: Springer-Verlag, 1995.

6748-28, Session 7
Assessment of quality parameters for a new generation hyperspectral imager
B. Aiazzi, Istituto di Fisica Applicata Nello Carrara (Italy); L. Alparone, Univ. degli Studi di Firenze (Italy); A. Barducci, S. Baronti, D. Guzzi, I. Pippi, P. Marcoionni, M. Selva, Istituto di Fisica Applicata Nello Carrara (Italy) This work focuses on measuring the quality of the data acquired by a new hyperspectral sensor characterized by a very high spectral resolution. By resorting to Shannons information theory, the concept of quality is related to the information conveyed to a user by multi-band data; that is to the mutual information between the unknown noise-free digitized signal and the corresponding noise-affected observed digital samples. Lossless data compression is exploited to measure the information content of the data. The bit rate achieved by a reversible compression process takes into account the contribution of the observation noise, i.e., information

130

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6748: Image and Signal Processing for Remote Sensing


regarded as statistical uncertainty, whose relevance is null to a user, and the intrinsic information of hypothetically noise-free samples. Once the parametric model of the noise, has been preliminarily estimated, the mutual information between noisefree signal and recorded noisy signal is calculated as difference between the entropy of the noisy signal and entropy derived from the parametric model of the noise. The noise is assumed to be possibly non-Gaussian and both spatially and spectrally autocorrelated. However, it is desirable also to know the amount of information that the digitized samples would convey if they were ideally recorded without experimental noise. A source entropy model is defined, the inversion of which yields an estimate of the information content of the noise-free source from the code rate and the noise model. Thus, it is possible to establish the extent to which an increment in the radiometric resolution, or equivalently in the signal-to-noise ratio (SNR), obtained thanks to technological improvements of the imaging sensor, can increase the amount of information that will be available to users applications. The above objective measurement of quality is increasingly matching the subjective concept of quality, i.e., the ability of achieving a desired objective, as the number of spectral bands increases. Practically, (mutual) information, or equivalently SNR, is the sole quality index for hyperspectral imagery, generally used for detection/identification of materials and spectral anomalies, rather than for conventional multispectral classification tasks. Noise variance and correlation coefficients, along and across track and in the spectral direction, are estimated. Entropy is measured on the observed data and inferred for the ideally noise-free source taking into account the contribution of additive zero-mean noise assumed to be signal independent and spatially stationary. SNR and mutual information are also derived. As a first case study, both L1A (raw) and L1B (corrected) data are analyzed in order to verify any effects resulting after the radiometric correction process. A certain influence on noise variance and on spatial correlation might be expected because of destriping. In a second test, different areas of the same image will be processed separately in order to assess the noise model on different landscapes. robust to the low snr in the atmospheric absorption images with low transmission. The correction, results in a hyperspectral cube with all the bands spatially aligned. The algorithm has been applied to the following hyperspectral imagers : AISA Hawk, AVIRIS ,CASI and HYPERION.

6748-31, Session 8
Genetic algorithms in estimating optimal neural network topologies for the classification of remotely sensed images
D. N. Stathakis, Joint Research Ctr. (Italy) The purpose of this paper is to summarize the most commonly used heuristics for the determination of optimal neural network topology in classifying remotely sensed imagery, and compare their performance to a proposed method based on searching by a genetic algorithm. Two decades following the introduction of neural networks in the classification of remotely sensed data tasks [1] one of their major drawbacks is the inexistence of an exact method to determine he hidden structure of the network. Several approaches have been devised to cover for this symptom. Some of them use the theorem of Kolmogorov [7] as the point of departure. The first, and most primitive, one is solely based on trial and error. The second, and most common, one is the deployment of specific heurists [2] to discover initial topologies for subsequent tuning by trial and error. Third is searching by constructive and pruning algorithms such as the widely adopted Optimal Brain Surgeon [6], [10]. Forth, is exhaustive search which is typically not an option for any real work application given the enormous time required to evaluate even a fraction of the solution space. In this paper, a different alternative is presented based on global search methods that have surprisingly received little attention until today in remote sensing. The method is based on the synergy of neural networks and genetic algorithms [3], [4], [5], [8]. The number of nodes of the first and second hidden layers are coded in the chromosome. A 10 bit chromosome is used holding 6 bits for the first and 4 bits for the second hidden layer. The topologies searched are 10:[20-83]:[0-15]:5. This range covers for the most known heuristics. In addition there is evidence that the results will show if the genetic algorithm is pushing for a solution outside those bounds. Several parameters in the genetic algorithm side have to be set but this it is somewhat easier compared to neural network parameter setting due to the existence of relevant theoretical evidence [3], [4]. The fitness function used is either solely classification error (the objective is to minimize it) or novel fitness functions promoting topologies that are not merely accuracy but also compact. The data set consists of a LANDSAT EMT+ image and SRTM [9] topographic ancillary data. CORINE land cover [11] is used as the reference source. Three sets (training, testing and validation) are formed following stratified random sampling to train and evaluate the network. The results show that the genetic algorithm is able to discover superior topologies compared to all known heuristics within ten generations (fig. 1). These topologies are discovered by evaluating only a fraction of the solution space. The heuristics are ranked according to performance for this data set (table 1) and their performance discussed. Classification results are also presented. Global search methods such as genetic algorithms provide a very efficient alternative that can permanently solve the question of how many hidden layers and how many hidden nodes. Many of the heuristics currently used yield suboptimal skeletons. The proposed method is systematic and relatively automated method with a solid theoretical framework.

6748-29, Session 7
Efficient regularized LDA for hyperspectral image classification
T. V. Bandos, Univ. de Valncia (Spain); L. Bruzzone, Univ. degli Studi di Trento (Italy); G. Camps-Valls, Univ. de Valncia (Spain) In this paper, we focus on different forms of regularization of the LDA. We conduct an exhaustive comparison of many LDAbased methods in the special case of classification of ill-posed remote sensing data. Second, we introduce an efficient version of the regularized LDA. Our third proposal includes, as efficient regularizer, the contextual information in the classifier through the novel formulation of composite covariances. Experimental results show the suitability of our proposals.

6748-30, Session 7
Co-registration of hyperspectral bands
Z. Figov, K. Wolowelsky, N. Goldberg, RAFAEL Armament Development Authority Ltd. (Israel) Hyperspectral imagery consists of a great number of spectral images. Although these spectral images are obtained simultaneously, due to a number of optical affects the hyperspectral bands are not co-registered. This is particularly prominent in images of regions near the absorption bands of atmospheric gases. This can be easily demonstrated when seeing a movie of hyperspectral bands of an object. We have developed a method using subpixel image correlation that enables us to identify the spatial misregistration between spectral bands and correct it. We use one band as a reference band and match all other spectral images to this band. We use the average row and column disparity to correct the spatial mis-registration. Other than the spatial correction that we obtain we also obtain a mapping of the atmospheric gas absorption features. This is done by plotting the average disparities as a function of wavelength. Local peaks in this plot are clear evidence of absorption features of the atmospheric gases. Our method is

6748-32, Session 8
Automatic snow extent extraction in alpine environments: short and medium term 20002006 analysis
G. Lisini, P. Gamba, E. Merlin, F. Riva, Univ. degli Studi di Pavia (Italy) Spatial snow characterization requires spatially distributed information. This can be provided by remote sensing data. Indeed, there are already a number of algorithms devoted to

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

131

Conference 6748: Image and Signal Processing for Remote Sensing


this task for data coming from various and different sensors in the visible and infrared regions [1-5]. Extracted information is subsequently used as one of the inputs to hydrological, particularly water discharge, models. For instance, they can be used as model validation data, or for forecasting long-term climate trends to be addressed in next decade water management policies at regional or national level. This is the reason why Lombardy Region, University of Milan Bicocca, Fondazione Lombardia per lAmbiente and ARPA Lombardia are currently funding a project, named Regional Impact of Climatic Change in Lombardy Water Resources: Modelling and Applications (RICLIC-WARM). Among the available spaceborne sensors, NOAA - AVHRR is usually considered as the most suitable for producing the required model parameters. In this paper a fully automatic operational algorithm to derive the snow covered area from optical data will be presented. Snow-cover maps are processed and provided to the user within a few minutes from the original NOAA - AVHRR data. Processing consists of calibration, geometric correction and classification, including automatic snow-cloud discrimination, which is one of the most important steps in the processing chain. To get maximum information and in order to reduce the cloud cover, the available images from NOAA site were analyzed synergistically. For operational applications high geometric accuracy is necessary. This is achieved by integrating an automatic iterative procedure using satellite orbit information and parametric geocoding. Further innovations of the algorithm are the threshold classification of areas with expected melting/temperated snow. For operational application, the processing chain includes a post-classification, and quality assessment of the result. The results of the analysis of two time series of images dating from the 2000 to 2006 will be discussed in the full paper. Both of them refer to the whole Northern Italy Alpine area. The first series compares snow extents extracted from nearly 100 images collected from 2000 to 2006 and enforces a medium term analysis. The second one focuses the analysis in the year 2002 using about 70 images for a more temporally detailed result. Beyond the analysis of the snow extent temporal behaviour, other interesting results could be extracted from these series, such as the statistical snow-coverage difference between Northern and Southern slopes during the whole winter season in multiple years. REFERENCES [1] Hall D.K., Riggs G.A., Salomonson V.V., Di Girolamo N.E., Bayr K.J. MODIS snow-cover products, Remote Sens. Environ., vol. 83, pp. 181-194, 2002. [2] Maurer E.P., J.D. Rhoads, R.O. Dubayah, D.P. Lettenmaier Evaluation of the snow-covered area data product from MODIS, Hydrological. Processes, vol. 17, pp. 59-71, 2003 [3] Rango A., Martinec J., Foster J, Marks D. Resolution in operational remote sensing of snow cover. IAHS Publ., vol. 145, pp. 371-381, 1983. [4] Schmugge T. J., Kustas W. P., Ritchie J. C., Jackson T. J., Rango A. Remote sensing in hydrology, Advances in Water Resources (Elsevier), vol. 25, pp. 1367-1385, 2002 [5] Swamy A. N., Brivio P.A. Modelling runoff using optical satellite remote sensing data in high mountainous alpine catchment of Italy, Hydrological Processes, vol. 11, pp. 14751491, 1997. evaluation of evidence-mass functions which are used as the frame of discernment events. This solution is based on a refinement of the frame of discernment based on the determination of all focal elements (singleton or composite hypothesis of non null masses). Thus, the combination of information source masses would involve only the focal elements masses. In the approach proposed here, the notions of fuzzy logic and possibility theory have been used for the calculation of masses and combinations between classes as an intermediary phase in arriving at belief functions. The result of the application of our fusion approach on information sources collection for a study area in the Ain-Drahim region (Tunisia) revealed a significant improvement in optimizing the calculation of mass evidence functions and thus achieving a satisfactory classification. These sources involved a highresolution Quickbird image, a medium-resolution SPOT image, a digital terrain model of the study area as well as a digitized forest inventory.

6748-34, Session 8
Land cover characterization through parametric modeling of intra-annual reflectance time series: a comparative study with MERIS data
H. M. Carro, Instituto Geogrfico Portugs (Portugal) and Institute of Statistics and Information Management (Portugal); P. J. Sequeira Goncalves, Ecole normale suprieure de Lyon (France); M. R. Caetano, Instituto Geogrfico Portugs (Portugal) Imaging and sensing technologies are constantly evolving so that, now, the latest generations of satellites commonly provide with Earth snapshots at very short sampling periods (daily images). It is unquestionable that this tendency towards continuous time observation will broaden up the scope of remote sensing activities: not only will it enable real-time detection of abrupt changes (e.g. forest fires, natural catastrophes) but also it will allow for accurate estimation of slowly varying trends (urban growth, desertification). In order to efficiently exploit this resulting amount of information, it is necessary to adopt a methodological approach that combine spatial image processing with time series analysis to extract pertinent features from noninformative noise and redundant observations. We believe that time series imaging is a great stake in Earth monitoring, and there is no doubt that theoretical models must accompany this challenge. In this article, we pursue the exploitation of multi-temporal images and investigate a mathematical function to methodically describe the intraannual evolution of spectral reflectance - or any vegetation index - measured at a pixel. We sought for a simple parametric model that is sufficiently parsimonious to concentrate all time information into a small set of pertinent parameters, but versatile enough to reproduce most observed profiles. It is a model corresponding to the nonlinear harmonic solution of a chaotic attractor. The linear component reflects the specific seasonal periodicity associated to any given land cover, while the nonlinear part allows reproducing waveforms exhibiting possible non-symmetries, phase shifts or amplitude modulations. The rationale of our approach can be summarized in three points: 1. Robustness - Parametric models are reasonably insensitive to additive noise in measurements; 2. Compactness - Inter-annual variations of reflectance and vegetation indices are normally cyclo-stationary processes with strong correlation. Parametric models reduce a redundant sequence of samples to a parsimonious set of descriptors; 3. Curse of dimensionality - In most machine learning classifiers there exist a trade-off between the searching space dimension and the number of classes to be differentiated. Parametric models perform a dimension reduction (similar to factorial discriminant analysis) to meet this trade-off and optimize classification performances. As land cover characterization is our first motivation for deriving a model that would match reflectance time series, we needed first to validate the mathematical model from an information perspective. To support adequacy of our model, we report on a land cover classification task based on monthly multi-spectral reflectance images acquired by MERIS sensor at a 300m nominal resolution throughout the years 2004 and 2005. For different combinations of the available spectral bands and dates,

6748-33, Session 8
Information sources fusion approach in forest stand classification
Z. Ben Dhiaf, Univ. de Tunis (Tunisia) and The Univ. of the Antilles and Guyana (France); J. Desachy, The Univ. of the Antilles and Guyana (France); A. Hamouda, Univ. de Tunis (Tunisia) The classification of remote-sensing images based on multiple information sources offers a consistent method for the automatic cartography of forest stands. However, fusion models reveal problems of combinatorial explosion due to the calculation of the assignment functions. This article proposes an information-fusion approach that responds to the need for updating the forest inventory, based on belief theory. It illustrates a solution that overcomes the problem of combinatorial explosion that arises with the

132

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6748: Image and Signal Processing for Remote Sensing


we compared classification performances using raw measurements and model-based synthesized data, in order to evaluate the agreement between our nonlinear model and reflectance time series of land cover classes. We then strive to reduce the full time information (i.e. one image per date) down to a parsimonious set of relevant model parameters to prevent the curse of dimensionality that penalizes most machine learning classifiers. The proposed model is simple (only five parameters), remarkably flexible and easily identifiable. Thus, after we validated the model, we trained and tested several classifiers with the estimated model parameters directly. The goal was to assess the gain in classification performances using different classifiers. In this study we report the comparison of Support Vector Machine (SVM), a new supervised classification system that is known to be quite immune to the feature space dimensionality, with most frequently used supervised parametric and non-parametric classifiers. Moreover, since two phase parameters of the model are periodic variables that lie on the unit circle, and any classifiers that rely on an Euclidean distance may irremediably fail at comparing non standard metrics, we tried to adapt machine learning classifiers to cyclic norms. [3] H. Greidanus et al. Benchmarking operational SAR ship detection, IEEE International Geoscience and Remote Sensing Symposium 2004, Vol. 6, pp. 4215-4218, September 2004. [4] V. Anastassopoulos and G. Lampropoulos, Optimal CFAR Detection in Weibull Clutter, IEEE Transactions on Aerospace and Electronic Systems, Vol. 31, No. 1, pp. 52-64, January 1995. [5] V. Tsagaris, V. Anastassopoulos and G. Lampropoulos, Fusion of hyperspectral data using segmented PCT for enhanced color representation, IEEE Transactions on Geosciences and Remote Sensing, (43), 10, pp. 2365-2375, 2005.

6748-36, Session 9
Fusion of AIS and SAR data for maritime surveillance
S. P. Coraluppi, NATO Undersea Research Ctr. (Italy) In recent years, global political changes have generated significant interest in surveillance applications to combat terrorism, smuggling activities, and illegal immigration. An important theater for these activities is the maritime domain. This has led to a number of national and multinational initiatives in maritime domain awareness (MDA), or maritime situational awareness (MSA), with the goal of having knowledge of all coastal and high-seas activities relevant to national security. As part of these efforts, NATO is pursuing research activities to exploit existing sensor systems in support of MSA. To date, these efforts have focused on exploitation of Automatic Identification System (AIS) transponder data, which is available in principle for all large vessels. As part of these NATO-wide activities, the NATO Undersea Research Centre (NURC) is conducting preliminary research activities to contribute to MSA. NURC has significant experience in multisensor tracking and fusion technology for active sonar-based undersea surveillance. In the maritime domain, sensor assets may include AIS tracks, as well as contacts from coastal radar, video, IR, and SARbased sensors. In this paper, we discuss the use of our distributed multi-hypothesis tracking (DMHT) technology for generating a real-time consolidated surveillance picture. Our initial focus is on the use of AIS and SAR data. NURC contributes to and has access to an AIS network controlled by the NATO Maritime Component Command (CC MAR) based in Naples. Additionally, NURC has a satellite ground station that allows for rapid acquisition and processing of satellite-based imagery. In the paper, using real data examples, we illustrate the use of SAR data to maintain continuous track of targets of interest in regions that lack AIS coverage. The fused AIS-SAR tracks allow for subsequent analysis and the determination of possible anomalous activities.

6748-35, Session 9
A novel ship identification system based on polarimetric ASAR data
G. Panagopoulos, Univ. of Patras (Greece); V. Tsagaris, Univ. of Patras (Greece) and Aratos Technologies (Greece); V. Anastassopoulos, Univ. of Patras (Greece) Introduction Modern SAR satellite sensors precisely measure thousands of square miles within very few minutes. SAR data can be acquired with great reliability to enable precision monitoring of surface processes. Space-based SAR imagery is available in all weather, day or night and has wide area coverage with swaths up to 500 km across. SAR images have high spatial resolution, so they can be used in sea surface ship monitoring. Ships are typically constructed from large flat metal sheets and hence are usually radar bright and therefore detectable in SAR imagery. Ocean going ships tend to be larger and made of metal, whereas coastal boats are smaller and may be made of wood or fiberglass; oil slicks are often associated with ships. In [1], authors provide a brief but useful summary of factors affecting the SAR imaging of ships. The methods for ship detection using SAR data are usually based either on Constant False Alarms Detectors (CFAR) or on image transformation and typically the Wavelet transform. A comprehensive review of detection algorithms can be found in [2] while a benchmarking of operational detection systems was carried out in [3]. Proposed system In this work a new ship identification approach is proposed which makes use of amplitude cross-polarized SAR data of VV-VH and HH-HV combinations. A typical ship detection system consists of the following parts (i) land masking, (ii) preprocessing - calibration (iii) prescreening - thresholding (iv) detection - false alarms rejection and (v) estimation - ship parameters estimation. Steps (i) and (ii) are rather typical while for step (iii) we use a clutter based CFAR detector proposed in [4]. The features for ship identification are mainly derived using shape characteristics from the dual polarized data considering the vessel as distributed targets. The vessel shape is obtained by fusing [5] the signatures of the ship in the cross-polarized images which have been processed for speckle noise reduction. Furthermore, for identification purposes new features are proposed in order to quantify the degree of rotation of EM polarization. The final feature vector contains both dependent as well as independent parameters from ship orientation. The proposed system is evaluated on ASAR data for the Patraikos Gulf in Greece. References [1] F. Askari and B. Zerr, Automatic approach to ship detection in space borne synthetic aperture radar imagery: An assessment of ship detection capability using RADARSAT, Technical Report SACLANTCEN-SR-338, 2000. [2] H. Zhaoying and Ch. Jinsong, A review of ship detection algorithms in polarimetric SAR images, 7th International Conference on Signal Processing 2004, pp. 2155-2158, 2004.

6748-67, Session 9
Coherent change detection and classification in synthetic aperture radar imagery using canonical correlation analysis
M. R. Azimi-Sadjadi, S. K. Srinivasan, Colorado State Univ. (USA) The all-weather, high-resolution Synthetic Aperture Radar (SAR) forms an ideal platform for providing 2-D imagery of stationary ground targets. For detection of moving targets, this would require continuous temporal monitoring of the surveillance area, which is often impractical. However, detecting a cue that threat activities have taken place by monitoring the soil disturbances that are left behind (e.g. vehicle tracks) provides valuable intelligence information. Moreover, using the sequence of the detected change features, it is even possible to back-track and locate the origin of the activities and/or their behavior. However, an automated procedure is needed to accurately detect the threat activities from a large sequence of SAR imagery and extract the relevant features for subsequent classification. Coherent Change Detection (CCD) is the process of detecting changes from pairs of SAR images of approximately the same location observed at two different instances. Several factors complicate this problem. These include: changing environmental conditions like soil moisture, vegetation and canopies, seasonal effects, and speckle noise in SAR imagery. These complications make the detection of weak target activities even more difficult. More importantly, traditional CCD

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

133

Conference 6748: Image and Signal Processing for Remote Sensing


methods are typically based on ad-hoc procedure of pixel intensity correlation, the effectiveness of which is highly dependent on the choice of a threshold. Moreover, these methods dont provide coherence feature vector needed for classification purposes. Extracting salient change features from CCD imagery that allow for successful discrimination of repetitive and benign change activities from possible threats is of great importance. This problem is complicated due to high false alarm rate in highly active areas as well as the presence of subtle threat activities. Consequently, any realistic CCD feature extraction and classification system should possess several key properties that are: (a) ability to effectively detect and characterize the changes due to man-made activities, (b) the ability to detect and extract coherent change features from weak or subtle target activities, (c) the ability to represent the coherent change features in reduced dimensional space. This is highly desirable property when coherent change features are to be archived or transmitted to other decision making systems, (d) the ability to reduce the false alarms by rejecting repetitive and benign activities that occur in the environment, (e) the ability to aggregate feature vectors over time and provide a temporal-spatial history of anomalous activities that can be useful to trace the behavior of the relevant threat activities, (f) the ability to perform such operations in cluttered environments. In this paper, we propose a novel canonical correlation-based approach for simultaneous coherent change detection, feature extraction and classification. Canonical coordinate decomposition allows us to quantify the changes between the pairs of SAR images and extract useful features directly without any need for to perform separate CCD and anomaly feature extraction. Moreover, if any visual analysis with/without human intervention is required, the CCD images can be reconstructed from the coherent change features. We have shown that canonical coordinate transformation of the raw SAR images allows us to maximally capture the divergence (or coherent change) information between the two images with a reduced feature dimension. As a result, even the weak coherent change information could be detected very effectively compared to any other CCD-based schemes. The inherent data reduction in the transformation of the raw SAR images to canonical coordinate change features, results in reduced bandwidth requirement in the case of data transmission or less database memory requirement in the case of archiving those changes. The reduced dimensional feature set and the extraction of canonical coordinate change features without performing CCD reduces the computational load considerably. The optimality of our proposed method in capturing the coherent change information with reduced dimensional feature vector is a major benefit of the proposed approach. The effectiveness of the developed algorithms are demonstrated on several SAR data sets. These data sets contain different types of activities of interest, e.g. the vehicle tracks, with different degree of difficulty. templates flat regions are larger, and more errors are corrected. Furthermore, because the rough blocks are smaller, the errors in the interpolation are also reduced. This method was applied to correct the NOAA images receiving in Tokyo, Bangkok and Ulaanbaatar. The results proved that this is a high accurate correction method.

6748-40, Session 10
GIFTS SM EDU Level 1B algorithms
J. Tian, M. J. Gazarik, R. A. Reisse, D. G. Johnson, NASA Langley Research Ctr. (USA) The Geosynchronous Imaging Fourier Transform Spectrometer (GIFTS) Sensor Module (SM) Engineering Demonstration Unit (EDU) is a high resolution spectral imager designed to measure infrared (IR) radiance using a Fourier transform spectrometer (FTS). The GIFTS instrument employs three Focal Plane Arrays (FPAs), which gather measurements across the long-wave IR (LWIR), short/mid-wave IR (SMWIR), and visible spectral bands. The raw interferogram measurements are radiometrically and spectrally calibrated to produce radiance spectra, which are further processed to obtain atmospheric profiles via retrieval algorithms. This paper describes the processing algorithms involved in the calibration stage. The calibration procedures can be subdivided into three categories. In the pre-calibration stage, a phase correction algorithm is applied to the decimated and filtered complex interferogram. The resulting imaginary part of the spectrum only contains the noise component of the uncorrected spectrum. Additional random noise reduction can be accomplished by applying a spectral smoothing routine to the phase-corrected spectrum. The phase correction and spectral smoothing operations are performed on a set of interferogram scans for both ambient and hot blackbody references. In the calibration stage, we first compute the spectral responsivity based on the previous results, from which, the calibrated ambient blackbody (ABB), hot blackbody (HBB), and scene spectra can be obtained. In the post-calibration stage, we estimate the Noise Equivalent Spectral Radiance (NESR) from the calibrated ABB and HBB spectra. We then implement a correction scheme that compensates for the effects of foreoptics. The FPA off-axis effect correction technique also will be presented.

6748-41, Session 10
Disturbances and their corrections in space observation with GOSAT Fourier transform spectrometer
T. Aoki, T. Yokota, National Institute for Environmental Studies (Japan); G. Inoue, Nagoya Univ. (Japan); K. Nobuta, A. Kotani, Fujitsu FIP Corp. (Japan) GOSAT is the satellite to measure the greenhouse gases from space with the Fourier Transform Spectrometer (FTS). It measures a spectrum of Earth-reflected solar radiations in high spectral resolution of about 0.2 cm-1, covering four spectral bands, including 0.76, 1.6, 2.0 and 14 micron wavelength regions. In first three bands, we have two detectors for measuring two components of polarization. For an acquisition of one interferogram, it takes 4 seconds or less, depending on the definition of measurement mode. Since the satellite moves in high speed, an image-compensation mirror (IMC) works to continuously stare a same position on the surface during this period of acquisition. For staring, the mirror is controlled by two-axis motors but it could slightly fluctuate around the right position, making the instantaneous filed of view (IFOV) vibrate around its mean position. Since the optical characteristics (such as the albedo) of the IFOV changes place-by-place, the intensity of IFOV could also change due to this fluctuation. The intensity of IFOV could also change due to the change of reflection angle, wind on the water surface, or other causes. During this time period of acquisition, the optical path length and Doppler shift caused by the satellite moving could also change. In this paper, we examine the effects of these kinds of disturbances to the signals of interferograms on resultant spectra and retrieval accuracies of CO2, and discuss about the correction method to the interferogram and spectra.

6748-39, Session 10
High accurate geometric correction for NOAA AVHRR data considering the variation of elevation effect
A. N. Van, Y. Aoki, Shibaura Institute of Technology (Japan) In recent years, NOAA images have been provided very useful information about ecosystems, climate, weather and water from all over the world. In order to use NOAA images, they need to be transformed from image into map coordinate system. This paper proposes a method that corrects the errors caused by this transform. According to the current method, errors are corrected in the image coordinate system by comparing the blocks in the image with the GCP templates, which contain GCPs and the flat regions around them. The corrected image is then transformed into map coordinate system. The map positions of the corner points in each block are precisely calculated based on their elevation and the satellites parameters; the other points are specified by bilinear interpolation. However, this interpolation process will be wrong with rough blocks, where the elevation is much different from one point to another. In the proposed method, the variation of the elevation is checked to find out rough and flat blocks. The rough blocks are divided again, one or more times, into smaller flat and rough blocks. Flat blocks are then combined with the templates to enlarge the flat regions. For this reason, the

134

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
Monday-Thursday 17-20 September 2007 Part of Proceedings of SPIE Vol. 6749 Remote Sensing for Environmental Monitoring, GIS Applications, and Geology VII

6749-01, Session 1
From space to place or from site to landscape? Mind the gap!
S. R. L. Campana, Univ. degli Studi di Siena (Italy) In recent years in the University of Siena we have been building up our experience in the use of integrated prospection approaches in the study of archaeological landscapes. We have been aware from the outset that our research would have to focus initially on settlements rather than on landscapes as such. Inevitably, the intensity of archaeological research would be less in the intervening spaces, which for the most part would therefore remain as areas of emptiness. We can start our discussion from Aiali, a place on the lowland between the medieval town of Grosseto and the Roman town of Roselle, in central Italy. The archaeological area here was first identified in 2001 during an Aerial Archaeology Research School organized by the University of Siena and colleagues from across Europe. Aerial survey during the School allowed us to recognize at Aiali an area within which the growth of the wheat varied in such a way as to reveal an articulated group of features that made up the plan of a complex of structures interpreted as a Roman villa, covering an extent of up to 4 hectares. In the following years Aiali has become the most important test site of the Universitys Laboratory of Landscape Archaeology and Remote Sensing. From 2001 onwards we have collected, processed and interpreted many different kinds of data for the site: Quickbird-2 satellite imagery; historical and recent vertical photographs (from 1954 to 2001); oblique photographs in a number of different years, seasons and lighting conditions; field-walking survey and geophysical survey (magnetometry, GPR, EM, ERT). Aiali has become a project upon which, on the one hand, we can apply the highest available level and intensity of archaeological prospection methods on a large, complex and stratified site, with settlement extending from the Etruscan period through Roman times and into the Middle Ages. At the same time Aiali can also be seen as a starting point for a wider approach to the study of the landscape between Grosseto and Roselle. This is an innovative kind of project in Italy but it is clear that we are trying to emulate the strategy applied for more than thirty years by Dominic Powlesland in his remarkable archaeological and landscape study at West Heslerton, in the Vale of Pickering in northern England. Clearly it will never be possible to extend this approach to Tuscany as a whole, bearing in mind the scale of archaeological mapping that we are attempting in this part of Italy. For example, the archaeological map of Grosseto province alone covers a total of 4030 sqkm while Tuscany as a whole extends to nearly 23000 sqkm. The archaeological objectives and outcome of the Aiali project will also have to take account of the critical impact of the kinds of information that are available for recording and assessing the potential and interpretation of the landscape. In attempting this it will be at least as important to appreciate what is not visible as to know what is visible. If we can achieve success in our sample area at Aiali through the application of the highest level of research intensity, and through conscious attempts to raise the level of visibility, and therefore understanding, in the emptinesses between identifiable sites, we will be able to make worthwhile contributions to better practice in a variety of fields: Conservation - increased awareness of the archaeological resource as a whole, so as to create more effective and betteradapted policies for landscape monitoring and conservation Academic issues - the recognition of emptiness (that is, absence of evidence) as an important element in the development of ideas about settlement patterns and landscape history Future archaeological research and the development of research strategies - the search for better visibility in our present emptinesses will hopefully produce new and perhaps different kinds of data, in turn creating new kinds of feedback into the investigative and interpretative process.

6749-02, Session 1
Techniques for improving the detection of archaeological features from satellite imagery
A. R. Beck, Univ. of Leeds (United Kingdom); G. Philip, Durham Univ. (United Kingdom) In common with all domains of remote sensing, residues indicative of past human activity can only be detected if they exhibit some form of identifiable contrast with their surroundings. Unlike many other domains these residues do not exhibit consistent spectral signatures: rather archaeological spectral responses are commonly expressed as subtle deviations from their surrounding matrix. This is true for crop marks, soil marks and thermal anomalies. The challenge is to collect imagery when the contrast between archaeological residues and the background matrix is maximized and thus to find algorithms that will enhance these, sometimes subtle distinctions so that they can be more readily detected. This paper will present work undertaken in the semi-arid environment of Homs, Syria. The project area includes two contrasting environmental zones with a differing repertoire of archaeological remains: a basalt zone (120 km2) and a marl zone (480 km2). Declassified Corona space photography and Ikonos satellite imagery (panchromatic and multispectral) were evaluated to determine their efficacy for detecting a range of different archaeological residues. No single image set was able to provide the best result for the two zones, as each required imagery collected under different environmental conditions. The specific requirements for each area will be reviewed, in particular: determination of the correct environmental conditions for enhanced archaeological contrast; image pre-processing techniques; presentation of a new moving kernel technique for enhancing subtle archaeological contrasts. Finally, thought will be given to how these techniques can be applied in other environmental situations and a framework presented for developing benchmark data which will aid in the programming of hyperspectral archaeological surveys.

6749-03, Session 1
Remote sensing and archaeological survey in the Hierapolis of Phrygia territory, Turkey
G. Scardozzi, Consiglio Nazionale delle Ricerche (Italy) The paper concerns the results of a research project on the application in archaeological survey of high resolution images acquired by QuickBird satellite. The research is carried out within the activities of the Italian Archaeological Mission at Hierapolis of Phrygia (Turkey). The use of satellite images with high geometric and radiometric resolution has constituted an important tool for archaeological research in the settlements and in the surrounding area, because vertical aerial photos and recent and detailed cartographies are non-available. In fact the exceptional spatial resolution of the images makes them comparable to aerial photos on a medium scale; this type of documentation has an enormous potential in the study of urban and territorial ancient contexts. The examination of these images has permitted to detect surface anomalies and traces linked to archaeological buried structures or to paleo-environmental elements; they also provide data which are highly interesting for the reconstruction of the ancient landscape. Moreover, particulary in the territory, the panchromatic images were georeferenced and used as the base field maps for the survey, in integration with a GPS sistem. The study of these images has made a fundamental contribution to the reconstruction of the urban layout of Hierapolis, with the integration of oblique aerial photos and ground truth verify. Also much interesting were the results obtained in the territory of the city, where are recovered numerous and important data on necropolis, aqueducts, roads, farms, quarries and villages dependent from Hierapolis. All the data collected are integrating into a GIS to produce archaeological maps.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

135

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology

6749-04, Session 1
Best practice approaches for applying satellite imagery for landscape archaeological applications: a case study around the World Heritage Site of Sanchi, India
A. R. Beck, Univ. of Leeds (United Kingdom); J. Shaw, Univ. College London (United Kingdom) Satellite imagery is an increasingly important tool for cultural and natural heritage management. It has particular relevance in those areas of the world where the heritage resource is poorly understood. In these areas there is little understanding of the range, nature and extent of heritage residues. Furthermore, what is known may be significantly biased: i.e. heritage management strategies may have been skewed towards a specific type of remain (normally monumental architecture). This paper will present work undertaken in the landscape around the UNESCO World Heritage site of Sanchi, Madhya Pradesh India. The project area consists of two major geological zones: Vindhyan sandstone scarpland and Deccan basalt. Although the Sanchi World Heritage site has been intensively studied the surrounding landscape has received significantly less attention The techniques presented extend the work conducted by Beck and Philip in the Homs region of Syria and importantly refine their techniques for an area with different environmental conditions and different archaeological characteristics. Rather than discuss the merits of individual sensors this paper takes a more holistic approach and examines the life-cycle of satellite imagery for an archaeological project. This means that satellite imagery is viewed not just as a source of archaeological information but also a data source that can be used to contextualise and interpret the archaeological resource. Hence this paper provides a framework which should allow archaeological investigators to select, manipulate and integrate different satellite sensors to provide information which is fit for purpose. This paper discusses the implications of satellite sensors for different activities including; archaeological prospection, landuse mapping and terrain modelling and considers how the synergies of different satellite and archaeological data can be exploited.

common remote sensing data source used for the study of ancient landscape. More recently, the application of satellite imagery can fruitfully aid the identification of large-scale cultural features, such as ancient land use patterns, roads, irrigations networks, paleohydrological systems. In particular, the recent availability of Very High Resolution (VHR) satellite images, such as IKONOS (1999) and QuickBird (2001), may provide new perspectives in the field of archaeological prospection ranging from small details (single building) to landscape archaelogy analyses which involve investigations of large areas. Compared to aerial photo, the VHR satellite data can be promptly georeferenced, offer a very large coverage, are available at lower costs and for a wider spectral range. The VHR satellite imagery could be ideal for investigations on regional scale as well as for studies performed in areas where aerial photography is restricted because of military or political reasons. Nevertheless, the possibility of locating unknown individual sites as well as large scale cultural features is highly dependent on many factors, such as, the image spatial resolution, extension of buried sites, ground characterists, illumination conditions. The practical potential of satellite-based approaches needs to be tested in relation to the specific archaeological problems of individual regions. This paper provides a review of satellite Quickbird data processing successfully applied in the achaeological context from site identification to landscape archaelogy analyses.

6749-07, Session 2
Finding archaeological cropmarks: a hyperspectral approach
A. Aqdus, W. S. Hanson, J. Drummond, Univ. of Glasgow (United Kingdom) Aerial photography has made the single most important contribution to our improved appreciation of the density, diversity and distribution of archaeological sites in Britain since WWII. This is particularly the case for areas of intensive lowland agriculture where ploughed-out sites are known only from marks in crops growing above them. However, reconnaissance for such cropmarks is not equally effective throughout the lowlands because of the particular conditions of drier weather, welldrained soils and arable agriculture required before they become visible. In Scotland, for example, there is considerable bias in the discovery and, consequently, known distribution of archaeological sites in favour of the drier eastern side of the country, with its higher percentage of arable agriculture, as opposed to the west with its wetter climate and greater proportion of grazing land. Given that the appearance of cropmarks is linked to moisture stress in growing plants, they are potentially detectable at bandwidths outside the visible and before they become apparent therein. Using a range of imagery (CASI 2, ATM and digital vertical photographic data) from two case study sites in Lowland Scotland to facilitate comparisons, one in the east and one in the west, this paper considers the extent to which hyperspectral imagery can enhance the identification of otherwise invisible archaeological sites.

6749-05, Session 2
Reconstructing the archaeological landscape of Southern Dobrogea: integrating imagery
I. A. Oltean, W. S. Hanson, Univ. of Glasgow (United Kingdom) The recent integrated aerial photographic assessment of Southern Dobrogea (Romania) is part of the first authors British Academy funded research programme Contextualizing change on the Lower Danube: Roman impact on Daco-Getic landscapes. This seeks to study the effect of the Roman conquest and occupation on the native Daco-Getic settlement pattern on the Lower Danube. The methodology involves integrating a range of remotely sensed imagery including: low altitude oblique aerial photographs, obtained through traditional aerial reconnaissance; medium altitude vertical photographs produced by German, British and American military reconnaissance during the Second World War, selected from The Aerial Reconnaissance Achive at Keele University; and high altitude de-classified military satellite imagery (Corona) from the 1960s, acquired from the USGS. The value of this approach lies not just in that it enables extensive detailed mapping of large archaeological landscapes in Romania for the first time, but also that it allows the recording of archaeological features permanently destroyed by more recent development across wide areas. This paper presents some results and addresses some of the problems raised by each method of data acquisition.

6749-08, Session 2
Undercanopy archaeology using airborne laser scanner to overcome the Mediterranean vegetation
S. R. L. Campana, Univ. degli Studi di Siena (Italy); D. Donoghue, N. Galiatsatos, Univ. of Durham (United Kingdom) The Department of Archaeology at Siena has been engaged for several decades in the testing of new methodologies, new approaches and new instruments to help reconstruct the archaeological record. Such methods and technologies have allowed us to map a large number of archaeological sites in Tuscany, up to about 18000. Most of the evidence come from ploughed areas or pasture land that represents about the 40% of Tuscan land cover. However, 50% of Tuscany is covered by forest and a quite similar percentage characterizes the majority of the Italian landscape. This means that conventional archaeological prospection techniques are not successful in areas covered by woodland. Nevertheless, we expect that the dense Mediterranean canopy hides a wealth of important archaeological evidence.

6749-06, Session 2
On the use of satellite Quickbird imagery in archaeological context
R. Lasaponara, Istituto di Metodologie per lAnalisi Ambientale (Italy); N. Masini, Consiglio Nazionale delle Ricerche (Italy) During the last century, aerial photos have been the most

136

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
The aim of this study is to identify the potential of dual-return airborne laser scanning technology (LiDAR) to map the undercanopy archaeology (about 10,000 km2). For this purpose, and in partnership with the UK Natural Environment Research Council, we acquired four sample areas which are representative of the different kinds of cultural sites and different kind of canopy cover between the west coast of Tuscany and the south countryside of Siena. At the present stage of research we can summarize the main objectives of using for archaeology by using sub-canopy laser penetration to characterise the underlying micro-relief. The results show how the laser signals penetrate different kinds of Mediterranean canopy cover and the results are interpreted in terms of archaeological sites typically found in Tuscany. The potential of LiDAR to advance knowledge is enormous because we hypothesise that many of the features are protected by dense woodland cover. This is in contrast to agricultural areas where the sites are frequently destroyed or flattened. Exploring the woodland areas should allow us to find sites before their destruction (partially or almost totally) and to plan good conservation policies. protection through World Heritage recognition to a number of islands and atolls in the central Pacific Ocean, including the Phoenix and Line Archipelagos. Most of these atolls and islands lack natural resource maps, needed for the identification of priority areas for inclusion in a marine reserve system. Our project provides assistance to UNESCOs World Heritage Centre by developing benthic habitat maps from high-resolution multispectral imagery. The approach involves: (i) the analysis of new Quickbird multispectral imagery for the Phoenix and Line Islands, a test area within the WHCPP; and (ii) the use of a MARXAN, a simulated annealing algorithm that uses a GIS interface. Analysis of satellite imagery was performed with ENVI(r), and includes standard radiometric and geometric corrections, removal of atmospheric effects using ATCOR (a MODTRAN4 radiative transfer model), de-glinting and water column correction algorithms, and a number of unsupervised and supervised classifiers. Ground-truth data collected in-situ is used to assess the accuracy of the resulting habitat maps, which are then used as input in to MARXAN, which identifies a proportion of each habitat in a spatial arrangement that accounts for connectivity among habitats and distribution along environmental gradients. The outputs of this research are being delivered to the UNESCO World Heritage Centre office and the Kiribati Government to assist in the marine reserve planning.

6749-09, Session 2
The effect of pixel resolution and spectral characteristics on the extraction of archaeological features from very highresolution remote sensing imagery: Sagalassos, southwest Turkey
V. L. De Laet, E. Paulissen, Katholieke Univ. Leuven (Belgium); K. Meuleman, Flemish Institute for Technological Research (Belgium); M. Waelkens, Katholieke Univ. Leuven (Belgium) Abstract The launch of several very high spatial resolution satellite (VHSRS) systems (Ikonos-2, Quickbird-2 and others) in the recent past also has provided new possibilities for archaeological research. The emphasis of this paper is to compare and evaluate the contribution of spectral characteristics and pixel resolution of Quickbird-2 and Ikonos2 for automatic extraction of ancient features from VHSRS imagery. The spectral characteristics of both images have been evaluated by a band-by-band comparison. Apart from a visual comparison, pixel- and object-based classification techniques are applied to assess the effect of different image characteristics. The study is carried out on the antique site of Sagalassos (southwest Turkey). A profound analysis of the VHSRS data reveals that Quickbird2 outperforms Ikonos-2 for the visual identification of ancient remains due to its enhanced ground resolution. For archaeological sites in general, the accuracy of the automatic extraction depends on several variables: the type and characteristics of VHSRS data, the classification method applied, the spectral variation within the site and the shape characteristics of the remnants. For Quickbird-2, object-based extraction appears independent of the site characteristics, which largely influence extraction on Ikonos-2. This study shows that object-based extraction on Quickbird-2 imagery better performs for archaeological applications in general.

6749-11, Session 3
Evaluation of spatial interpolators in the framework of a geographical information system: a case study using oceanographic data
D. Kitsiou, T. E. Nitis, M. N. Kostopoulou, N. Maravelakis, Univ. of the Aegean (Greece) Spatial interpolation methods are widely applied in marine studies to evaluate the spatial distribution of oceanographic parameters and inter-compare time series of maps of selected variables. A variety of such methods are nowadays available and selection of the most appropriate for a specific case study is not an easy task. Moreover, the accuracy of interpolated maps is considered of high importance, since reliable evaluation of the spatial structure of oceanographic phenomena is required. Within geography and other spatially oriented disciplines, and most of the times in the framework of a Geographical Information System, several attempts have been carried out to assess the efficiency of various spatial interpolators using diverse methodologies. In this paper, an attempt was carried out to evaluate the accuracy of two spatial interpolators for mapping the distribution of organic carbon (weight %), an important indicator of marine sediments, in the marine environment. Measurements of organic carbon were carried out in a network of 20 sampling sites in the Gulf of Gera, which is representative of a semi-enclosed and shallow marine ecosystem at the south-western part of the island of Lesvos, Greece. For the interpolators under study, the cross-validation error was calculated at each sampling station. The process involved removing one sampling station from the network at a time and then applying the spatial interpolator to the organic carbon values measured at the remaining 19 sampling stations. This process was therefore repeated 20 times and the crossvalidation error at each station was the difference between the measured and the interpolated value. Subsequently, calculation of the RMSE (root-mean-square error), the MAE (mean-absolute error) and the MBE (mean-bias error) was carried out to assess the accuracy of the performance of the interpolators under examination. The calculation of the three above mentioned error statistics was necessary, since their efficacy in describing the interpolators performance error varies. The result of the application of the two interpolators under study was illustrated on thematic maps and discussion on the potential increase of their accuracy was carried out.

6749-10, Session 3
Using high resolution multispectral imaging to map Pacific coral reefs in support of UNESCOs World Heritage Central Pacific project
D. Siciliano, R. C. Olsen, Naval Postgraduate School (USA) Concern over worldwide declines in marine resources prompted the search for innovative solutions for their conservation and management, particularly for biodiversity hotspots such as coral reef ecosystems. Effective protection of marine environments calls for representation of all habitat types, connected biologically and distributed along environmental gradients. Rapid advances in sensor resolution and image analysis techniques, tailored to the unique optical problems of marine environments, have allowed the derivation of detailed benthic habitat maps representing coral reefs main ecological communities from multispectral satellite imagery. UNESCOs World Heritage Central Pacific Project (WHCPP) plans to afford

6749-12, Session 3
Ship traffic monitoring using satellite SAR images in combination with AIS reports
D. J. Weydahl, R. Olsen, Norwegian Defense Research Establishment (Norway) Ship traffic monitoring may be performed using satellite SAR data. The advantage with the SAR sensor is the all weather and day/night imaging capability. However, the SAR backscatter contrast between a vessel and the surrounding sea state may

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

137

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
be small in high wind conditions and at small incidence angles. The present and future SAR satellites will have the capability of imaging the earth surface with several incidence angles, and with dual-polarimetry (HH/HV, VV/VH or HH/VV). The ship/clutter contrast may therefore be increased by using a higher incidence angle during the acquisition, and/or by applying different polarisation combinations. After detecting the ship candidates with the satellite SAR system, one may identify the candidates by using an elevated AIS sensor. There are plans for placing AIS systems onboard satellites. It will then be possible to combine the information from satellite SAR with those from satellite (or ground-based coastal) AIS and thereby identify all the detected ships within a SAR image. This data fusion will enable us to develop further knowledge about SAR backscatter properties from vessel types that may not be detected so well using the SAR data only. On the other side, it will be possible to pin-point those ship candidates that do not carry an AIS system, and thereby take appropriate security or rescue actions. assess the impact. The expanse of area in the post treatment period in respect of irrigated crop agro-horticulture and water bodies allow to infer that the intervention of the implementing agency in the watershed was instrumental for such aerial expansion. This also signifies that there is augmentation of soil moisture condition, which is reflected in the statistics on coverage of irrigated crops and overall increase in the biomass of the watershed. The output of the impact analysis could be taken up for quantitative assessment of various measures that were responsible for significant improvement in the watershed as well as to take up corrective measures where improvement was insignificant.

6749-15, Session 3
Regional ecological risk assessment by remote in Bosten Lake
W. Liu, Xinjiang Univ. (China) Taking the area around Bosten lake as an example, this paper tries to make a systematical appraisal on regional risk toward the area around the largest inland freshwater lake in arid regions with the method of ecological environmental evaluation and the Remote Sensing and GIS. Its purpose lies in offering the theory and technical support for regional risk management, which will play an important realistic role in improving the ecological environment of Bosten lake. Studies shows that, suffering from high mineralization and eutrophication, meanwhile, the reed growth declines constantly with the water area dwindling and water environment worsening. The region around Large Lake river mouth where enters Kaidu river and West Pump station is a high taking place area of flood, the area of Southwest Small Lake, with serious mineralization, is withering constantly, and the average water quality of Large Lake is the III grade. Through the projects of wetland recovery and bio-diversity protection; through East Pump station project , water circulation along the lake is strengthened and floodpreventing standard raised as well in Large Lake region, and with the safe water convey of downstream Peacock river by pumping up the water in pump station, salt river and other pollutants have been successfully drained off in time. It is pointed out that only by taking the way of regional sustainable development, can the greatest benefit of the ecological environment and social economy be achieved.

6749-13, Session 3
Long-term monitoring of sea-based oil pollution in the European seas
G. Ferraro, S. Meyer-Roux, O. Muellenhoff, D. Tarchi, K. Topouzelis, Joint Research Ctr. (Italy) Accidental pollution at sea can be reduced but never completely eliminated, on the other side, deliberate illegal discharges from ships can indeed be reduced by the strict enforcement of existing regulations and the control, monitoring and surveillance of maritime traffic. Notwithstanding, operational oil discharges is a common practice and represents the main source of marine pollution from ships. Consequently, JRC has focused its attention on the need to monitor in the long term the problem of sea-based oil pollution in all European seas. For this reason, JRC has collected aerial surveillance data and satellite imagery from different actors and archives, even if the data for the seas around Europe are not homogeneous. In this paper, JRC intends to present the first results of the analysis of the created database on oil spills. In conclusion the long term oil spill monitoring is the key instrument to assess the implementation of maritime transport regulation such as the Directives on port reception facilities and on ship-source pollution and can be seen as an environmental indicator of the European Seas.

6749-16, Session 4 6749-14, Session 3


Remote sensing and GIS application for impact evaluation of watershed development programme
S. N. Das, A. K. Barman, Indian Agricultural Research Institute (India) Soil erosion due to abrasive action of runoff is very acute in India as many rivers and tributaries have criss-crossed the country. It is reported that around 150 m ha of land suffers from soil erosion out of 329 m ha, the total geographical area of the country. Watershed development programme is one of the major initiatives taken by various ministries and departments viz. Ministry of Agriculture, Ministry of Rural Development and Ministry of Environmental Forest towards Natural Resources Management as it allows management of resources following natural harmony. Thus, tremendous efforts are made to mollify the processes of soil erosion so as to enhance the agriculture production vis-vis restoration of fragile eco system. The interventions made by various implementing agencies are monitored and evaluated through conventional means. The advent of remote sensing technology and Geographic Information System allow to monitor and evaluate the impact of various soil and water conservation measures through scientific means. The paper deals with the impact evaluation studies of watershed development carried out in a watershed of Simoga district, Karnataka. Digital image analysis of satellite data pertaining to pre-treatment and post treatment period have been carried out for impact evaluation. The change in the land use and land cover during pre and post treatment period alongwith transformation of NDVI of both the periods were analysed to

Remote sending and GIS for land-use planning in industrial and urban areas
A. Marino, M. Ciucci, M. Mariani, A. Moccaldi, Istituto Superiore per la Prevenzione e la Sicurezza del Lavoro (Italy) A correct urban and land-use planning can be supported by several innovative tools and methodologies. In this paper the potential of remote sensing and GIS technologies have been emphasized. The proposed methodology has been applied to the River Pescara Valley, considered as the appropriate coverage to define environmental features that influence the industrial areas. For this research MIVIS hyperspectral (at a 1500 and 3000 m elevation) images have been used. The obtained images have been georeferenced. From the processing and classification of these images some information layer have been obtained: thematic maps of land-use (industrial areas identification), vegetation conditions, thermal pollution, quality parameters (temperature, organic matter, chlorophyll, sediments) for river and sea waters. Thematic maps obtained from remote sensing have been inserted in a GIS, that means a system to insert, store, integrate, extract, retrieve, manipulate and analyze georeferenced data layers in order to produce interpretable information. Then the data base has been integrated with further information inserted as continuous layers; thematic layers; vector layers; punctual data; attributes. Some specific operators have been applied that allowed to integrate the information contents and therefore to obtain final thematic maps (environmental quality maps, vulnerability and risk maps, visualization of models related to accidental events). The innovative technologies proposed facilitate and optimize the duties required from actual regulations, as for a recent Italian regulation aimed to the verification and research of compatibility of major hazard industrial plants with land-use and environment. Therefore it is useful to develop methodologies supporting industry and Competent Authorities.

138

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology

6749-17, Session 4
Mountain pine beetle detection, monitoring and modeling: enhancement, interpretation and evaluation of airborne imagery
A. C. Roberts, C. Bone, S. Dragicevic, J. Northrup, Simon Fraser Univ. (Canada) The processing and evaluation of digital airborne imagery for detection, monitoring and modeling of mountain pine beetle (MPB) infestations is evaluated. The most efficient and reliable remote sensing strategy for identification and mapping of infestation stages (current to red to grey attack) of MPB in lodgepole pine forests is determined for the most practical and cost effective procedures. Evaluation of applied RS tools (imaging systems, analytical procedures and predictive infestation spread models) for the general monitoring and control of forest pathogens and pests as well as related procedures that apply to other environmental monitoring practices are considered (e.g. riparian vegetation, erosion, logging practices, suspended sediment concentrations, stream morphology and habitat). This research was planned to specifically enhance knowledge by determining the remote sensing imaging systems and analytical procedures that optimize resource management for this critical forest health problem. Within the context of this study, airborne remote sensing of forest environments for forest health determinations (MPB) is most suitably undertaken using multispectral digitally converted imagery (aerial photography) at scales of 1:8000 for early detection of current MPB attack and 1:16000 for mapping and sequential monitoring of red and grey attack. Digital conversion should be undertaken at 16 to 20 microns for B&W multispectral imagery and 18 to 24 microns for colour and colour infrared imagery. From an operational perspective, the use of twin mappingcameras with colour and B&W infrared film will provide the best approximation of multispectral digital imagery with near comparable performance in a competitive private sector context (open bidding).

6749-21, Session 5
Space oil spills atlas of the offshore United Arab Emirates
S. M. Issa, United Arab Emirates Univ. (United Arab Emirates) Space Oil Spills Atlas of the Offshore United Arab Emirates PIs Name* (Salem Issa), Department (Geology), E-mail: salem.essa\@uaeu.ac.ae U.A.E. University, Al-Ain, P.O. Box: 17551, U.A.E. 1st Co-Is Name (Hussein Harahsheh), Global Scan Technology, 2nd Co-Is (Manabu. Shiobara), Japan Oil Development Co., Ltd., 3rd Co-Is (Takashi. Nishidai), JGI. Abstract The purpose of this study was to locate most oil-threatened areas and to document occurrences of oil spills through the production of space oil spills Atlas in the Arabian Gulf, offshore UAE. A number of satellite data were processed and visualized from early satellites of the mid-seventies through to the most recent Earth observation satellites. Shipping routes bordering the UAE with relatively frequent incidences of oil spills included the offshore area running parallel to the coast of Abu Dhabi, Dubai, Sharjah and Ajman, and the routes through the Strait of Hormuz. Visual analysis revealed a close relationship between spills, oil fields and shipping routes. Monitoring of natural seepage of oil in the area offshore of Abu Dhabi showed numerous small oil slicks caused by natural seeps from the seafloor. This oil spills Atlas achievement by the UAEU-remote sensing lab is an important step towards the rise of the awareness in the country and the establishment of an early warning system in the Gulf region.

6749-22, Session 5
Soil erosion extraction and the key factors in complex topographical region
J. Pan, Chongqing Jiaotong Univ. (China) Comparing the flat topographical area, it is difficult to extract soil erosion in complex one. RUSLE is the most comprehensive model for soil erosion, but it just supplies in the normal condition. So it is necessary to study suitable factor algorithm for the complex topographical region. The paper proposes new algorithm for topographical factor and cover-management factor that are affected by complex topography seriously, including topographical factor calculation for steep slope and vegetation fraction extraction based on remote sensing classification. The test indicates the algorithm is more effective than the normal algorithms of RUSLE

6749-18, Session 4
Landslide susceptibility analysis using an artificial neural network model
S. Mansor, B. Pradhan, Univ. Putra Malaysia (Malaysia) This paper presents landslide susceptibility analysis around Cameron highland area, Malaysia, using an artificial neural network model; a Geographic Information System (GIS) and remote sensing techniques. Landslide locations were identified in the study area from interpretation of aerial photographs and from field surveys. Topographical, geological data and satellite images were collected, processed, and constructed into a spatial database using GIS and image processing tools. The factors chosen that influence landslide occurrence were: topographic slope, topographic aspect, topographic curvature and distance from drainage, all from the topographic database; lithology and distance from lineament, taken from the geologic database; land cover from TM satellite images; and the vegetation index value from Landsat satellite images. These factors were used with data mining model by using an artificial neural network to analyze landslide susceptibility. Each factors weight was determined by the back-propagation training method. Then the landslide susceptibility indices were calculated using the trained back-propagation weights, and the landslide susceptibility map was created in GIS environment. Landslide location data was used to verify results of the landslide susceptibility maps and to compare them. GIS data was used to efficiently analyze the large volume of data, and the artificial neural network proved to be an effective tool to analyze landslide susceptibility.

6749-23, Session 6
MoZis: mobile zoo information system
U. Michel, Univ. Osnabrck (Germany) This paper describes a new project of the Institute for Geoinformatics and Remote Sensing, funded by the German Federal Foundation for the Environment (DBU, Deutsche Bundesstiftung Umwelt www.dbu.de ). The goal of this project is to develop a mobile zoo information system for Pocket PCs and Smart phones. Visitors of the zoo will be able to use their own mobile devices or use Pocket PCs, which could be borrowed from the zoo to navigate around at the zoos facilities. The system will also provide additional multimedia based information such as audio-based material, animal video clips and maps of their natural habitat. People could have access to the project at the zoo via wireless local area network or by downloading the necessary files using a home internet connection. Our software environment consists of proprietary and nonproprietary software solutions to be as flexible as possible. Our first prototype was developed by Visual Studio 2003 and Visual Basic .Net.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

139

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology

6749-24, Session 6
Contents correlation and genetic algorithmsbased remote sensing images fusion
Y. Na, Xidian Univ. (China); M. Ehlers, Univ. Osnabrck (Germany); H. Ji, L. Shi, Xidian Univ. (China) High spatial resolution Panchromatic images can provide detail geometric information and lower spatial resolution multi-spectral images can provide detail spectral information. High spatial resolution multi-spectral images can be obtained with images fusion methods. Many images fusion methods have been presented in recent years, such as wavelet transform based fusion method, IHS transform based fusion method and PCA transform based fusion method. The disadvantage of IHS transform is that it can not decompose an image into different scale, so this transform can not be used to enhance image details. PCA transform based fusion method can not well maintain spectral information. Both IHS transform based fusion results and PCA transform based fusion results have serious spectral distortion. Wavelet transform can decompose image into different scale, but wavelet transform based fusion method only implement low frequency components average and high frequency components substitution, not use the redundancy and supplement relations of images contents. A contents correlation and Genetic Algorithms based remote sensing images fusion method is presented in this paper. Based on the imaging properties of Panchromatic images and multispectral images, contents correlation measurement concept is introduced. The fusion procedure is as follows: wavelet decomposition of Panchromatic and multi-spectral images, determination of redundancy and supplement relations of images contents with contents correlation method, the construction of fusion regulations to redundancy components and supplement components, fusion algorithms optimization by using Genetic Algorithms. Finally, a fused image can be obtained with inverse wavelet transform. Experiment results show that contents correlation and Genetic Algorithms based remote sensing images fusion method can provide a very good fusion image. This method is better than ordinary wavelet transform based fusion method, IHS transform based fusion method and PCA transform based fusion method.

the Hyperion sensor on EO-1 as part of the Hurricane Katrina support effort. In addition to multiple Hyperion data sets, both Ikonos and QuickBird data were also acquired over the New Orleans area. Following registration of the data sets, multiple high-spatial-resolution CRISP-generated hyperspectral data sets were created. In this paper, we present the results of this study which shows the utility of the CRISP-sharpened products to form material classification maps at four-meter resolution from spacebased hyperspectral data. These products will be compared to the equivalent products generated from the source 30-m-resolution Hyperion data. In addition, we will show the results of a change detection experiment in which two resolution-enhanced Hyperion scenes taken two days apart after the Katrina disaster were used to assess the water level changes. The CRISP approach can be used with existing sensor pairs such as the Hyperion/Ikonos and Hyperion/QuickBird pairs we used in our study. We also investigated the use of Ikonos spacebased data to sharpen hyperspectral data collected using the AVIRIS high-altitude aircraft. The method also explores the option of a hybrid multi-spectral/hyperspectral satellite, on which a compact high-spatial-resolution multi-spectral sensor employing time-delay and integration techniques could be combined with a lower-spatial-resolution hyperspectral instrument.

6749-26, Session 6
Fusion of MODIS land channels to produce Canada-wide time series of surface albedo at 250m and 10-day intervals for climate change and terrestrial monitoring applications
A. P. Trishchenko, Y. Luo, K. V. Khlopenkov, W. M. Park, Canada Ctr. for Remote Sensing (Canada) The Moderate Resolution Imaging Spectroradiometer (MODIS) onboard Terra and Aqua satellites is a unique sources of reach spectral information useful for many applications. The MODIS instrument provides observations in 36 spectral bands ranging in wavelengths from 0.4 um to 14.4 um. Two bands are imaged at 250 m resolution at nadir, five bands at 500 m, and the remaining 29 bands at 1 km. The standard MODIS data processing system covers the basic operational needs for a number of data products and applications. Implemented globally it, however, cannot always make the best use of MODIS 250m and 500m land channels required for terrestrial monitoring and climate change applications. To address the need of national users in medium resolution data of improved quality, especially in terms of spatial resolution, the Canada Centre for Remote Sensing (CCRS) with support from the Canadian Space Agency (CSA) has developed independent technology for processing MODIS imagery over Canada starting from level 1B top of the atmosphere swath data. The system includes the following steps: 1) fusion (downscaling) of MODIS 500m land channels B3-B7 with 250m bands B1-B2 to obtain consistent 250m imagery for all seven bands B1-B7; 2) re-projection of 250m bands into standard geographic projection; 3) scene identification at 250m spatial resolution to obtain mask of clearsky, cloud and cloud shadow pixels; 4) compositing clear-sky pixels over 10-day intervals; 5) atmospheric correction; 6) landcover-based bi-directional correction procedure. Details of each component listed above will be described. Comparison with standard MODIS archived products and highresolution LANDSAT imagery will be presented to demonstrate features of newly derived land products. Baseline clear-sky composites can be accessed from the NRCan GEOGRATIS archive at http://www.geogratis.ca/ geogratis/en/index.html. This work was conducted at CCRS as part of the Project J35 of the Program on Enhancing Resilience in a Changing Climate, Earth Sciences Sector of the Department of Natural Resources Canada. Project is supported by the Canadian Space Agency through the Government Related Initiatives Program (GRIP). The MODIS data were acquired through the NASA Distributed Active Archive Center (DAAC), the excellent collaboration and the assistance of the NASA personnnel are greatfully acknowledged.

6749-25, Session 6
Resolution enhancement of hyperion hyperspectral data using Ikonos multispectral data
E. M. Winter, Technical Research Associates, Inc. (USA); M. E. Winter, Univ. of Hawaii at Manoa (USA); S. G. Beaven, Space Computer Corp. (USA); A. J. Ratkowski, Air Force Research Lab. (USA) The resolution of hyperspectral sensors is typically much less than that achieved by panchromatic and multi-spectral sensors due to fundamental noise limitations. We have developed a new and innovative technique for combining a high-spatial-resolution multi-spectral image with a lower-spatial-resolution hyperspectral image. The approach, called CRISP, compares the spectral information present in the multi-spectral image to the spectral content in the hyperspectral image and derives a set of equations to approximately transform the multi-spectral image into a synthetic hyperspectral image. This synthetic hyperspectral image is then recombined with the original lowspatial-resolution hyperspectral image to produce a sharpened product. The result is a product that has the spectral properties of the hyperspectral image at a spatial resolution approaching that of the multi-spectral image. To test the accuracy of the CRISP method, we applied the method to synthetic data generated from single aircraft acquired hyperspectral images. These high-spatial-resolution images were used to generate both a lower-spatial-resolution hyperspectral data set and a four-band multi-spectral data set. With this method, it is possible to compare the output of the CRISP process to the truth data (the original scene). In all of these controlled tests, the CRISP product showed both good spectral and visual fidelity, with an RMS error less than one percent when compared to the truth image. We then applied the method to real world imagery collected by

140

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology

6749-27, Session 7
Change detection in urban scenes by fusion of SAR and hyperspectral data
D. C. J. Borghys, M. Shimoni, C. Perneel, Royal Belgian Military Academy (Belgium) Urban areas are rapidly changing all over the world. Remote sensing has been used for many years to monitor these changes. The urban scene is characterized by a very high complexity, containing objects formed from different types of man-made materials as well as natural vegetation. Hyperspectral sensors allow mapping the surface materials present in the scene and therefore allow identifying the main object classes in the scene in a relatively easy manner. However amguities persist where different types of objects are constructed of the same material. This is for instance the case for roads and rooftops. Although higher-level image processing (e.g. Spatial reasoning) might be able to relief some of these constraints, this task is far from easy. In the current paper the authors have fused the information of the hyperspectral sensor with that high-resolution polarimetric SAR data. SAR data give information about the type of scattering on the ground with in turn is related to the geometry and dielectric properties of the imaged objects in the scene. The information from SAR is therefore complementary to that which is obtained from hyperspectral data. In this paper a supervised classification based on logistic regression for the SAR and on classical methods for the hyperspectral data is applied to obtain two classification results. These are then fused into a single classification in order to map the main classes in the scene. The result is compared to a ground truth map that was constructed by combining a digital topographic map and a vectorized cadastral map of the same are. The comparison allows to identify possible changes between the map and the available image set. These candidate changes need further analysis by a human expert, but allow focusing the attention of the human photo-interpreter to regions of interest. The method was applied on a dataset consisting of hyperspectral data from the HyMAP sensor (126 channels in VIS-SWIR) and E-SAR data. The E-SAR data consist of fullpolarimetric L-band and VV,HH X-band data. The SAR data were acquired from two viewing directions. Both data sets were acquired by the Greman Aerospace Agency DLR in the frame of a Belgian National Project funded by the Federal Science Policy.

6749-29, Session 7
Geospatial intelligence about urban areas using SAR
B. van den Broek, R. J. Dekker, TNO (Netherlands) Situational awareness and accurate up-to-date intelligence are keywords in present day urban area security and military operations on urbanized terrain. For gathering up-to-date (geospatial) intelligence with respect to new and/or damaged infrastructure in cities, detection of changes using satellite data is an efficient approach. Radar satellites are important here since they can collect data at day and night and independently of weather conditions ensuring that the information can be obtained at regular intervals and in time. A drawback of SAR imagery is the poor ability to recognize the detected changes in the scene. Classification of the detected objects therefore has to be performed using other information sources such as geographical and context information, which can be derived from maps, geographical information systems and from earlier optical imagery at times when cloud cover was no problem. At TNO we have developed and implemented a change detection algorithm for efficiently co-registering imagery and extracting areas of change in radar data. For radar imagery a dedicated filtering mechanism suppresses speckle in order to reduce unwanted false alarms. The study presented here comprises a test case in the city of Baghdad on basis of multi-date change detection with Radarsat fine beam imagery with a resolution of 10 meters and optical imagery of 1 meter for monitoring infrastructural change and damage assessment. Using the tool developed at TNO we show that the detected changes from the radar data can be characterized using complementary data and other context information. Special attention will be paid to the use of high-resolution 3-dimensional data in urban areas from an IKONOS stereo pair for the classification of changes. The study was performed in the framework of the NATO/SET working group on High Resolution SAR-Systems for Geospatial Intelligence (SET102) and the EU network of excellence on Global Monitoring of Security and Stability (GMOSS).

6749-66, Poster Session


Laser-Scanner and GIS instruments to analize, validate and prevent the hydro-geologic risk: applied example in Pordenones urban area, Friuli-Venezia Giulia region, Italy
M. Maso, A. Rosina, F. Baldan, T. Busa, M. Carone, R. Costantino, N. Randelli, Univ. Iuav di Venezia (Italy) Abstract The combination of meteorological events, which are becoming more extreme, and unplanned urbanisation, which increases hydro-geological risk, requires the application of new technologies and instruments in order to forecast natural phenomena and safeguard the territory. This study suggests a methodology that uses GIS instruments and Laser-Scanner surveys for: a) detecting areas prone to flooding, b) helping public administration to create Civil Protection Plains, c) validating PAI (hydro-geologic plan), d) providing input data to territorial planning. The area of study is Pordenone, in FriuliVenezia Giulia, situated along the Noncello torrent, an affluent of Livenza river. This river has an almost constant discharge but, in the occurrence of high intensity rains, its affluents (which show a torrent-like regime) cause high variation to discharge and hydraulic criticism in the Pordenone area. A high accuracy and geometric precision DTM has been realised using data from a Laser-Scanner survey with 1x1 m grid. A static hydraulic model has been applied on two study areas, for a scenario of uniform flooding which depends only on the a.s.l.. Using the spatial analyst tool and a specific model builder of GIS software the critic area of Pordenones depurator was studied and a break in the river-bank was simulated in a residential zone. Results of this study have: a) detected buildings and streets prone to the effects of extreme events, b) evidenced the affected PRG areas, c) validated the PAI plan, d) identified other critical areas to such an event.

6749-28, Session 7
Analysis of polarimetric SAR response from man-made objects in an urban area.
D. J. Weydahl, Norwegian Defense Research Establishment (Norway) ENVISAT ASAR gives the opportunity to analyse and interpret urban features using multi-polarisation satellite SAR data. It is believed that ENVISAT Alternating Polarisation (AltPol) mode can be of importance if one chooses to use ASAR images to build up general knowledge of infrastructure in an area. This work is focusing on ASAR backscatter change from urban categories as a function of polarisation and incidence angle, as well as the ability to disciminate between a set of urban classes. Many ENVISAT ASAR AltPol scenes are acquired over the Oslo region in Norway during the summer and autumn 2006. The ASAR AltPol data sets are geocoded and calibrated. The ASAR backscatter variations are analysed with respect to incidence angle variations and polarisation combinations. Contrast and structural information from man-made objects are also investigated. By combining ASAR AltPol backscatter from HH/HV and VV/ VH, it may be possible to say something about the complexity of an object and the direction (horizontal/vertical) of its dominant components. Strong cross-polarisation backscatter will be evaluated with respect to building types. The work will also focus on which incidence angles (or incidence angle combinations) that will give the maximum information of manmade structures from an urban area when using ENVISAT ASAR data.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

141

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology

6749-67, Poster Session


Comparison of the performances of middleresolution multi-spectral images for vegetation cover rate extraction based on mixture
Y. Suga, Hiroshima Institute of Technology (Japan); T. Konishi, Computer Aided Design & Information Ctr. (Japan); S. Takeuchi, Hiroshima Institute of Technology (Japan) This study aims to establish a practical image analysis method for the use of middle-resolution multi-spectral images acquired by Landsat TM/ETM+, Terra ASTER and ALOS AVNIR-2 as the complementary data sources of higher resolution images like IKONOS for the purpose of environmental monitoring of widerange areas. For this purpose, an image analysis based on mixture analyses is tested as one of the effective approaches. As the information target, vegetation cover rate (VCR) in urban areas is selected because it is one of the important environmental factors to affect urban environment like heat island. In order to realize easy and efficient computation for estimating the mixture rate of vegetation categories, the linear mixture model using two main categories, vegetation and nonvegetation, is employed combined with the least square estimation of multi-regressive coefficients for vegetation cover rate (VCR) and non-vegetation cover rate (non-VCR) with multiple bands data. In addition, two sub-categories for both of vegetation and non-vegetation categories are introduced to specify representative pixel values as correct as possible, that is, trees and grasses for vegetation, and buildings and baresoils for non-vegetation respectively, and their optimal mixture rates are estimated as well as the mixture rate of vegetation and non-vegetation categories. For this purpose, an iterative procedure is employed, in which each mixture rate of two subcategories for vegetation and non-vegetation are varied by ten percent steps and the least square estimation is applied with all combinations of mixture rates of sub-categories for vegetation and non-vegetation. The experiments for VCR extraction were conducted with the test area of Hiroshima-city and with the multi-spectral data acquired by Landsat TM/ETM+, Terra ASTER, and ALOS AVNIR2. The accuracy for VCR extraction was evaluated based on the comparison with the VCRs obtained by pixel-wise vegetation/non-vegetation classification of an IKONOS multispectral image. The result suggests that the performances of above three sensors are related not only to the number of bands but also to their wavelength regions. At the conference, detailed experimental results will be presented.

6749-70, Poster Session


Contribution of the data thematic mapper to the fracturation of the region of Ain Tmouchent, Low Tafna: relationship between the failles and the eruptive centers
S. A. Ahmed Zine Eddine, Ctr. National des Techniques Spatiales (Algeria) In the area of An Temouchent, Landsat 5 Thematic Mapper satellite data were to us a great utility in the description of the faults, because of the complexity of the zone of study and the recent age of the grounds which makes the structural analysis particularly difficult to establish. It is true that recent tectonics of the area are not as spectacular as that which results from the preceding paroxysmal phases, but it does not remain less significant about it if we want to improve the geodynamic model of total tectonics. Thus, we will try to attach the result obtained by remote sensing to the work already carried out in the sector of study. For this, on one hand. To place this one within its regional framework and to establish a relation between the direction of the characterized faults and their relation with the eruptive centers.

6749-71, Poster Session


Analysis of key issues about the selection methods of suitable scales in monitoring land use in the Tarim River Basin, Xinjiang, China
J. Zhao, X. Chen, A. Bao, Xinjiang Institute of Ecology and Geography (China); Y. Duan, Xinjiang Management Bureau of Tarim River (China) Selection of suitable scales is one of the key issues in the studies of ecology, geography, etc. Currently, many study achievements about the parameters (such as the status, structure and pattern) in the studies of suitable scale selection have been achieved. However, the selection of optimal scales in models (process) is not paid great attention to yet although a suitable scale of the process is the foundation of scale transition. After reclaiming wastelands in the Tarim River Basin in recent 50 years, stream flow of one third river section of the Tarim River (the longest continental river in China) is cut off, the natural ecosystems are seriously degenerated, and the situation of land use is tempestuously changed. Therefore, it is urgent to resolve the problems that whether there are the change laws of scales and what kinds of scales are suitable in monitoring land use. In this study, the scale change laws of general land use and 4 main land use types including woodlands, shrubbery lands, sandlands and saline or alkaline lands in the mainstream area of the Tarim River in recent 50 years are lucubrated by interpreting the land use data in the 1950s, 1970s, 1990s and 2000 using the available maps and RS images. By regarding the areas of land use types as the scales, the parameters are selected, the histograms of the areas of patches are charted, the normalized scale variances under 9 scales (1:2000000, 1:1000000, 1:500000, 1:250000, 1:100000, 1:50000, 1:25000, 1:10000 and 1:5000) are calculated, and the features and scale change laws of land use in the Tarim River Basin are lucubrated using the landscape indexes including the Shnnon-Weaver diversity index, Simpson diversity index and fractal dimensions. The results are as follows: (1) The difference in areas of patches in the mainstream area of the Tarim River is in great disparity, the patch of sandlands is the largest and its proportion in the year of 2000 was 43.77%; (2) The frequency distribution of the areas of patches, sandlands, saline or alkaline lands, woodlands and shrubbery lands in the mainstream area of the Tarim River in all the study periods was in normal distribution, however, the positions of peak values and the distribution patterns were different, the positions of the peak values of shrubbery lands in the 1990s and those of sandlands and saline or alkaline lands in the year of 2000 were in a range of 0.1~1 km2, those of others were in a range of 1~10 km2, only the positions of the peak values of woodlands did not change and were in a range of 1~10 km2 in recent 50 years; (3) The normalized scale variance table reveals that it is the most suitable that the general scale of land use in the mainstream area of the Tarim River is at 1:50000 or in a rang of 50000~100000, the general patterns of the normalized scale variances in the Tarim River Basin in the 4 study periods were similar, the normalized scale variances were the highest in the 1990s, and then in the 1950s, 1970s and 2000; (4) In the normalized scale variances, shrubbery land was

6749-69, Poster Session


A distance variable to simulate the urban population
X. Wang, M. Ma, Cold and Arid Regions Environmental and Engineering Research Institute (China) The urban population density declines from Central Business District (CBD) to city edge. Some studies have explored several mathematical forms to describe the relationship between population density and location. But these models assume that the city is single center and the urban population decline with concentric circle format. It is difficult to satisfy this requirement in actual cities. A distance variable, city edge distance, has been developed to solve the problem in this paper. Using city edge distance can express multi-centered city situation. The irregular city is easily transformed into a virtual concentric circle city. Then the Clark model and power exponential model are transformed into new format based on the city edge distance. A case study indicates that the transformed formats could successfully simulate the urban population density based on remote sensing and Geographical Information System methodologies. But more applications are needed in different city types to validate and improve this distance variable.

142

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
the only land use type with the change of suitable scale in recent 50 years, that is it was decreased from 1:50000 to 1:25000; the optimal scale of sandlands at scale of 1:100000 was in a range of 1:50000~1:100000, the figures of normalized scale variances in the 4 study periods reveal that the change of the normalized scale variances of sandlands was tempestuous and was significantly higher than those of other land use types, however, the curve patterns of the normalized scale variances in the 4 study periods were similar; the most suitable scale of saline or alkaline lands and woodlands was at scale of 1:50000 all along; (5) Comparing with the normalized scale variances, there were no significant distribution trends of three landscape indexes, the change of curve patterns was smooth, and the maximum values of the Shannon-Weaver and Simpson indexes were all at scale of 1:50000 except those in the year of 2000, which were similar to that of normalized scale variances; the normalized scale variance and diversity index of sandlands were 1:100000 and 1:50000 respectively and at the same scale; the results of diversity indexes and normalized scale variances of saline or alkaline lands in the 1950s and 1970s were similar; those of woodlands and shrubbery lands were different all along, the normalized scale variances and diversity indexes of woodlands were at scales of 1:50000 and 1:10000 respectively, and those of shrubbery lands were at scales of 1:25000 and 1:10000 respectively. The results of the diversity indexes and the normalized scale variances were not so similar, and even the results of the two diversity indexes at most of the time were neither similar. The values of fractal dimensions were generally low and their curves were smooth, which reveal that the areas of land use types are distributed under the same structure in the regions where the land use types are not so many. 1980s has lead to a dramatic built-up area expansion in Hong Kong and Macao. This paper based on MSS data and Landsat TM images, analyzed the structural features and spatial patterns of the builtup area expansion of Hong Kong and Macao from 1973 to 2004. Besides, the driving mechanisms behind the expansion are also discussed from several aspects. The data used are one MSS data with a spatial resolution of 80~80m obtained in 1973 and four TM images with a spatial resolution of 30~30m obtained respectively in 1989,1999,2000 and 2004 for the two Special Administrative Regions. The result showed that Both of Hong Kongs and Macaos builtup area underwent large changes during the period from 1973 to 2004. The built-up area of Hongs was about 59.95 sq. km in the early 1970s, and 207.70 sq. km in 2004. The built-up increased by 2.46 times and increased averagely by 4.77 sq. km per year. While the built-up area of Macaos was about 6.45 sq. km in the early 1970s, and 22.47 sq. km in 2004. The builtup increased by 2.48 times and increased averagely by 0.52 sq. km per year. The built-up area expanded around the Hong Kong Island and Kowloon Peninsula, and the expansion rate was distinct in different time. The urban construction of Macao formed according the transport condition historically.

6749-74, Poster Session


Synthetic generation of airport familiarization using geo information systems
N. Zimmer, M. Launer, J. Schiefele, Jeppesen GmbH (Germany) Airport Qualification describes a required process which determines the landing and starting authorization on an airport. This Qualification can be accomplished using a depicted picture as familiarization or get a familiarization by an authorized person. At some airports, one is not allowed to approach without such a familiarization. The FAA identified these airports and called them Special Pilot Qualification Airports. At present, Jeppesen has the Airport Familiarization Charts in the market, in which airport photos are used that are captured on approaches. Before the approach, Pilots make use of these photos to get familiarized with the airport, the runway layout, the approach and terrain. These charts cover all qualification airports as well as all other airports supported by Jeppesen. Within the scope of a research project a first prototype for generating Airport Familiarization pictures and videos using Flight Information Data as well as Remote Sensing Data and their derived data within Geo Information Systems has been researched, developed, implemented and validated. The paper describes a new feasibility using Remote Sensing Data like gap filled and improved SRTM data, Airport Mapping Data, flight important Obstacle Data just as modelling three-dimensional airport buildings within GIS Systems to replace analogues photos with synthetic pictures and also generate new Airport Familiarization Videos. Finally, an overview of the potential of the Airport Familiarization approach for several scientific training options is presented.

6749-72, Poster Session


Landscape pattern and process of modern oasis in Xinjiang, China
J. Huang, Y. Lin, S. Dong, Xinjiang Institute of Ecology and Geography (China) Based on Landsat/MSS/TM images and CBERS-02 CCD data of Beitun Oasis in Ertix River Watershed in the years of 1972, 1989, 1999 and 2005, the landscape patterns for the past 30 years were analyzed. Using the GIS data collective platform, we calculated the landscape pattern conversion probability matrix, landscape pattern index, and contribution rates of major landscape components to characterize the impacts and responses of landscape pattern changes and landscape ecological processes. The results indicate that in this region the areas of farmlands, urban & rural residential lands and waters are increasing, the area of woodlands is decreasing, and that of grasslands is decreasing and then increasing. In the desert landscape patterns, the areas of sandlands and Gobi deserts & nude lands are decreasing after increasing, and those of saline or alkaline lands & marchlands are increasing obviously in the latter period. The features of landscape ecological process of urban & rural residential lands are concentrated in spatial pattern, but for grasslands and woodlands, those are fragmentized in spatial patterns. The landscape components convert very frequently, and the landscape pattern is not stable. Woodlands ecosystem function reducing and soil salinization and alkalization result in a negative influence on the local ecological system. It is essential to adjust the landscape patterns to rehabilitate and construct the fragile ecological system of modern oasis landscape ecosystem in arid area and use water resources reasonably, so that ecological environment and social economy is healthy and stable with sustainable development.

6749-75, Poster Session


Analysis of wetland landscape change in Sheyang Lake region
C. Ke, Nanjing Univ. (China) The information sources, including topographical maps in the different time, Landsat TM images, were selected to preliminarily investigate the wetland landscape change of Sheyang Lake Region in the past 76 years by means of GIS and landscape indices analysis methods. The results showed Sheyang Lake in 1996 reduced as 437.8 km2, which accounted for 38.1% of total area in 1920. Meanwhile, marsh area continuously dramatically decreased in the past 76 years; but the change of water body area was fluctuant. It decreased from 1920 to 1958, afterward continuously increased because of the necessity of local economy development, many fishing ponds were dug in marshes or fertile agriculture land for fast development of aquaculture. Patch quantity of water body is 16, 48, 60 and 188 from 1920 to 1996, respectively, average patch size 6.907, 2.057, 1.838 and 0.966, respectively, patch density 0.145, 0.486, 0.544 and 1.036, respectively. These landscape indices showed

6749-73, Poster Session


A comparative study of urban expansion on Hong Kong and Macao special administrative region in the past three decades
F. Mu, Institute of Remote Sensing Applications (China) Urbanization has been a prominent phenomenon in China economic development since the country adopted the reform and openness policy in 1978. Hong Kong and Macao are two Special Administrative Region (SAR) of the Peoples Republic of China. The unprecedented combination of economic and population growth since the early

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

143

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
that the patch quantity of water body became more and more, patch size smaller and smaller from 1920 to 1996. That is, water body became more and more fragmentized, higher and higher spatial heterogeneity. Patch quantity of marshes is 28, 44, 51 and 150 from 1920 to 1996, respectively, average patch size 37.469, 20.678, 14.608 and 3.535, respectively, patch density 0.027, 0.048, 0.068 and 0.283, respectively. The landscape indices of marsh showed the same results as water body. In a word, in the past 76 years human activities dominated the wetland landscape changes in Sheyang Lake Region. alternating cross polarization mode (AP mode) by the direct down-link at Hiroshima Institute of Technology (HIT) in Japan. VV/VH and HH/HV polarization modes of ASAR data were processed and analyzed to investigate temporal change of SAR backscattering coefficient of rice-planted fields during the rice growing cycle in Hiroshima district, Japan. In addition, multitemporal, multi-direction (descending and ascending) and multiangle (IS2, IS3, IS4 and IS5) ASAR/AP mode images were used in this study. ASAR data were acquired in the mode of VV/VH in 2005 and HH/HV in 2006 respectively. Ground truth data ware measured such as height of plant, vegetation cover rate (VCR) and Leaf Area Index (LAI) corresponding to ASAR observation, and also the correlation between ASAR backscattering coefficient and those parameters of rice crop growing were analyzed. The results indicated the superiority of VV/VH data compared with HH/HV data for the purpose of monitoring of rice crop growth in Japan. That is reason why HH/HV polarization backscatter of rice-planted fields tended to be fluctuated by the difference of multi-direction and multiangle. On the other hand, VV/VH polarization backscatter was more stable compared with HH/HV. In particular, backscatter of VH polarization was linearly increased during the rice growing cycle. In conclusion, rice crop growth can be successfully monitored using multi-temporal, multi-direction and multi-angle ASAR/AP mode data.

6749-76, Poster Session


Photoswitchable surfaces: a new approach to chemical sensing
A. Radu, S. Scarmagnani, R. H. Byrne, C. Slater, N. Alhashimy, D. Diamond, Dublin City Univ. (Ireland) The importance of large-scale deployments of wireless sensing networks (WSNs) is clear. Ubiquitous sensing linked to digital communications has a great impact to personal health, environmental control as well as an indicator for terrorist activity. While the deployment of WSNs for physical parameters is well established, it still presents a significant challenge if it is to link successfully into the molecular world of chemistry and biology. The so called chemical sensor paradox is maybe the limiting factor for wider application of chemo/bio sensors in WSNs. [1] Generation of chemically or biologically inspired signals inevitable lead to the slight changes of sensing surface. Sensors therefore suffer from baseline drifts and variation in sensitivity, so they must be periodically calibrated, or in the other words, the sensing surface must be removed and characteristic of sensor checked. Development of adaptive materials is a direction undertaken in our group in order to approach the solution to the exploitation of chemo/bio sensors in WSNs. We have developed a polymer material based on spiropyran - a molecule that can undergo reversible transformation from neutral, passive form into zwitterionic, active form with binding groups exposed upon illumination with light of a certain wavelength. We have demonstrated that surface immobilised films containing spiropyrans derivatives can be light-activated and used for sensing for example, heavy metals. Light-deactivation of the polymer when the requirement for sensing stops, leads to the reformation of the original passive spiropyran surface. Moreover, activation and deactivation of the sensing surface can be achieved by employing simple, low-cost, low-power light emitting diodes (LEDs). This has several implications: Sensing surface is maintained in an inactive/passive form (binding groups are hidden) as long as there is no need for sensing. Photonic activation of the surface is user-controlled in terms of when and where the activation occurs. Subsequent deactivation of the surface to the passive form in principle should extend the sensor lifetime and reduce the need for sensor calibration, thereby significantly simplifying the sensors operation. By employing LEDs and by establishing spatial control over the activated surface we envision the possibility to construct simple, low-cost, low-power chemo/bio-sensing devices with greatly prolonged life-time. [1] Byrne R., Diamond D.;Nat. Mater. 2006, 5, 421 Acknowledgements: We acknowledge support from Science Foundation Ireland through the Adaptive Information Cluster Award (SFI 03/IN.3/ 1361), and the Environmental Protection Agency through grant number EPA - 2004-RS-AIC-M4 .

6749-78, Poster Session


Landuse transformation and ecosystem health assessment from 1986 to 2005 in Zhejiang coastal zone
Z. Chen, State Oceanic Administration (China) and Zhejiang Univ. (China); Z. Mao, J. Chen, Q. Zhu, J. Zou, State Oceanic Administration (China) Because of global climatic variations and anthropogenic influence on local environment, as the active interaction belt between ocean and land, the coastal ecosystems are sensitive and some of which are under pressure. It is of very necessary to diagnose whether they are in healthy stage. The Zhejiang coastal zone situated at about 120E - 123E and 27N - 31N, the landuse have changed largely during the past 20 years. The objective of this paper is to find out how the landuse change during 1986 to 2005, and try to establish an ecosystem health assessment framework. The remote sensing data collected include Landsat TM in 1986 and 1995, ASTER in 2005, AVHRR in those years, MODIS from 2000. And there is also some auxiliary information from social and economic statistics. The landuse/landcover continuous transformation in study area can be obtained according to the classification derived from the satellite data. Landscape variation can be quantificationally calculated using the landscape indices in the software of Fragstat. The selected ecosystem health indicators consist of the aspect of social (population density, construction density), disturbance (the frequency of flood, red tide, typhoon), environmental (greenness of terrestrial plants, contamination diffusion), coastal water quality (suspended solids, Secchi disk depth, 5-day biochemical oxygen demand, total nitrogen, total phosphorus, chlorophyll-a). According to this research, we can acquire the landuse/landcover transformation from 1986 to 2005 when the Chinese economic develop rapidly. Assess whether the human habitat and environment is healthy, how pressure the coastal zone sustain. We can also seek some strategy for the human and nature to keep harmonious relationship.

6749-79, Poster Session


Ion-selective electrodes as detector systems autonomous environmental sensing devices
T. Radu, A. Radu, D. Diamond, Dublin City Univ. (Ireland) Wireless sensor networks (WSNs) will play a vital future role in areas such as environmental control, personal health monitoring, or detection of potential terrorist threats. Deployment of such networks is currently overwhelmingly restricted to transducers for detecting physical parameters such as temperature, pressure, light or movement. Unfortunately, the application of chemo/bio sensors in WSNs is limited mainly

6749-77, Poster Session


Rice crop growth monitoring using ENVISAT-1/ ASAR AP mode
T. Konishi, Computer Aided Design & Information Ctr. (Japan); Y. Suga, Hiroshima Institute of Technology (Japan); S. Omatu, Osaka Prefecture Univ. (Japan); S. Takeuchi, K. Asonuma, Hiroshima Institute of Technology (Japan) ENVISAT-1/ASAR is the most advanced space-borne SAR, its capabilities include the different incidence angles and the dual polarization acquisitions. This study focuses on the validation for monitoring of rice crop growth using ENVISAT-1/ASAR

144

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
due to the lack of small and simple chemo/bio sensors that have the required sensitivity, selectivity and lifetime, coupled with low-cost and low-power consumption [1]. Recent developments in the theory of ion-selective electrodes (ISEs) suggest that these devices are capable of detecting much lower concentrations than previously thought, and a number of groups have demonstrated sub-micromolar detection limits with these sensors in the lab. [2] Their low cost and relatively simple mode of use, coupled with inherently low power consumption makes them interesting candidates for potential integration into autonomous sensing devices that can be part of larger WSNs. In this work, we explore the possibility of using ISEs for environmental analysis. informed by recommendations based on recent advances in making ISEs with sub-micromolar detection limits. [3] In this particular case, we have studied the use of ISEs as potential detectors for an autonomous, deployable system for continuous monitoring of heavy metal contamination of soil samples from the area of Silvermines, Co Tipperary, Ireland. In this locality, years of mining have left specific hot-spots with very high levels of contamination, which pose a potential threat to the adjoining districts. In this paper, we present initial results with the ISEs and discuss their potential future use in an autonomous WSN deployment that could quantify the degree of heavy metal pollution in the area arising from dust blow-offs, and capture dynamic changes in this profile over time, thus potentially serving as an early warning system for the local community and specialists in various Government agencies. [1] Byrne R., Diamond D.;Nat. Mater. 2006, 5, 421 [2] Bakker E., Pretsch E.; Trends Anal. Chem. 2005, 24, 199 [3] Radu A., Peper S., Baker E., Diamond D.; Electroanalysis, 2007, 19, 144 Acknowledgment We acknowledge support from Science Foundation Ireland through the Adpative Information Cluster Award (SFI 03/IN.3/ 1361), and the Environmental Protection Agency through grant number EPA - 2004-RS-AIC-M4. *to whom the correspondence should be addressed e-mail: dermot.diamond\@dcu.ie this approach is applicable. Drought was not very severe in 2006. However, its spatial variation was very obvious. It seemed that Leizhou Peninsula in western Guangdong had higher level of drought severity. Temporally drought severity was comparatively higher in spring than in the summer when rain events were more frequent. Though applicable, the approach still needs many efforts to improve for a better monitoring of drought events in the province.

6749-81, Poster Session


Retrieving spatial-temporal variation of land surface temperature in Tibet Plateau of China for the years 2005-2006 from MODIS satellite data
M. Gao, H. Tang, Z. Qin, Chinese Academy of Agricultural Sciences (China); H. Pei, Nanjing Univ. (China); J. Qiu, Chinese Academy of Agricultural Sciences (China) Tibet Plateau serves as the sources of several big rivers such Yangtze River and Yellow River. Moreover, the plateau plays important role in driving atmospheric circulations in East and South Asia regions. Land surface temperature (LST) has long been recognized as a key parameter in forming the geoatmospheric characteristics of the plateau, where the cold air dominates the surface, making it a white plateau covered with snows and glaciers. In recent decades the accelerative melting of glaciers in Tibet Plateau has been highly concerned not only within China but also around the world. Monitoring the spatial and temporal dynamics of ground temperature in Tibet Plateau has been a very necessary action to investigate the changes of snow cover and glaciers. In the study we present our study on mapping land surface temperature variations for the years 20052006 in the plateau. We use the thermal infrared band data of MODIS satellite images for the mapping. Since the plateau has a very rough ground surface that is difficult to estimate the necessary parameters for the mapping, we have developed a practical approach for LST retrieval from the MODIS thermal band data. The approach was alternated from the split window algorithm proposed by Qin et al. (2001) for NOAA-AVHRR data. Detailed methods for atmospheric transmittance and ground emissivity have been presented in the paper. Results from our study indicate that ground temperature in the plateau is featured with obvious spatial and temporal variations. Generally the temperature in winter and spring is less than 0C and it is also not very high in summer, due to the high altitude. Annual vibration of the temperature is generally from -10C to +5C. Because of topological form, Chaidamu Basin of the plateau has the highest temperature in summer, usually high up to above 40C. Another feature is that ground temperature in 2006 was slightly higher than that in 2005, indicating an omen of global warming. Our study provides an alternative to understand the environmental changes in the plateau that shape significant impacts on atmospheric dynamics of East Asia and South Asia.

6749-80, Poster Session


Remote sensing of agro-droughts in Guangdong Province of China using MODIS satellite data
M. Gao, Chinese Academy of Agricultural Sciences (China); X. Zhou, H. Zhang, Guangzhou Geographic Institute (China); H. Tang, Z. Qin, Chinese Academy of Agricultural Sciences (China) Drought has long been one of the most serious natural disasters shaping important impacts on agricultural farming in Guangdong province of China, where the dominant subtropical cropping system is highly prone to be threatened by drought events. Each year about 1/3 of total cropping area in the province is under drought impact and the most severe drought event in last 60 years happened in 2004 when above 0.7 million hectares of cropping area, accounting for 68% of the total, was under severe impact. Therefore, drought monitoring has been highly concerned in the province. The government strongly requires information on dynamic changes of drought events for administrative decision-making to prepare anti-drought actions. A research project has been established for the monitoring. In the paper we intend to present our progress in development of a practical remote sensing approach for drought monitoring in the province. The basic idea is to integrate remote sensing and GIS for the agro-drought monitoring. We use MODIS satellite data to extract the required information on crop growth, surface temperature and moisture on the basis of daily dynamics. The information is integrated into the available GIS database on crop types and their spatial distribution, soil types, regional water resources, topology, and meteorological elements for drought monitoring. A quantitative model was established for assessment of drought severity on the basis of equilibrium in crop water demand and soil moisture supply. An operational monitoring system has been developed for Guangdong province, with the model as the core component. Using MODIS satellite data as the main input, we carried out the preliminary operation of the system for the year 2006. We computed the required data for assessment of drought severity in each 10-days for the year 2006. Our results indicated that

6749-82, Poster Session


Mapping vegetation cover variation of grassland ecosystem for desertification monitoring in Hulunber region of Inner Mongolia, China
Y. Zhu, Nanjing Univ. (China); H. Tang, Z. Qin, Chinese Academy of Agricultural Sciences (China); X. Lei, Nanjing Univ. (China); B. Xu, Chinese Academy of Agricultural Sciences (China) Grassland ecosystem degradation and desertification has been highly concerned in China for years because such degradation is perceived to directly relate with the occurrence of sandstorms invading into north China. In this study we intend to map the spatial-temporal variation of vegetation cover density from remote sensing data in Hulunber region, a typical grassland ecosystem with the highest biomass productivity in Inner Mongolia of China. Theoretically the degradation can be measured through observing the decline of vegetation density of grassland ecosystem. Therefore the mapping was done through an approach of combining MODIS images with Landsat TM/ETM and ASTER images. Since NDVI is a good indicator of vegetation, a practical algorithm has been developed in the study to map the spatial-temporal variation of the grassland ecosystem in recent years. Remote sensing images of the region for the period 2000-2006 have been obtained and analyzed for

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

145

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
the mapping. Results from our study indicated that the vegetation cover rate has been steadily decreasing in recent years, with relatively high spatial and temporal variation. Vegetation growth period of the ecosystem is generally from April to early October. Thus vegetation cover rate usually reaches the maximum in August in the region. Our study reveals that the rate on average has a trend of steadily decreasing in recent years. In 2000 the rate was above 60% on average, while it decreased to below 55% in 2006. Though many factors may lead to the decrease of vegetation cover rate, especially drought as a consequence of annual precipitation vibration, this decrease still implies such a fact that the grassland ecosystem is under the process of degradation and desertification. Therefore, our study shows that the change of vegetation cover rate is an applicable indicator for grassland ecosystem degradation monitoring. The approach combining MODIS data with Landsat TM and ASTER images provide an alternative approach to monitor the changes of desertification the spatial and temporal variation of snow cover in arid and semiarid environment. Inner Mongolia grassland in 2006 is slightly worse than that in 2005. Vegetation growth in Tibet remains almost unchanged in 2006 compared to that in 2005. The vegetation growth in the early period of 2006 is closed to that in 2005. However, in the late period, the status is worse in 2006 than in 2005. Our study provides useful information required for better grassland ecosystem administration.

6749-85, Poster Session


The application of GIS in the regionalization of plastic and sunlight greenhouse
F. Guan, R. Wei, Meteorological Institute of Hebei Province (China) The meteorological factors for developing the plastic and sunlight greenhouse were derived by analyzing the microclimate data of the plastic and sunlight greenhouse, and studying the meteorological condition needed by the vegetables and production reduction brought by the low temperature and spare sunlight. The percentage of days without any sunshine hours from October to the March of next year was taken as the sunlight factor, the average temperature in the coldest month as the temperature factor, and taking the altitude of the site into account also. The area suitable for building sunlight greenhouse is determined by the comprehensive methods under support of GIS technology. The result shows that the central and east part of Hebei province is suitable and south part is moderately suitable, but it is not suitable in north of the Hebei province.

6749-83, Poster Session


Land-use and land-cover changes affect local and regional climates in Piracicaba, Brazil
P. P. Coltri, S. Freitas, Instituto Nacional de Pesquisas Espaciais (Brazil); V. A. Demetrio, Univ. de So Paulo (Brazil); N. J. Ferreira, Instituto Nacional de Pesquisas Espaciais (Brazil) Land cover substitution and land use modification have been considered important on local and regional climate changes, especially in tropical regions. Piracicaba is a city in southeastern Brazil and its economic activity is based primarily on sugar cane culture. In fact, the sugar cane culture seasonality affects the land use in most cities due to changes in land use and cover. Here, we studied the seasonal variation of urban heat-islands and regional climate using remote sensing, GIS and atmospheric modeling. Urban heat-islands were analyzed with Landsat 7 (Enhanced Thematic Mapper +) data of sugar cane crop (from January to March) and of non-crop period (from August to November). Infrared thermal band images were converted to surface temperatures (Celsius degrees) and the warmer regions were identified. The results indicated that the non-crop period had average temperatures 3.5C greater than those observed during other months. Therefore, the heat-islands intensity is intimately linked to the sugar cane culture seasonality. In order to analyze land cover effects on regional climates, numerical simulations are being performed using the atmospheric model BRAMS (Brazilian RAMS). Taken together, the results suggest that land use/cover changes are related with the dynamics of sugarcane culture and markedly affect climate at local and regional levels

6749-86, Poster Session


Coupling relationship of land surface temperature, impervious surface area and normalized difference vegetation index for urban heat island using remote sensing
X. Wu, Q. Cheng, Zhejiang Gongshang Univ. (China) This paper compares the normalized difference vegetation index (NDVI) and percent impervious surface as indicators of surface urban heat island effects in Landsat imagery by investigating the relationships between the land surface temperature (LST), percent impervious surface area (%ISA), and the NDVI. Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper Plus (ETM+) data were used to estimate the LST for the Hangzhou City, China. A map of percent impervious surface with a standard error of 7.95% was generated using a normalized spectral mixture analysis of July 2003 Landsat TM imagery. The results indicate there is a strong linear relationship between LST and percent impervious surface, whereas the relationship between LST and NDVI is much less strong. This result suggests percent impervious surface provides a complementary metric to the traditionally applied NDVI for analyzing LST quantitatively over the seasons for surface urban heat island studies using thermal infrared remote sensing in an urbanized environment.

6749-84, Poster Session


Monitoring vegetation growth in the grassland of China using MODIS satellite data
B. Xu, X. Yang, Z. Qin, Chinese Academy of Agricultural Sciences (China) Vegetation is the most important factor determining economic and ecological functions of grassland. Monitoring vegetation growth can provide the essential information required for grassland ecosystem administration. In this paper we present our study on monitoring vegetation growth in the grassland of China. The aim is to compare the general growth status of vegetation in various grassland regions. A practical approach has been developed to use the daily MODIS satellite images for the monitoring. Some conclusions can be obtained from the monitoring in 2006. (1) The general status of vegetation growth remains unchanged in 2006 when compared to that in 2005. (2) Temporally the status is slightly different between 2006 and 2005. Compared with that in 2005, the growth status in 2006 can be featured with similar growth in early period and better growth in middle period and worse growth in later period. (3) In 2006, the better growth mainly distributed in Xilinguole region of Inner Mongolia and the northern region of Tibet. The worse growth can be observed in the northeastern region of Inner Mongolia and the southwestern region of Tibet. (4) Spatially the main grassland provinces have great difference in vegetation growth status. For instance, the overall growth of

6749-87, Poster Session


Urban applications of multi-typed space-borne imageries using standard mobile 3D open graphic API
K. Lee, Hansung Univ. (South Korea) In recent, platforms and devices for space-borne remote sensing are diversified to face both multi/mobile uses and various application domains. The real-time mobile 3D modeling and rendering is one of core technologies for mobile applications with geo-based spatial data sets. However, mobile platforms and devices for mobile 3D applications show some limitations to design and implement: limited CPU and memory, absence or limited performance of graphic accelerators, small size of display panel, and so on. At the industry standard level, mobile 3D graphic rendering API such as OPENGL|ES (OPENGL for Embedded System) by Khronos group stimulates and helps mobile 3D development. In this study, an integrated prototype for 3D authoring and rendering system by feature-based approach and 3D graphic and pixel pipeline with LOD-based images was designed and implemented using OPENGL|ES API: data model and design of spatial features, editing and manipulating of 3D urban objects, generating geometrically complex type features, supporting of

146

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
both database and file system, and texture mapping of spaceborne image library to 2D/3D geo-based features. This system also tested in hand-held PDA of specifications as follows: Window Mobile 2003 SE, Intel Strong ARM 312 MHz, 320 by 240 65K display, and NAND memory of 64MB and SDRAM 64 MB. As the results of this prototype implementation and experiments with actual satellite imagery, the possibilities of the mobile 3D urban applications were investigated in the consideration to 2D location-based architectures/services and with COLLADA as new open format and database schema for 3D contents/models creation. a binary classifier, which finds the maximal margin (hyperplane) between two classes. SVM can classify non-linearly separable data sets by plotting the data into a high-dimensional feature space using kernels. In comparison to other data mining techniques such as ANNs, it is easier to use, and only a few parameters need to be adjusted by the users. Precisely the study used SVM with Gaussian kernel with different bandwidth to enhance the performance of the interpretation process; the results are compared in details. The hyperspectral imagery contains 224 bands; after cross track illumination correction there are 174 representative bands remaining from 224 total bands, excluding the overlapped and not calibrated bands, so the whole remaining bands are used to classify 13 different geological units. Those units are characterized by experts in the study area .the whole geological units were well classified and recognized by the proposed SVM using different data sets with different sizes for training as well as for testing. The Learning process is performed using the SVM neural network. The network is trained using the kernel-Adatron algorithm with the Gaussian kernel. With three data sets for training the first consists of 11310 samples the Mean Square Error (MSE) was 6.800E-05 and recognition rate of 84%, the second has 22620 samples MSE was 1.600E-05 and recognition rate of 91.5%; and the third has 33930 samples MSE was 9.500E-05 and recognition rate of 79.23%.

6749-88, Poster Session


A threshold method for coastal line feature extraction from satellite imagery
L. V. I. Zoran, C. Golovanov, Univ. Politehnica Bucuresti (Romania); M. A. Zoran, National Institute of Research & Development for Optoelectronics (Romania) The coastal zone of world is under increasing stress due to development of industries, trade and commerce, tourism and resultant human population growth and migration, and deteriorating water quality.Satellite imagery is used for mapping of coastal zone ecosystems as well as to assess the extent and alteration in land cover/land use in coastal ecosystem. Beside anthropogenic activities, episodic events, such as storms, floods, induce certain changes or accelerate the process of change, so in order to conserve the coastal ecosystems and habitats there is an urgent need to define coastal line and its spatio-temporal changes. Coastlines have never been stable in terms of their long term and short term positions. Coastal line is a simple but important type of feature in remote sensed images. In remote sensing have been proposed many valid approaches for automatically identifying of this feature, of which the accuracy and speed is the most important. The aim of the paper is to develop a threshold-based morphological approach for coastline feature extraction from optical remote sensing satellite images (Landsat ETM and IKONOS) and to apply it for Romanian Black Sea coastal zone. At first coastal line is initially extracted by intensity threshold approach, afterwards, morphological operators are used to precisely process coastal region and to improve accuracy and speed . Then a morphological operator is used to fill the isolated regions, and precisely acquire the coastal line feature at last.

6749-91, Poster Session


A WebGIS system on the base of satellite data processing system for marine application
G. Fang, D. Wang, H. Huang, J. Chen, State Oceanic Administration (China) From 2002 to 2004, a satellite data processing system for marine application had been built up in State Key Laboratory of Satellite Ocean Environment Dynamics (Second Institute of Oceanography, State Oceanic Administration). The system received satellites data from TERRA, AQUA, NOAA-12/15/16/ 17/18, FY-1D and automatically generated Level3 products and Level4 products(products of single orbit and fused multi-orbits products) deriving from Level0 original data, which is controlled by an operational control sub-system. Currently, the products created by this system play an important role in the marine environment monitor, disaster monitor and researches. Now a distribution platform has been developed on this foundation, namely WebGIS system for querying and browsing of oceanic remote sensing data. This system is based upon large database system-Oracle. We made use of the space database engine of ArcSDE and other middleware to perform database operation in addition. J2EE frame was adopted as development mode, while Oracle 9.2 DBMS as database background and server. Simply using standard browsers(such as IE6.0), users can visit and browse the public service information that this system provide, including browsing for oceanic remote sensing data, and enlarge, contract, move, renew, traveling, further more data inquiry, attribution search and data download etc. The system is still under test now. Founding of such a system will become an important distribution platform of our countrys satellite oceanic environment products of special topic and category (including Sea surface temperature, Concentration of chlorophyll, and so on), for the exaltation of satellite products utilization and promoting the data share and research level of the oceanic remote sensing platform.

6749-90, Poster Session


Analysis of hyperspectral data for lithological identification using support vector machines
A. M. N. Ahmed, M. A. E. Farag, M. A. Yahia, H. H. Ramadan, M. S. Abd Elwahab, Ain Shams Univ. (Egypt) Remote sensing hyperspectral data has many applications especially in the field of earth science Utilization of this technology has shown a rapid increase in many areas of economic and scientific significance. Hyperspectral sensors capture the detailed spectral signatures that uniquely characterize a great number of diverse surface materials. Classification, clustering, and visualization of these very highdimensional signatures need untraditional methods. Different approaches for spectral image interpretation have been studied using Artificial Neural Networks (ANNs) and Support Vector Machines (SVM) to meet the challenge of high-dimensionality. The study used SVM for geological mapping of hyperspectral imagery at Abu Zenima area, western Sinai, Egypt, the hyperspectral data has been captured in 2003 by Hyperion instrument on the United States Geological survey (USGS) Earth Observing1(EO-1)satellite. Hyperion data provides continuous spectral coverage over 220 bands with a 10-nm (average) sampling interval over the contiguous reflected spectrum from 356 to 2577 nm, with a ground sample distance (GSD) of 30 meters for all bands. The Hyperion scene is collected as a narrow strip, covering a ground area of 7.7 km approximately in the across-track direction, and 42 km or 185 km in the alongtrack direction. The image file is provided in scaled at-sensor radiance values, with the data stored as 16-bit signed integers. The support vector machine (SVM) algorithm, a supervised machine learning method based on the statistical learning theory, is a new method in artificial intelligence. SVM is basically

6749-92, Poster Session


Ring structure characteristics of South Ordos Basin and its control factors in the deep
F. Yang, Shandong Univ. of Science and Technology (China) According to the synthetical analysis of remote sensing information and geophysical data, the late-reformation characteristics of the South Ordos basin in the Late Mesozoic and the Cenozoic is researched with the former research results and the field work. The result shows that the margin and inside of the basin are all reformed since late Mesozoic. Two different kinds of structure belt exist around the Ordos Basin. They are

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

147

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
compressional thrust nappe belt and extensional faulted depression belt. The ring structure which is found in the southern Ordos basin has a diameter of more than 300km, and its movement and rotation are on the assumption that the ring has dominated by the mantle plume in the deep earth and relative movement on the surface of earth. It is testified that the ring structure is a relatively independent block, and it is significant for the coexistence of multi-energy resources in the same basin. There are also differences between the south and north part of Ordos basin in large scale lifting. That is the reason why there are differences between the south and north part of Ordos basin in the distribution of mineral resources. The seismic profile and the magnetotelluric sounding data of the ring structure show that there is a high depth of the Moho boundary comparing with the north part of Ordos basin, it convinced us that the mantle plume lifting does exist in the location of ring structure in the south part of Ordos basin. The lifting movement of Ordos basin in the Late Mesozoic and the Cenozoic may be an important factor of the changing channel of Yellow River cross the Ordos basin, it is significant for the formation and evolution of Yellow River. ground objects and the extraction of biophysical of vegetation. It is an important obstruction of developing the quantification of RS technology. In this paper, sparse Tamarix spp in the Tarim River Basin were taken as a research object. Two different approaches are presented. The first is simulated the spectrum of the object with GORT model based on ground observation. In the second approach, the spectral responses of sparse crown types of vegetation and bare soil have been simulated pixel by pixel using linear spectral mixture model. Coupling the simulating spectra of object obtained from two different models with MODTRAN4 (for atmospheric spectral and directional effects) to generate top-of-atmosphere radiance images respectively. Comparison of simulated images with actual remote sensing data (EO-1,Landsat ETM,CBEAS), the spectral differences of these data are analyzed. This paper describes the applied methodology, the study area with the input images, the set-up of the actual image simulations, and discusses the final results obtained.

6749-95, Poster Session


An application of the VIC-3L hgdrological model for Lake Poyang Basin based on remote sensing
S. Huang, Meteorological Research Institute of Jiangxi Province (China) Lake Poyang is the largest lake in China and has an integrated basin in Jiangxi province, China. In order to simulate the hydrological character of large lake basin, a hydrological based three-layer variable infiltration capacity(VIC-3L) land surface model is applied to Lake Poyang basin, which has a 160 000km2. The required soil parameters are taken from the soil classification information of Global 5min data provided by the National Atmospheric and Oceanic Administration (NOAA) hydrology office. The vegetation and land cover parameters are derived from the Landsat TM data and UMD Global land cover information, and the forcing data are obtained through GIS interpolation methods based on 96 meteorological stations over the study area. All of the data required by the VIC-3L model are compiled at 1km~1km resolution for Lake Poyang basin and the daily meteorological data are available for the period of 1961-2005. However, the land cover in Lake Poyang basin is mainly mixed forest. The unrealistically large land cover are deciduous broadleaf forest from UMD data source resulted in an unreasonable evapotranspiration and stream flows, and resulted in a change of water resource in Lake Poyang basin. The study indicates that Remote Sensing plays an important role in proving the precision of model simulations.

6749-93, Poster Session


Consideration of grazing treatments for aboveground biomass estimation from MODIS data in Inner Mongolia of China
W. Tao, B. Xu, X. Yang, Z. Qin, Chinese Academy of Agricultural Sciences (China) Spatial and temporal changes of aboveground biomass is very important for grassland ecosystem management especially in Inner Mongolia steppe where over-grazing has been recognized as a one of the key factors leading to ecosystem degradation under arid and semi-arid environment. In this study we intend to develop a practical approach through consideration of grazing treatments for aboveground biomass estimation from MODIS satellite images in Inner Mongolia steppe region. The NASA/MODIS reflectance product (MOD13A2) was used to compute four vegetation indices: the normalized difference vegetation index (NDVI), the enhanced vegetation index (EVI), the modified soil-adjusted vegetation index (MSAVI), and the optimized soil-adjusted vegetation index (OSAVI). Totally 206 ground samples were obtained from typical steppe in the region. Three grazing treatments were considered for the sampling: ungrazed, moderately grazed and overgrazed. Two datasets were created from the ground samples. 156 samples were used to build the estimation models called model dataset and the rest 50 samples were for validation called validation dataset. The computed vegetation indices from MODIS reflectance products were used to correlate with the actual aboveground biomass from the ground sampling. Both linear and nonlinear regression models were established to reveal the relations between aboveground biomass and the vegetation indices. Establishment of the models was with consideration of different grazing treatments. The results showed that the aboveground biomass was nonlinearly correlated with all the four indices for the model dataset. However, EVI has the best correlation among the four. The correlation of EVI with the biomass was R2=0.80, 0.81 and 0.68 respectively for the ungrazed, moderately and overgrazed treatments. Results from validation indicated that estimation accuracy of the EVI models was about 80% on average. Therefore, the nonlinear models of EVI can be applied to biomass estimation in the study region. Our study reveals that grazing intensity has great impacts on aboveground biomass and the information about grazing intensity is very useful for accurate estimate of aboveground biomass in Inner Mongolia steppe of China.

6749-96, Poster Session


A meteorological assessment method of ecological quality based on remote sensing and GIS
S. Huang, Meteorological Research Institute of Jiangxi Province (China) considering the meteorological effect on ecological environment, Based on remote sensing, GIS and data merged technology, the paper analyses the environmental factors, such as vegetable cover degree, land degradation, water area, natural disaster and humid condition, and studies a synthesis assessment method of ecological quality. The study also makes certain a technological flow chart and method for monitoring ecological quality in operation, which can be used to monitor the changes of ecological quality. The paper gives an example for applying the monitoring and assessment method on ecological quality of Jiangxi Province.

6749-94, Poster Session


Remote sensing data of sparse crown assimilation using GORT and LSMM model in Tarim River Basin, Xinjiang, China
G. Jiapaer, X. Chen, Xinjiang Institute of Ecology and Geography (China); S. Liu, Beijing Normal Univ. (China) The sparse crown along both riversides of the Tarim River play an important role in firming sand and restraining desertification. The low percentage of coverage of the sparse vegetation result in obtained the spectrum information form remote sensing data difficultly, which affects the classification accuracy of identifying

6749-97, Poster Session


Scale effect of geological environment monitoring
Z. Zhou, Hohai Univ. (China) Geological environment means the summation of every geological body and geological actions which influences the existence and development of mankind, including regional geological environment, urban geological environment, mine geological environment, geological environment of Geologic vestiges protect zone and so on. The content of geological environment is rich, and the factors affecting the evolution of

148

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
geological environment are complex. Exacerbation of environment brings different scales geological hazards and harmful effect to the existence and development of mankind. In general geological environment monitoring the monitoring target and domain is determined according to specific geological hazards problem, but in fact some others factors with more bigger extent which is ignored determine the geological environment evolution. In this paper the main content and method of geological environment monitoring is described, and the scale effect of geological environment monitoring is researched, then the importance and necessity of analysis the geological environment evolution using monitoring data of different scales geological environment is proved. medium, on the antenna frequencies and on the acquisition data parameters (i.e. stacking or number of traces/sec). Usually, this parameter is related to the path of the antenna at different investigation depths. In this work, the horizontal resolution is determined experimentally and compared with different models considering the path. In these studies three GPR systems antennas were used with different centre frequencies: 900 MHz, 1 GHz and 1.6 GHz. The experimental procedure consists in to place different elements into a homogeneous media and to detect them with the GPR system. The used media were water and sand. The elements were metallic bars with different diameter and PVC tubes with different diameters and filled with: water, air, sand or salt. The bars or the tubs were placed at different distances and at different depths in the medium. The resolution value was considered as the distance between elements that allows the antenna to detect them separately. This result was compared with the path of the antenna, calculated using the wave velocity and the depth to the anomalous bodies. Coupling effects with the surface reflection and the direct wave were also analysed.

6749-98, Poster Session


Environmental monitoring and data sharing
Z. Zhou, Hohai Univ. (China) Environment means the situation, influence and force in surrounding place. The evolution of environment follows the natural law, but it also influenced by actions of mankind. The environment evolution is the representation of inner interaction of hydrosphere, biosphere, atmosphere and lithosphere. The general environment monitoring is limited in some specific profession field or some specific terrain tract, so the monitoring data is difficulty to generate organic relations and brings effect on the comprehensive correct cognition to natural laws and scientific problems of mankind. Only erect global environment monitoring net and data sharing system, we can really understand the environment evolution law. In this paper the earth is as an object, and based on inner interaction of hydrosphere, biosphere, atmosphere and lithosphere, the content and scale of environment monitoring is analyzed and the importance of erecting global environment monitoring data system is expounded. Keywords: Environment monitoring, data sharing

6749-101, Poster Session


Morphotectonics in the broader area of Athens using Aster epipolar imagery
M. Foumelis, I. Fountoulis, Univ. of Athens (Greece); I. Parcharidis, Harokopio Univ. of Athens (Greece) With the advent of instruments that produce high quality stereo images, remote sensing has evolved into an important tool to ground observations and particularly to the study of terrain feature related to the topography. Remote sensing stereo image pairs can be used to derive 3D terrain information similar to conventional photogrammetry. The present study is focusing on the extration of morphotectonic features using ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) Nadir-Back loking Band 3 data, free of clouds, on the basis of automated digital stereo-correlation image matching technique. The ground instantaneous field of view for ASTER data is 15 m over 60 km of swath offering a unique potential for morphological interpretation, overcoming difficulties due to area accessibility. Anaglyph 3D representation of the generated ASTER epipolar images by means of stereo glasses is shown to be an excellent tool for delineating morphotectonic structures. Additionally, the ability of optical interpretation from various viewing angles allows the recognition of structures of different orientation. The study area selected is situated in mountain Parnitha, at the NW of Athens building shell, ranging up to approximately 1400 m above sea level. Unexpected seismic activity in region during September 1999 underlines the need for re-examination of the area in terms of morphotectonic structures. Finally, ground-truthing was obtained from ralated fieldwork and various scale geological and topographic maps. Concluding remarks on the method applied and its complementarity with conventional mapping techniques are also drawn.

6749-99, Poster Session


Characteristics of the GPR field pattern antennas
V. Prez-Gracia, R. Gonzlez-Drigo, D. Di Capua, L. G. Pujades, Univ. Politecnica de Catalua (Spain) Ground-Penetrating Radar has become a popular nondestructive and non-invasive tool in different kind of applications: civil engineering, archaeology, concrete and masonry analysis, etc. The selection of the antenna frequencies depends on the application, but each antenna has a radiation pattern and some characteristics that have influence in the final interpretation and in the model obtained for the studied medium. The knowledge of these features and its coupling effects with the medium could improve the results of the GPR prospecting studies. In this work, some experimental procedures were carried out in order to obtain the 1.6 GHz centre frequency antenna characteristics in the air and in one material medium and to compare them. First, the study of the attenuation due to geometrical spreading was performed. This result was compared with the amplitude attenuation in a material medium, deduced from the GPR experimental data. Second, the shape of the radiation pattern was estimated in laboratory for different distances between the target and the antenna. Near field and far field were considered during the experimental data acquisition. Third, the relative amplitude of the reflected wave (in dB) was obtained depending on the relative position of the antenna over the target. The shape of the radiation pattern and the relative amplitudes obtained in the air were compared with those obtained in a slow medium (water). This slow medium was characterized with the wave velocity and the attenuation factor of the GPR signal.

6749-102, Poster Session


Hybrid control and data acquisition system for geographically sensors for environmental monitoring
F. Garufi, Univ. degli Studi di Napoli Federico II (Italy); A. Boiano, Istituto Nazionale di Fisica Nucleare (Italy); F. Acernese, Univ. degli Studi di Salerno (Italy); R. De Rosa, Univ. degli Studi di Napoli Federico II (Italy); R. Romano, F. Barone, Univ. degli Studi di Salerno (Italy) In this paper we describe the architecture and the performances of a hybrid modular acquisition and control system prototype developed by our group for the implementation of geographically distributed monitoring and control systems. The system, an improvement of a VME-UDP/IP based system developed by our group for interferometric detectors of gravitational waves, is based on a dual-channel 18-bit low noise ADC and 16-bit DAC module at 1 MHz, managed by an ALTERA FPGA, that can be used standalone or mounted as mezzanine (also in parallel with other modules) on a motherboard. Both the modules and the motherboard can send/receive the configuration and the acquired/correction data for control through a standard EPP parallel port to an external PC, where

6749-100, Poster Session


Experimental analysis of the resolution is shallow GPR survey
V. Prez-Gracia, R. Gonzlez-Drigo, D. Di Capua, L. G. Pujades, Univ. Politecnica de Catalua (Spain) Horizontal resolution indicates the capability of the system to detect separately two close features in the medium. This parameter depends on the wave velocity travelling trough the

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

149

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
the real-time computation is performed. Experimental tests have demonstrated that this architecture allows the implementation of distributed control systems, using a standard laptop PC for the real-time computation, with delay time of less than 25us on a single channel, that is a sustained sampling frequency of more than 40 kHz. Each module is also equipped with a 20-bit slower ADC necessary for the acquisition of an external calibration signal. The system is now under extensive test in two different experiments, that are the remote control and data acquisition of a set of geophysical sensors like seismometers, velocimeters and accelerometers for geophysics with different transmission standards and the remote control of the end mirrors of a suspended Michelson Interferometer through electrostatic actuators for Interferometric Detectors of Gravitational Waves.

6749-105, Poster Session


An approach for GIS-based statistical landslide susceptibility zonation: a case study in the northern part of El Salvador
V. Kopackova, J. Sebesta, Czech Geological Survey (Czech Republic) The study area represents the northern part of El Salvador and similar, as the whole country, the region is particularly prone to natural hazards caused by seismic, volcanic and tectonic activities (endodemic processes) and associated exodynamic phenomena such are landslide hazards, rock failures, floods and so on. The study was carried out with standard remote sensing and GIS tools for the processing of satellite imagery and the construction of digital elevation models (DEMs). The analysis was based on digital data sets including various derivatives of digital elevation models (DEMs) as well as Landsat image data, seismic database, geological and morphological maps. All the relevant thematic layers representing various factors (slope, aspect, relative relief, lithology, buffer zones along thrusts, faults and lineaments, drainage density, land cover and so on) were relatively weighted and classified due to its disposition to cause landslides. Principle Component Analyses (PCA) was used as a statistical method that allowed to decorrelate the individual hazard triggers. As result, a landslide susceptibility map integrating morphological, geological and seismic phenomena was computed. Delineated hazard zones were validated with landslides mapped in the field (USGS 2001, CGS - 2005, 2006) and both, the model and terrain mapping, showed a good agreement.

6749-103, Poster Session


Laser interferometric sensor for seismic wave velocity measurement
F. Acernese, Univ. degli Studi di Salerno (Italy); R. De Rosa, F. Garufi, Univ. degli Studi di Napoli Federico II (Italy); R. Romano, F. Barone, Univ. degli Studi di Salerno (Italy) Laser interferometry is one of the most sensitive methods for small displacement measurement for scientific and industrial applications, whose wide diffusion in very different fields is due not only to the high sensitivity and reliability of laser interferometric techniques, but also to the availability of not expensive optical components and high quality low-cost laser sources. Interferometric techniques have been already successfully applied also to the design and implementation of very sensitive sensors for geophysical applications, like monolithic accelerometers. In this paper we describe the architecture and the expected theoretical performances of a laser interferometric velocimeter for seismic waves measurement. We analyze and discuss the experimental performances of the interferometric system, comparing the experimental results with the theoretical predictions and with the performances of a state-of the art commercial accelerometer. The results obtained are very encouraging, although few problems, mainly related to the sensitivity and stability of the interferometric system, have to be better studied.

6749-106, Poster Session


Monitoring water coastline thermal anomalies by remotely-sensed data
S. Sampieri, A. Allegrini, D. Guglietta, E. Mercuri, P. P. M. Merola, Consiglio Nazionale delle Ricerche (Italy) The InfraRed (IR) electromagnetic spectrum band is particularly useful to measure water surface temperature, because in this spectrum region water radiance has a characteristic minimum value. We focused our attention on the monitoring of local sea thermal anomalies using an airborne hyperspectral sensor MIVIS (Multispectral Infrared and Visible Imaging Spectrometer), which has both spectral and spatial high resolution. MIVIS sensor is a modular instrument composed of 4 spectrometers, which simultaneously measure the electromagnetic radiation coming from the Earths surface by recording 102 spectral bands, i.e. in the Visible (0.43 - 0.83 m), the Near-Infrared (1.15 - 1.55 m), the Mid-Infrared (2.0 - 2.5 m), and the Thermal-Infrared (8.2 - 12.7 m). The different types of thermal anomalies, represented by hot and cold inputs in water (plumes), can be identified in the Thermal IR range: hot inputs can be related to industrial discharges and polluted river outpours while cold inputs can be represent river mouths or fresh water outlet. By processing the localized plumes, it is possible to classify and, consequently, interpret their extension, shape, and temperature gradient. In the present work, we analysed sea surface temperature and monitored coastal water plumes alongshore Southern Italy.

6749-104, Poster Session


Average coherence image-derived observations over an urban area: the case of Athens
I. Parcharidis, Harokopio Univ. of Athens (Greece); M. Foumelis, Univ. of Athens (Greece); P. Kourkouli, Harokopio Univ. of Athens (Greece) Interferometric coherence is a measure of the phase properties of SAR images indicating displacement and/or changes of the scattering elements within the scene. Absolute values of the interferometric coherence vary between 0 - 1 and can be seen as a measure for the accuracy in the estimation of the interferometric phase. The stability of the phase signal is lost when the physical (land cover) and/or geometrical properties (satellites viewing geometry) of the ground changes. In the present study coherence results obtained using 20 Cband ERS SAR Single Look Complex (SLC) VV-polarization images acquired in descending mode over the metropolitan area of Athens covering the period 1992 - 1999 are presented. A straightforward approach using a single master SAR image on which the other images are mapped was adopted ensuring perfect registration of the interferometric results. After generating single coherence images, with temporal separation varing between 138 and 1335 days, an averaging procedure followed leading to the average coherence image. Proper positioning and visualization of the SAR average coherence image into the Google Earth environment was performed by generating a relative .kml file. The final geocoding of the average coherence image has been improved using common features in the coherence image and a very highresolution QuickBird image. Overlay of coherence product on the QuickBird image inside Google Earth environment allow correlating the level of coherence with characteristics and properties of the urban shell. As urban areas are considered of high coherence, observations of this type permit to investigate and evaluate their phase stability in details.

6749-108, Poster Session


Change detection analysis of multi-temporal images in Naples area
D. Guglietta, P. P. M. Merola, S. Sampieri, A. Allegrini, Consiglio Nazionale delle Ricerche (Italy) Naples city is located about 200 km south of Rome, in Campania region. The city, founded in VII century b.C., is characterized by an endless city planning continuity and so it is difficult to trace its regular urban development. It is possible to discovery the

150

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
different urban settlements: the primitive Greek nucleus; the Greek-Roman center; the medieval city; the Swabian and then Aragonese city and the city of the XIX and XX century which extends until the boundaries of Campi Flegrei. Today the city is one of the most important ports of the Mediterranean area and its hinterland is densely inhabited. Monitoring the urban changes the researchers of IIA-CNR used multispectral images by MIVIS sensor. MIVIS (Multispectral Infrared and Visible Imaging Spectrometer) airborne hyperspectral sensors is a modular instrument composed of 4 spectrometers, which simultaneously collect radiation coming from the Earths surface in 102 bands (Visible, Near-InfraRed, Mid-InfraRed and Thermal-InfraRed. The MIVIS hyperspectral data, covering the Naples city were collected during different airborne campaigns carried out. Our goal is to investigate, to detect and map the urban dynamic evolutions to quantify change extent, distribution and trajectories of change. To highlight the urban changes identified in the MIVIS images, caused to natural and anthropogenic factors at various temporal and spatial scales, different hyperspectral data processing procedures were used: - Computing of Vegetation Index; - Principal Components Analysis (PCA); - Supervised classification; - Thermal Capacity (CT). (USA) The contrast of features in emissivity spectra retrieved from thermal-infrared images for rough surfaces retrieved from thermal-infrared images is lower than that for smooth surfaces of the same material. This is due to the interactions between neighboring facets due to multiple reflections, often called the cavity effect. It is in principle possible to compensate spectra for this effect provided the roughness of a surface can be characterized adequately. Fractal characterizations of geologic surface suggest that shape is independent of scale. However, for real surfaces, coarse- and fine-scale topography typically has a different fractal dimension, because the basis function for surface elements is different: for example, hills or sand dunes have fundamentally different shapes and slopes than boulders or sand grains. Thus, we might expect retrieved emissivities to differ as a function of pixel size and the characteristics of subpixel roughness to be more complex than simply estimating root-mean-square of surface height. We investigate this problem using 5-m airborne MASTER thermal-infrared images of Death Valley, California, and low-resolution images (10-500 m pixels) derived by aggregating MASTER pixels to plot the change in emissivity contrast as a function of pixel resolution. We use existing 10-m DEMs and high-resolution (~3-30 mm) LiDAR DEMs to measure power spectra for the imaged surfaces over a wide range of scales, and to calculate thermal radiosity in order to predict loss of emissivity contrast for these surfaces. Contrast loss due to roughness must be accounted for in field validation of thermal infrared images.

6749-109, Poster Session


Thematic maps production for emergency management purposes: the ITHACA project
F. Perez, A. Ajmar, F. Disabato, Politecnico di Torino (Italy) The ITHACA (Information Technology for Humanitarian Assistance, Cooperation and Action) project has the main goal to carry on operational and research activities in the field of geomatics for analysis, assessment and mitigation of natural and manmade hazards. The project has been developed within an association founded by Politecnico di Torino and SiTI (Istituto Superiore sui Sistemi Territoriali per lInnovazione) in cooperation with WFP (World Food Programme), the largest UN operational Agency and some private and public organisms. Thematic utility maps production play a key role in the emergency management field. ITHACA is therefore developing different projects dealing with small scale map production for the early impact phase. In this paper are described the following applications: - rapid mapping for flooded areas detection. An example dealing with a heavy flood that affected a huge area of the Mozambique (Zambezi River basin) at the beginning of the 2007 is showed. Radar SAR images, acquired before and after the catastrophic event, are used for flooded areas extraction using change detection techniques. Orthoprojected Landsat 7-ETM+ data are used as a reference map for the geometric correction operations as well as to supply control data. High resolution satellite data are also used to estimate the damages; - production of snow maps directly obtained using free-ofcharge data (Terra/MODIS), easily available for a worldwide extent. Snow mapping is required to verify accessibility to those areas interested by catastrophic events. An innovative procedure based on satellite data aimed at the daily monitoring of snow covers is shown. Operative procedure is completed by specific processing modules dedicated to data geometric correction, mosaicking, masking and vectorization of snow cover polygons. Results dealing with a test performed over the Himalayan area, with particular reference to the Hindu Kush (a mountain range in Afghanistan and Northern Areas of Pakistan) are showed.

6749-111, Poster Session


On the use of data fusion tecniques for improving satellite QuickBird capabablity for archaeological prospection
R. Lasaponara, A. Lanorte, R. Coluzzi, Istituto di Metodologie per lAnalisi Ambientale (Italy); N. Masini, Consiglio Nazionale delle Ricerche (Italy) The recent availability of Very High Resolution (VHR) satellite imagery, such as IKONOS (1999) and QuickBird (2001), may be able to open new perspectives in the field of archaeological remote sensing. In particular, QuickBird offers panchromatic and multispectral imagery with the highest spatial resolution currently available within the satellite sensors. It has panchromatic and multispectral sensors with spatial resolutions (SR) of 61-72cm and 2.44-2.88m, respectively, depending upon the off-nadir viewing angle (0-25 degrees). The high spatial resolution and spectral capability can make the VHR satellite images a valuable data source for archaeological investigation ranging from synoptic view (i.e. identification of high probability locations of ancient buried sites) to small details (i.e. single subsurface building). Nevertheless, the satellite-based detection of archaeological marks faces several challenges, particularly in the case of buried remains. In fact, the presence of underlying structures produces weak signals, that can be easily covered by noise. The data fusion can fruitfully enable the integration of the high spatial resolution of the panchromatic image with the spectral capability of multispectral images. Thus, allowing to achieve improved capabilities that are not possible using solely the individual datasets. In this paper, the most common data fusion algorithms were tested in order to evaluate their performance for the detection of archaeological features. The evaluation was performed by using satellite QuickBird imagery acquired for some study cases located in the South of Italy.

6749-112, Poster Session


Accurate pose and location estimation of digital camera using urban building natural features
W. Xie, Old Dominion Univ. (USA) This paper proposes an improved calibration method based on multi-views vanishing point to estimate the accurate pose and location parameters of camera in building scene. This method put the lens distortion and orientation parameters of camera directly into the calibration mode. Furthermore, using

6749-110, Poster Session


Roughness effects on thermal-infrared emissivities estimated from remotely-sensed images
A. Mushkin, I. Danilina, A. R. Gillespie, Univ. of Washington (USA); L. K. Balick, M. F. McCabe, Los Alamos National Lab.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

151

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
the line feature existing in the building surface, geometric constraints are introduced into the calibrating model. The orientation parameters of camera can be estimated accurately by this method. information extracted by the Hyperion and MIVIS surveys by using a Residual Error index (RE). The plot of the RE values vs. the corresponding mean Euclidean Distance (ED) values obtained for the same areas by classifying the MIVIS data, allowed to determine a threshold value that identify the areas on which succesfully apply the unmixing procedure . This outcomes that thresholding the ED-RE regression line it is possible to derive, from Hyperion satellite hyperspectral data, land cover information comparable (in term of abundance of classes), within an known error, to the ones obtained by an airborne survey of MIVIS sensor.

6749-113, Poster Session


Camera calibration with long focal length using building natural features
W. Xie, G. Zhou, Old Dominion Univ. (USA) In recent years, many researches are about building 3D object model in the fields of computer vision and photogrammetry, and camera calibration becomes the key problem. For long focal length digital camera calibration, theres its particularity. In this paper, aiming at long focal length lens, a camera calibration method based on point-line combination with vanishing points is proposed. This approach overcame the demerit of the conventional calibration theory with vanishing points; therefore the precision of calibration parameters became better.

6749-117, Poster Session


Virtual Huanghe River system: framework and technology
H. Lu, Institute of Geographical Sciences and Natural Resources Research (China); G. Liu, Henan Univ. (China) Virtual Reality provides a new approach for geographical research. In this paper, a framework of the Virtual Huanghe River System was first presented from the view of technology, which included five main modules data sources, 3D simulation terrain database, 3D simulation model database, 3D simulation implementation and application system. Then the key technologies of constructing Virtual Huanghe River System were discussed in detail: 1) OpenGL technology, the 3D graphics developing instrument, was employed in Virtual Huanghe River System to realize the function of dynamic real-time navigation. 2) MO and OpenGL technologies were used to make the mutual response between 3D scene and 2D electronic map available, which made use of the advantages of both 3D scene and 2D electronic map, with the macroscopic view, integrality and conciseness of 2D electronic map combined with the locality, reality and visualization of 3D scene. Also at the same time the disadvantages of abstract and ambiguity of 2D electronic map and the direction losing of virtual navigation in 3D scene were overcame.

6749-115, Poster Session


Indexes for assessing the spectral range more sensitive to the detection of buried structures
R. M. Cavalli, L. Fusilli, Istituto sullInquinamento Atmosferico (Italy); A. Palombo, Istituto di Metodologie per lAnalisi Ambientale (Italy); S. Pascucci, F. Santini, Istituto sullInquinamento Atmosferico (Italy) Buried man-made structures, like archaeological handiworks, altering the natural trend of the superficial soil, can yield tonal anomalies on remotely sensed images. These anomalies can differ in size and/or intensity according to either the environmental conditions at the time of acquisition or the spectral and spatial characteristics of the images. The research challenge is to identify the best wavelength to detect these anomalies. In this paper we have introduced two new parameters for identifying and assessing the potential of anomaly detection applied to a wide range of images: the Detection Index (DI), which counts the pixels related to the marks, and the Separation Index (SI), which relates to the difference in brightness of the marks with respect to the background. They have been tested on airborne hyperspectral data acquired on remains not yet excavated of a few archaeological sites. Results have shown that such indexes are an efficient, flexible and quick tool for assessing the image potential to detect buried structures. Moreover, when they are applied to hyperspectral data, they have allowed for identifying a spectral range more sensitive to the detection of the buried structures and more suitable remote sensing sensor.

6749-118, Poster Session


Strategies of remote sensing monitoring of changes in NATURA 2000 sites: a practical assessment in coastal mountains of NW Iberian Peninsula
R. A. Diaz Varela, P. Ramil Rego, S. Calvo Iglesias, Univ. de Santiago de Compostela (Spain) Monitoring of landscape and vegetation dynamics needs costeffective methods for the analysis and management of large multitemporal datasets. Medium resolution satellite imagery temporal series of programs like Landsat or Spot, offers attractive possibilities for automatic temporal change detection in broad areas. In the present work we aimed at the identification of land cover changes, particularly those involving environmental impacts, during the period 1991-2003 in a Natura 2000 site in the Northern Mountains of Galicia (NW Iberian Peninsula). We targeted changes related to new industrial and intensive agricultural activities that affect natural and semi natural valuable ecosystems as well as dynamics of traditional agricultural systems. We tested different methods, involving the generation of change images from PCA, selective PCA and NDVI differencing on multitemporal compositions of Landsat TM images. The effects of different image radiometric corrections on some of the methods were also assessed. Object oriented classification was used for the classification of continuous change images in change/no change thematic categories. Although the general agreement in results, we founded differences depending on the change detection method and radiometric correction. Major differences occurred when different methods of change image generation were compared while the influence of radiometric correction got only importance in the case of uneven haze effect removal on NDVI based methods. The use of PCA on high dimensionality Landsat Tm bands composition allowed the removal of atmospheric effects not related to effective land cover changes.

6749-116, Poster Session


Retrieving vegetation cover types in the complex natural ecosystem of the Pollino National Park (south Italy) through hyperion data
L. Fusilli, C. Bassani, S. Pascucci, Istituto sullInquinamento Atmosferico (Italy); S. Pignatti, Istituto di Metodologie per lAnalisi Ambientale (Italy) The effectiveness of Hyperion hyperspectral spaceborne sensor in mapping the land covers in patchy areas was evaluated by comparing its products with the thematic information content derived by processing hyperspectral airborne data (MIVIS sensor). The efficiency of the Hyperion data in discriminating the standard hierarchical nomenclature levels (CLC2000) in use in the European countries were exploited on a sector of the Pollino National Park (Italy) characterized by a fragmentized landscape. Four Standard supervised classifiers have been compared in terms of algorithm performance and class level accuracy by applying statistical metric. Moreover a sub-pixel analysis was performed on selected area of Hyperion data with uneven ground cover by selecting the endmembers on the base of territory knowledge offered by the most accuracy MIVIS classification. Accuracy of the mixture modelling was evaluated with respect to the thematic

152

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology

6749-119, Poster Session


Gas concentration detection based on Kalman filtering theory
Z. Bi, F. Fan, Zhejiang Univ. (China) As a monitoring technique with spectroscopy, the differential optical spectroscopy technique (DOAS) utilizes the narrow molecular absorption spectral characteristics of various gases to retrieve their concentrations. Although widely used in practical atmosphere gas detection application, however, due to its rough manipulation in filtering ambient and system noise, DOAS can not work well in situations where serious noise present. To further improve the performance of the traditional DOAS technique, a numerical method-Kalman Filter is introduced as the solution for the first time in this paper. KF (Kalman Filter), a discrete-data linear filtering algorithm, which can well estimate the influence of noise through fast and stable divergence, is applied to overcome the flaw of static least-squares method. Considering all the possible disturbing factors, the evolution of gas concentration with time is modeled as the condition formula, and the measurement formula is also derived from the detection principles of DOAS. An expression of the condition space for gas concentration detection system is therefore built. Then we analyze system uncertainty and the properties of both process and measurement noise. Finally, a recursive condition computation is applied based on the adaptive estimation of noise and the optimum gas concentration values could be attained. A DOAS experimental system is set up in lab for verification of our method and data under various ambient conditions are collected. As these data shows, in relatively low SNR situations, KF demonstrates remarkable superiority than the traditional DOAS in performance, which reveals its potentials to enhance the stability of DOAS under unknown conditions.

6749-120, Poster Session


Study on simulating spatial distribution and varying patterns of population in Urumqi, China
X. Chen, Xinjiang Institute of Ecology and Geography (China) 1 Introduction Population, resources, energy, food and environment are the five major issues the world facing today, and the population is the key issue. Spatial distribution of the population refers to population distribution in various districts at a point of time. Traditionally, population data comes from the countrys population census on the basis of administrative districts. In practical application, it suffers from various problems. (1) Low spatial resolution (county, district, etc.) Population data on the basis of districts displays homogenous distribution in a census unit, which masks spatial heterogeneity of population distribution and therefore it cannot provide accurate non-district data for policy makers and urban planners.(2) Census unit boundaries often do not coincide with others factors, which makes data integration problematic.(3) The traditional population density data of administrative districts can only provide limited and discrete data samples and be unable to provide reliable data for related population studies. And the spatialization of census data is an effective means to resolve above problems. In previous study, due to the scarcity of census data and related spatial data, there are a few studies on modeling of urban population spatial distribution in China. With the aids of GIS (ArcGIS and GIS programming), land-parcel and statistical population data, this paper aim to develop a quick, effective and easily applied method simulating urban population spatial distribution. Then, it produces Urumqis 100m x 100m raster population surface based on the above method. Finally, it makes a quantitative study of population spatial distribution patterns of Urumqi. 2 The Population Spatial Distribution Patterns of Urumqi Urumqis population spatial distribution mainly presents Ttype. The population density declines from south to north. Such population distribution pattern is closely related to the terrain. Urumqi is located in Urumqi valley and its downstream flood plain. Hills and mountains on both sides of Urumqi city limited the whole urban growth. The population resides in the flood plain from south to north. The southern suburban and Wulabo

area is the water conservation region so that can not be fully developed. There are hills and mountains in the east and west of Urumqi and most of the land can not be used for construction. Based on the generated population surface, I further conduct a neighborhood function in GIS to calculate mean population density in a circle with 800 meter radius centered at each cells centroid to uncover agglomeration and dispersion effects of population distribution within the district . Form the analysis of buffer zone of urban road network; we can see that about 45% of urban population resides in the residential parcel within 100 meters on both sides of the road, which indicates that traffic roads have played an important role in the distribution of urban population. In order to further analyze the variability of population density within different distance around CBD, we choose the six clockwise directions (A-F) from CBD. Then, we extract 100 population density values from all directions and generate a map which shows the relationship between distance and population density variability.Obviously, Urumqis population density spatial distribution does not confirm to Clarks urban population density spatial distribution model completely (Urban population density tends to exponential decay with the increasing distance from the urban center.). The center of the city (irregular pentagon region including Xihou street, Donghuan road, Renmin road, Hongqi road, Wenyi road and Qingnian road as center) does not form high population density. This is mainly due to this region is an important regional business center with senior services and relatively small quantity of residential parcel. 3 Conclusion This paper use urban land use and statistical population data through the GIS to propose a new approach simulating spatial distribution of urban population. Then, the population spatial distribution of Urumqi as a case is simulated by the above approach and we get more accurate and natural 100m~100m raster population surface. Concerning the population spatial distribution pattern of Urumqi, the population density declines from south to north and the population distribution mainly presents T-type, the population distribution presents multicentre agglomeration and the population distribution of the districts shows different features. Form the analysis of buffer zone of urban road network; we can see that about 45% of urban population resides in the residential parcel within 100 meters on both sides of the road. The population density varies significantly with the increase of the distance from CBD. Finally, this paper finds that the development history of districts, terrain and traffic road are main factors influencing Urumqis population spatial distribution.

6749-121, Poster Session


Application of CIRILLO, a new atmospheric correction tool on Castel Porziano Beach (CPB)
M. Musacchio, S. Amici, M. Silvestri, Istituto Nazionale di Geofisica e Vulcanologia (Italy); S. Teggi, Univ. degli Studi di Modena e Reggio Emilia (Italy); M. F. Buongiorno, Istituto Nazionale di Geofisica (Italy); M. A. Sgavetti, S. Silenzi, S. Devoti, Univ. degli Studi di Parma (Italy) Hyper/Multispectral data provides information about characteristic of natural and antropic surfaces. In order to retrieve the mineralogical species composing the Castel Porziano Beach (CPB), remote sensed data needs to be atmospherically corrected. In this work a new tool for the atmospheric correction for spaceborne EO data, based on MODTRAN and 6S codes, and developed on IDL/ENVI platform will be proposed and tested using NASA HYPERION and ASTER data. In this paper the capability to identify mineral association composing the sand of the CPB emerged beach, using hyperspectral data is shown. In order to define the mineralogical composition of the collected sample, SEM EMPA (Scanning Electron Microscopy and Electron MicroProbe Analyser) and optical polarizing microscopy analysis have been done. Results have been compared with 300 measurements performed directly on the CPB sand and 300 measurement acquired in the laboratory, both using an ASD-Fieldspec This methodological approach illustrate the high potential of hyper/multispectral remote sensing for the classification of the beaches and for the evaluation of beach nourishment

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

153

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology

6749-122, Poster Session


The application of the covariance matrix statistical method for removing atmospheric effects intended for environmental applications
D. G. Hadjimitsis, Cyprus Institute of Technology (Cyprus); C. R. Clayton, Univ. of Southampton (United Kingdom) The Covariance Matrix Method (CMM) uses the statistical relationship between all the selected bands of a satellite sensor simultaneously, rather than one at a time as in the regression method. It examines the set of variances and covariance between all band pairs in the image data and CMM provides an average pixel correction for a specified part of a satellite image. It is necessary to know a priori a value for the atmospheric path radiance on one spectral band. From this, CMM enables the estimation of the atmospheric path radiances in all the other bands. Dark pixels must be present in the CMM technique. Indeed, the authors suggest an algorithm for choosing the suitable dark objects in the scene prior to the inclusion on the CMM atmospheric correction approach. This methodology has been presented as an improved revised version of the primary CMM atmospheric approach. The authors provide a critical assessment of the suitability of the CMM atmospheric correction for a series of case studies using both Landsat TM and SPOT image data.

6749-123, Session 5:
Multi-temporal satellite observation of Pirotan Island (Marine National Park, Gulf of Kachchh, Jamnagar, India) monitoring using remote sensing
S. Chavan, GEER Foundation (India); S. Sharma, Space Applications Centre (ISRO) (India) Coral Reefs are the closest known counterparts to the tropical rainforests in terms of species richness and biological productivity in the ocean. The combination of local stresses from over fishing, physical destruction, coastal pollution and sedimentation, together with the growing threat from climate change, may result in permanent degradation of the coral reef ecosystem at a planetary scale. In fact, coral reefs may be the first major biological system to respond to human impacts at this scale. In the face of such rapid, large-scale change, there is an urgent need to improve and integrated observations capabilities for coral reefs and related coastal ecosystems such as seagrass beds and mangroves, and to integrate satellite based and in situ observing programs in support of management action. The aim of this study is to monitor coral reefs and associated ecosystems (mangroves, seagrass beds) by integrating remote sensing data and field measurements. Multi temporal coral reef study is required for monitoring the climate and human impact changes. Image processing was undertaken on IRS-LISS III sensor images to produce the ecogeomorphological maps. This process utilized the supervised classification using maximum likelihood method. These maps are intended to serve as baseline information for the baseline information for future GIS study. The next step involved the accuracy assessment of the maps using ground verification points. This study also involved the cluster analysis was used for classify habitat type. The present study involved the use of Indian remote Sensing satellite data for monitoring the Pirotan Island. The Pirotan reef is in major threats due to over fishing, high sedimentation rate, human influences, mud over reef etc. Different zones are classified using the satellite data like algal ridge, mud over reef, sandy muddy reef flat, live coral zone, seagrass, algae on live corals, algae on muddy reef flat are distinguished very clearly using the satellite data). Study of benthic features with higher precision, extensive ground survey for training area is required and only a limited number of areas can be investigated.

Bridges are a core element of infrastructure in industrial countries, as well as for newly industrializing and developing countries. Concerning natural hazard detection and disaster management it is very important to have as much as possible information about the actual situation concerning these infrastructure elements. Here Synthetic Aperture Radar (SAR) sensors come into play. The main advantage of SAR is the availability of data under nearly all weather conditions. Information can be gathered on demand and exploited to extract the needed information. However, due to the side looking imaging geometry, SAR images are difficult to interpret and there is a need for support of human interpreters by automatic image analysis algorithms. State of the art airborne SAR sensors provide spatial resolutions in the order well below a half meter. In such data many features of urban objects like bridges can be identified and used for recognition. In case of bridges over water, the before mentioned side looking imaging geometry can lead to special characteristics in the SAR image. If the imaging geometry, bridge material and construction match certain prerequisites, a bridge can appear as bright stripes in the SAR-magnitude image. These stripes can be segmented and some features like height and width can be derived. In case of spaceborne systems like RADARSAT, the coarse resolution allows this investigation only for very large bridges. But modern sensors achieve much finer resolutions. Actual spaceborne systems like TerraSAR-X will deliver images with metric resolution and allow this feature extraction also for smaller objects. In this paper the special appearance of bridges over water in SAR data is discussed. Geometric constraints concerning the appearance of the three bright stripes are investigated and simulated by an adapted ray tracing approach. The aim of these investigations is to predict if these stripes can be expected for a given imaging constellation (e.g. aspect, depression angle, material) and a certain bridge. So in case of natural hazard detection and disaster management it can be predicted if the actual image acquisition parameters allow a detection of possible changes. Information about the significant features of a certain bridge has to be derived before and stored in a database. This can also be supported by automatic approaches. So additionally to the investigations about geometric constraints, an approach is presented to detect bridges over water and to derive features like width and height from spaceborne and airborne SAR magnitude images. A part of the data was delivered in the framework of the NATO/SET working group on High Resolution SAR-Systems for Geospatial Intelligence (SET102).

6749-32, Session 8
Evaluation of a hyperspectral scanner allowing for deterioration status assessment of asbestos-cement roofing sheets
S. Pascucci, C. Bassani, L. Fusilli, Istituto sullInquinamento Atmosferico (Italy); A. Palombo, Istituto di Metodologie per lAnalisi Ambientale (Italy) Aim of this study is the identification of the hyperspectral scanner operational characteristics allowing for asbestos cement (AC) roofing sheets deterioration status assessment that is related to the asbestos fibers abundance. At this purpose we made laboratory measurements on AC samples with different deterioration status collected in two industrial areas in Italy. The asbestos occurrence in the AC samples was recognized using XRD and FTIR instruments and the abundance of surfacing asbestos fibers was performed by using a high resolution scanner. The samples optical characteristics and the brightness gradients, that can affect the AC samples, were identified using a portable field spectrometer (ASD). The results of the ASD measurements (i.e. band-depth ratio of the continuum removed calculated for the asbestos diagnostic band at 2.32m considering the different illumination or viewing geometry) were related to the relative percentage of surfacing asbestos fibers (i.e. the AC deterioration status). Since laboratory measurements confirmed that optical measurements are sensitive to variations in asbestos fiber abundance, detection limit analysis was used for defining the requirements (signal-to-noise ratio, band FWHM, and sampling range) of an optimal hyperspectral sensor most suitable for

6749-31, Session 8
Detection and feature extraction of bridges in airborne and spaceborne SAR image data
K. Schulz, A. Thiele, E. Cadario, U. Thoennessen, H. Gross, H. Hammer, Forschungsgesellschaft fr Angewandte Naturwissenschaften e.V. (Germany); U. Soergel, Leibniz Univ. Hannover (Germany)

154

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
detecting the diagnostic asbestos absorption feature in the SWIR range.

6749-35, Session 9
Accessing climate model input parameters from various remote sensing sources for the city of Beijing
M. S. Mller, Univ. Salzburg (Austria) and Univ. Bonn (Germany) Climate models are used for the prediction of regional climate variations. As a part of the Urban Environmental Monitoring Program (UEM) at Arizona State University together with the University of Beijing a climate model for the Beijing area should be created. This specific model should then be used for the calculation of climate values, the distribution of climate effects with a special focus on the Olympic Games 2008. Remote sensing imagery can be used for the acquisition of data about the actual land use land cover type. In this study we could obtain a recent ASTER scene acquired in July 2006 and classified the imagery into several urban and sub-urban classes. Due to the lack of a standardized classification schema valid for China, we adopted the U.S. national land cover data schema (NLCD). The classification was performed based on an object based image analysis (obia) approach and lead to acceptable results (80% classification accuracy). Second task in the project was the development of a method for the estimation of urban roughness parameters. The urban roughness is an important factor influencing the wind wave pattern, wind speed and finally the urban climate. Urban roughness is strongly correlated with building heights and the variations in building heights. And the various building heights are finally a function of homogeneity over a specific area. Texture is an equivalent to that value. Based on a panchromatic SPOT image we could define some textural measurements matching specific building roughness classes as input parameters for the urban climate model.

6749-33, Session 8
The construction of as-built digital urbanscape: from building components to a city
N. Shih, National Taiwan Univ. of Science and Technology (Taiwan) The purpose of this research is to build digital urban landscape (or urbanscape), based on as-built environmental information. A long-range 3D laser scanner is used to record buildings, plants, and open spaces in static configuration, plus the records of pedestrians, vehicles, objects in dynamic form. As the scan tolerance can be controlled at 4 mm / 50 m, the final urban information is made of precise object description with colors and textures included for infrastructure monitoring. The urbanscape is contributed by static and dynamic scenes of environments. Scan data are used to analyze shared design principles. A very special study was made to compare two plafonds (a ceiling wood frame in well shape) designed and engineered by the same person based on as-built geometries. A historical district was also scanned for its temple, residential houses, and public buildings. Regional cultural and commercial characteristics help defining the old street appearance, the historical identity of different periods, and the urban development clues. As a result, studies have extended the record of 3D geometries into following information related issues, such as digital contents definition, land registration information integration, urban elements integration, characteristics abstraction, and scan noise made by pedestrians or vehicles for circulation system evaluation. In order to manage and display large amount of as-built data, the information system is made by 3D database, model web pages with measuring function, as well as model editing works for different data types and service possibilities.

6749-36, Session 9
The effect of land-cover change on the urban heat island phenomenon: Greater Cairo case study
S. El Ghazali, Information Technology Institute (Egypt); A. Shakweer, Information and Decision Support Ctr. (Egypt); A. A. Darwish, Information Technology Institute (Egypt) On warm summer days with clam winds, the temperature in the city is usually higher than surrounding areas. This phenomenon is called Urban Heat Island (UHI) and it occurs mainly due to the presence of fewer trees and other natural vegetation to shade buildings, block solar radiation and cool the air through evapotranspiration in urban areas. This phenomenon increases: discomfort for local residents, the demand for energy needed to cool homes and buildings and the amount of ground level ozone (smog). Therefore it is clear that not only does UHI has a negative impact on local residents and the environment but also it has an economic cost. With a total area of more than 251 square kilometers (most of which is built up) and a population of more than 11 million inhabitants (as of 1996), Greater Cairo Metropolitan Area (GCMA) is deemed the largest metropolitan area in Africa and the Middle East. The continuous and rapid change in its urban fabric and the associated informal development has resulted in the severe damage of GCMAs environment. Among those effects is the increase in the UHI phenomenon. Although, many studies have used remotely sensed imagery to record the UHI in towns and cities of the developed world, similar studies on the developing world are rare. The main objective of this paper is to test the effect of the change in GCMAs land-cover on the UHI phenomenon. For this, two Landsat images (acquired in 1989 and 1999) are classified to produce two land-cover maps. Also, the thermal band, in each image, is used to produce the Land Surface Temperature (LST) map at the time of acquisition. The LSTs maps are used to compare the change in LST over a period of ten years and then this change is analyzed and linked to the correspondent land-cover maps. Finally, the UHI in GCMA are compared to results obtained from similar results that were carried out in the developed world. Initial results indicate that there is an obvious increase in the LST and consequently the UHI phenomenon is more obvious.

6749-34, Session 9
Monitoring informal settlements from high resolution satellite data using an object-based approach
T. Blaschke, I. Georg, Paris-Lodron-Univ. Salzburg (Austria); P. Hofmann, Leibniz Univ. Hannover (Germany) Informal settlements are phenomena of developing countries and emerging nations. They stand for slums, favelas or shanty towns. The UNHCR expects the number of inhabitants worldwide to grow from one Billion in 2005 to 1.5 Billion in 2020. The demand to monitor and to control these developments is increasing. We report on comparative studies in South Africa and Brazil using an object based image analysis (OBIA) approach. Depending on the terrain and the structure of the informal settlements the resulting classification accuracies are very different. The differences caused by using different sensors (Ikonos and Quickbird) are nearly neglectable. Conversely, the differences in classification accuracy between relatively simply structured townships in South Africa and extremely complicated hilly favelas in Rio de Janeiro are huge. These favelas are difficult to classify since building material is often similar to regular settlements, and housing structures seen from a nadir perspective are not much different, too. Variations in microstructure, shape, and pattern complicate classifications. OBIA provides solutions if auxiliary GIS data sets are available. Additionally, when classifying image objects instead of pixels, complex object-features can be developed, e.g. structure of super-objects in which the objects to be classified are embedded. This way we develop texture measures based upon the shape of sub-elements of an object, which are more suitable to describe such complex phenomena. These descriptions are fused to a generic segregation indicator which is then used to improve the initial image segmentation through an iterative knowledge based segmentation.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

155

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
From the carried out analysis it is clear that the urbanization in parts of the GCMA had raised the LST more than in areas that had witnessed less or no urbanization. Finally the analysis has clearly shown that the UHI is more obvious in GCMA than in other cities. intelligence, Geographic Information Systems (GIS), and various scientific disciplines face an increasing need to extract feature information from imagery and to coordinate that information with other geospatial data. The process of locating and extracting features such as roads, buildings, and rivers is often complex and time-consuming. ITT Visual Information Solutions has developed an object-based feature extraction tool for ENVI, the leading remote sensing software solution for obtaining information from imagery. The ENVI Spatial Feature Extraction Module is a semi-automated, workflow-based tool to extract both linear and area-based features from high resolution panchromatic and multispectral imagery. This tool is built with a users workflow in mind, leading the user from the segmentation and vectorization of imagery, through supervised and rule-based classification methods. The vector-based output can then be easily coordinated with other geospatial information. The module also allows users to extract similar features across multiple large images, and to then share those same rules with colleagues, saving valuable time by eliminating the need to manually digitize individual features. The ENVI Spatial Feature Extraction module uses advanced, fast, segmentation algorithms to extract a variety of features. By adjusting just the scale level parameter, users can get features at multiple scales and resolutions. The variety of features users may extract include roads, building rooftops, vehicles, water bodies, forests, farmlands and fields, and other land use/land cover applications.

6749-37, Session 9
Analysis of urban heat island effect by remote sensing satellite data
M. A. Zoran, National Institute of Research & Development for Optoelectronics (Romania) Remote sensing is a key application in global-change science, being very useful for urban climatology ,especially in the context of rapid increasing of urbanization in most parts of the world. Urbanization, the conversion of other types of land to uses associated with growth of populations and economy has a great impact on both micro-climate as well as macro-climate.By covering with buildings, roads and other impervious surfaces, urban areas generally have a higher solar radiation absorption, and a greater thermal capacity and conductivity, so that heat is stored during the day and released by night. Therefore, urban areas tend to experience a relatively higher temperature compared with the surrounding rural areas. This thermal difference, in conjunction with waste heat released from urban houses, transportation and industry, contribute to the development of urban heat island (UHI).Multi-spectral and multitemporal satellite imagery provide the most reliable technique of monitoring of different urban structures regarding the net radiation and heat fluxes associated with urbanization at the regional scale. Investigation of radiation properties, energy balance and heat fluxes for Bucharest urban area , Romania is based on satellite data from various sensors Landsat TM, ETM, AVHRR, MODIS. So called effect of urban heat island must be considered mostly for summer periods conditions. The results of this research have a number of applications in urban climatology, climate change, and urban vegetation studies.

6749-40, Session 10
Change detection for updates of vector database through region-based classification of VHR satellite data
A. P. Carleer, E. Wolff, Univ. Libre de Bruxelles (Belgium) Up-to-date knowledge of land cover is an important tool for the various planning authorities with responsibilities for the management of territory. The geospatial objects are changing over time and the land cover information (vector geospatial database) has to be up-date periodically. Until now, interpretation of aerial photographs is a standard tool for monitoring land cover change where fine spatial resolutions are required and this task is expensive and timeconsuming. Though, from a spaceborne perspective, the VHR satellite data are, since 1999, capable to meet the mapping and monitoring needs of municipal and regional planning agencies. Indeed, these data from the sensors Ikonos, QuickBird, OrbView-3, and in near future, the Pliades-HR French sensors, have spatial resolution lower than 5 m in multispectral mode and lower than 1 m in panchromatic mode. These new sources of data combine the advantages of satellite data with the very high spatial resolution. In spite of these advantages, the use of VHR satellite data involves some problems in traditional per-pixel classification often used in change detection techniques. There are still two occurring classification problems that can strongly deteriorate the result of a per-pixel classification of the VHR satellite data: spectral variability and poor spectral resolution. A solution to overcome these problems is the region-based classification that can be integrated in the common change detection techniques. The segmentation, before classification, produces regions which are more homogeneous in themselves than with nearby regions and represent discrete objects or areas in the image. Each image region then becomes a unit analysis and makes it possible to avoid much of the structural clutter. Image segmentation provides a logical transition from the units of pixels to larger units in maps more relevant to detect the changes in these. In this context, this research project suggests to use region based classification of VHR satellite data in different change detection processes for updates of vector database. The region based classification can be integrated in common change detection techniques that require classification. Moreover, the integration of the region based classification has other advantages; it allows to measure and use a number of features,

6749-38, Session 10
Atmospheric data access for the geospatial user community
J. van de Vegte, W. J. Som de Cerff, Koninklijk Nederlands Meteorologisch Instituut (Netherlands); R. M. van Hees, SRON Nationaal Instituut voor Ruimteonderzoek (Netherlands); M. E. Schaepman-Strub, M. Hoogerwerf, Wageningen Univ. (Netherlands); B. Domenico, Univ. Corp. for Atmospheric Research (USA); S. Nativi, Consiglio Nazionale delle Ricerche (Italy); O. Wilhelmi, Univ. Corp. for Atmospheric Research (USA) The atmospheric and geospatial communities are still separate worlds with their own tools and data formats. It is extremely difficult to easily share data among scientists representing these communities without performing some cumbersome conversions. ADAGUC aims to reduce the need for scientists to invent their own converter tools. Selected space borne atmospheric datasets will be made accessible to a GIS system in order to be submitted to data comparison, resampling, selection, manipulation and visualization. The user community will be intensively involved in the project to obtain a high fitness for use. The first ADAGUC workshop (Oct 2006) was attended by a large group of users from both the atmospheric and GIS community and resulted in a better understanding of user needs that are currently translated into specifications for the user requirements document of ADAGUC. The deliverables of this project are: Open Source conversion tools, selected atmospheric datasets in a GIS-friendly format and a web service to demonstrate the usability of the above to the geospatial and atmospheric community.

6749-39, Session 10
Obtaining information from imagery using object-based feature extraction
C. Darnel, ITT Visual Information Solutions (USA) Imagery dependent communities such as defense, security,

156

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
on top of spectral features. These features can be the surface, the perimeter, the compactness (area/perimeter(c)), the degree and kind of texture and the context. The segmentation is one of the only methods that ensure to measure the morphological features (surface, perimeter, shape...) which may be especially useful when very high spatial resolution data are available. The use of these additional features could allow to compensate for the poor spectral resolution of VHR satellite data and to increase the classification accuracy for spectrally heterogeneous classes. providing the possibility of vegetation-mapping at a high level of specificity. However, because imaging spectrometer data availability/coverage is currently limited, a need exists for effective exploitation of multispectral satellite imagery with broad-area spatial coverage. This study considers a Cerrado region under pressure from agricultural activities. The objective was to assess the utility of hyperspectral and multispectral satellite imagery in discriminating among Cerrado subtypes and agricultural subclasses. Implicit in this analysis was the consideration of the optimal feature set configuration. Hyperspectral Hyperion and multispectral China-Brazil Earth Resources Satellite (CBERS) data were acquired over part of the Brazilian Cerrado, with respective sensor overpasses ~17 minutes apart. Temporally coincident field transect data for Cerrado physiognomies and agricultural sites were collected, which included various biophysical metrics. Nonmetric multidimensional scaling and hierarchical cluster analysis were used to identify potential environmental gradients of biophysical groupings. Four Cerrado subclasses were identified: Campo Limpo (Open Cerrado Grassland), Campo Sujo (Shrub Savanna), Cerrado Tpico (Wooded Cerrado), and Cerrado Denso (Cerrado Woodland). Image classification was based upon autonomous image endmember-extraction, as well as unsupervised clustering. Results suggest that both Hyperion and CBERS data were sufficient in distinguishing between and Cerrado and agricultural land cover, but the Hyperion-derived classification was more accurate for more specific Cerrado subclasses.

6749-41, Session 10
Individual object delineation revising cadastral boundaries by means of VHSR data
S. S. Lang, D. Tiede, D. Hoelbling, Univ. Salzburg (Austria); M. Moeller, Austrian Academy of Sciences (Austria) Per-parcel classification approaches are two-fold in terms of their usability. On the one hand, e.g. in the case of crop monitoring on agricultural fields, a per-parcel approach facilitates the classification process by providing pre-defined, ready-to-use boundaries. On the other hand, pre-existing boundaries may not fully reflect the required ones. Thus it may be necessary to either remove redundant boundaries within homogenous land cover types, or add relevant boundaries which are otherwise missing. These challenges of the per-parcel approach become obvious in the case of using fine-scaled cadastre information. In this case, only a limited set of boundaries may be stable throughout the classification process, some will be dissolved, new ones will be introduced. In this paper we discuss an adaptive per-parcel approach of object generation based on multispectral satellite data but respecting given limits from digital cadastre data. Using objectbased image analysis (OBIA) we semi-automatically delineated spectrally homogenous and reasonably compact objects, incorporating the given outlines. We differentiate between three cases: (1) the parcel corresponds to a homogenous image object; (2) a set of parcels needs to be merged because of homogeneity, yet respecting a specific maximum size and compactness threshold; (3) a single parcel is spectrally heterogeneous, and therefore needs to be split and new boundaries have to be generated by object-specific segmentation. Whenever possible, object boundaries should coincide with the original cadastral borders, so only if cadastral limits do indicate a change in category, the respective boundary is retained. Otherwise, if changing categories are not reflected by the cadastral data (e.g. different forest types) a new object border will be generated. The study was carried out in an 850 sqkm sized administrative district in the Stuttgart Region, Germany. In this area, as in many others of the federal state of Baden-Wurttemberg, property had been constantly split and divided while inherited over centuries. We used orthorectified, mosaicked SPOT 5 multispectral data (5m GSD), co-registered to an existing orthophotomosaic (0.25m GSD). Orthorectification was done using a 5m DEM and a parameterised orbital pushbroom model implemented in the Leica Photogrammetry Suite. The digital cadastre information dated from 2005 and partly from 2004. We based our work on Cognition Network Language (CNL) a modular programming language in the Definiens Software environment, which allows addressing single objects and their specific behaviour. The results have met the demand of the given task, even there are limitations due the different resolutions introduced by combing SPOT and cadastral data. The shift in scales was accommodated in such a way that very small objects were dismissed and a minimum size of generated objects was considered.

6749-44, Session 11
Model and measurements of linear mixing in thermal IR
L. K. Balick, M. F. McCabe, W. B. Clodius, C. A. Jeffery, J. Theiler, Los Alamos National Lab. (USA); A. R. Gillespie, A. Mushkin, I. Danilina, Univ. of Washington (USA) Hyperspectral thermal IR remote sensing is an effective tool for the detection and identification of gas plumes and solid materials. Virtually all remotely sensed thermal IR pixels are mixtures of different materials and temperatures. As sensors improve and hyperspectral thermal IR remote sensing becomes more quantitative, the concept of homogeneous pixels becomes inadequate. The contributions of the constituents to the pixel spectral ground leaving radiance are weighted by their spectral emissivities and their temperature, or more correctly, temperature distributions, because real pixels are rarely thermally homogeneous. Plancks Law defines a relationship between temperature and radiance that is strongly wavelength dependent, even for blackbodies. Spectral ground leaving radiance (GLR) from mixed pixels is temperature and wavelength dependent and the relationship between observed radiance spectra from mixed pixels and library emissivity spectra of mixtures of pure materials is indirect. A simple model of linear mixing of subpixel radiance as a function of material type, the temperature distribution of each material and the abundance of the material within a pixel is presented. The model indicates that, qualitatively and given normal environmental temperature variability, spectral features remain observable in mixtures as long as the material occupies more than roughly 10% of the pixel. This conclusion is undergoing more quantitative testing. Field measurements of known targets serve as a reality check on the model. Target spectral GLR from mixtures as a function of temperature distribution and abundance within the pixel at day and night are presented and compare well with model output.

6749-45, Session 11
Unimixing and mapping of alteration zones using EO-1 hyperion data in Erongo, Namibia
M. M. Oskouei, W. Busch, Technische Univ. Clausthal (Germany) The study aims to detect and map alteration halos in Erongo Namibia using Hyperion dataset. Detail surveys and investigation have been possible considering hyperspectral sensors capability to render plenty of spectral information from observing surface of earth. In the term of mineral detection there are particular challenges. Main problem is due to very small size of mineral grain comparing to even those data possessing finest ground resolution.

6749-43, Session 11
Hyperion and CBERS satellite image classification intercomparison for Cerrado and agricultural mapping
A. M. Filippi, C. Brannstrom, D. M. Cairns, D. Kim, Texas A&M Univ. (USA) The Cerrado is a savanna ecoregion with grassland and woodland subtypes covering ~one-quarter of Brazil, and it is considered to be a global biodiversity hotspot under imminent threat from land-use conversion. Hyperspectral remote sensing offers a means of spatio-temporal monitoring the Cerrado, while

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

157

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
One of the methods has been invented for objects smaller than ground resolution of dataset is linear mixture model (LMM). This allows us to estimate abundances of targets in each pixel of scene after determination of endmembers. Regarding to the challenge about mineral detection task, as a matter of fact, finding pure pixels of the data in mineralogical scale is often impossible. Therefore we are able to determine the purest pixels which are a mixture of several minerals themselves and spectral profile of them consists of absorption features of those detectable minerals. During in the investigation, after atmospheric correction by ACORN and data quality assessment, the endmembers inside the scene were distinguished and then closest minerals by means of spectral feature fitting (SFF) were determined, and then concerning of some additional information like genesis of geological structures and possible alterations each of those endmembers named by one of alteration kinds. The hierarchical algorithm developed in ENVI has been used for the determination of endmembers. This algorithm includes a dimensional reduction process by minimum noise fraction method (MNF), pixel purity index (PPI) indicator, and NDimensional visual interpretation. After that, alteration abundances and consequently their map was produced by means of matched filtering (MF). Multiple regression is a common technique used when performing digital analysis on images to obtain continuous, quantitative, variables (as temperature, biomass, etc). In this scenario pixels are treated as samples from which several independent variables are known; when remote sensing images are available, the different spectral bands offered by a given sensor are often used as independent variables. The dependent variable is also a quantitative variable, such as a forest inventory variable [1] or a climate variable (e.g. temperature) [2], [3], [4]. This paper presents a methodological proposal to evaluate the implications of JPEG2000 [5] lossy compression when applied to these regression processes. Compression can be applied over the original images (so affecting the fitting process) or only over the images used to apply the model obtained from original data, without compression. The Digital Climatic Atlas of the Iberian Peninsula [6] obtains different climatic continuous quantitative variables (annual and monthly average temperature, precipitation, etc.) from multivariable regression so it is a perfect test bed for our research interests. Original images were compressed using compression ratios from 50% up to 0.1% of the original file size. Results have revealed that, even at the maximum compression ratios, practically the same accuracy, measured with independent reference climatic stations, is obtained, so JPEG2000 seems an interesting technique not heavily affecting these common modeling approaches. Bibliography [1]. Baulies X, Pons X (1995) Approach to forestry inventory and mapping by means of multi-spectral airborne data, International Journal of Remote Sensing, 16:61-80. [2]. Guijarro J.A. (1986) Contribucin a la Bioclimatologa de Baleares. Tesis doctoral. Palma de Mallorca. [3]. Snchez Palomares O., Snchez Serrano F., Carretero Carrero M.P. (1999) Modelos y cartografa de estimaciones climticas termopluviomtricas para la Espaa peninsular, Ministerio de Agricultura, Pesca y Alimentacin. Instituto Nacional de Investigacin y Tecnologa Agraria y Alimentaria, Madrid. 192 pp. [4]. Ninyerola M., Pons X., Roure J.M. (2000) A methodological approach of climatological modelling of air temperature and precipitation through GIS techniques, International Journal of Climatology, 20:1823-1841. [5]. ISO/IEC. JPEG2000 image coding system. Part 1: Code coding system. Technical report, International Standard Organization / International Electrotechnical Commission, 2000. [6]. Ninyerola et al. (2005) Atlas Climtico Digital de la Pennsula ibrica. Metodologa y aplicaciones en bioclimatologa y geobotnica, Pub. Univ. Autnoma de Barcelona. Bellaterra. 44 p. ISBN: 932860-8-7

6749-46, Session 11
An integration of ASTER and Landsat imagery for remote sensing based geological interpretation, NW Iran
K. Martnek, Charles Univ. in Prague (Czech Republic); V. Kopackova, V. Metelka, Czech Geological Survey (Czech Republic) Geological mapping project in NW Iran is focused mainly on mapping of general geology, economic geology, geohazards, and environmental features in the scale 1 : 25 000. Study area is located in Eastern Azarbayjan Province in NW Iran and covers 600 km2. Part of the study area is unsatisfactorily covered by topographic maps - only maps in the scale 1 : 50 000 are available. Therefore airborne images and ASTER DEM were used for extraction of detailed topographical and geological information. Structural indices were mapped from SRTM3, ASTER and Landsat 7 ETM+ data combining visual interpretation and computer based automatic feature extraction method. Lithological indices were mapped from Landsat and ASTER imagery using combination of histogram manipulation and spectral transformations (principal component analysis PCA, intensity-hue-saturation). Indices of hydrothermal alteration zones were delineated using feature oriented PCA, band ratios, image slicing and spectral mapping using comparison with spectral libraries. The results of various approaches were implemented in a GIS environment, which is further used as a basis for field geological mapping.

6749-49, Session 12
Algorithm development for land surface temperature measurement from GOES-R satellite
Y. Yu, D. Tarpley, R. Mundakkara Kovilakom, H. Xu, J. Privette, National Oceanic and Atmospheric Administration (USA) The Geostationary Operational Environmental Satellite (GOES) program is developing a new generation sensor, the Advanced Baseline Imager (ABI), on board the GEOS-R satellite to be lunched in approximately in 2012. Comparing to the current GOES imager, ABI will have significant advantages to measure land surface temperature as well as to provide qualitative and quantitative data for a wide range of applications. In specification, spatial resolution of the ABI sensor is 2 km, and the infrared window noise equivalent temperature is 0.1 K, which are very close to the polar-orbiting meteorology satellite sensors such as AVHRR. Most important, ABI senses the cover region every five minutes, that not only providing more cloud-free measurements but also making the daily temperature variation analysis possible. In this study we developed a split window algorithm for the LST measurement from the ABI sensor. We generated the ABI sensor data using MODTRAN radiative transfer model and NOAA88 atmospheric profiles and run regression analysis for the LST algorithm development. The

6749-47, Session 12
Removing long-term errors from AVHRR-based brightness temperature (BT)
M. Z. Rahman, L. M. Roytman, City College/CUNY (USA); F. N. Kogan, National Oceanic and Atmospheric Administration (USA) Empirical distribution functions were applied for removing longterm errors from BT data derived from AVHRR sensor on NOAA environmental satellites. This paper investigates BT stability in the NOAA/NESDIS Global Vegetation Index (GVI) data set during 1982-2003. This period includes five NOAA satellites. Degradation of BT over time for each satellite was estimated for geographical location in China. The method of matching empirical distribution functions improves the time relative stability of BT data for all satellites, especially NOAA-9, -11 and -14.

6749-48, Session 12
Implications of JPEG2000 lossy compression on multiple regression modeling
A. Zabala, X. Pons, F. Aul-Llins, J. Serra-Sagrist, Univ. Autnoma de Barcelona (Spain)

158

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
algorithm is developed from optimizing existing SW LST algorithms published in various literatures and adding a path length correction term to minimize the retrieval errors due to difference atmospheric path absorption from nadir view to the edge-of-scan. The algorithm coefficients are stratified for dry and moist atmospheric conditions, as well as for the daytime and nighttime. The algorithm sensitivity to land surface emissivity uncertainty is analyzed to ensure the algorithm performance.

6749-54, Session 13
Simulation of SAR images for evaluation purposes
R. H. Speck, H. Suess, DLR Standort Oberpfaffenhofen (Germany) Well-suited simulation tools for the generation of SAR images with spaceborne systems are necessary for image quality analysis and evaluation. Ground-based experimental tests in many cases are physically impossible and very costly. For an all-embracing test of the SAR system capabilities the simulated data should reflect the real SAR mapping capabilities to the highest possible degree. This paper describes a simulation tool for the generation of SAR data from realistic scenario. The precise mathematical modeling of the complete SAR-system chain is described by the following modules: orbit, target and background, radar hardware, image formation and image quality evaluation. Illustrative simulation examples based on real scenarios will be presented. The simulation results demonstrate the degradation the image quality by orbital and motion errors, insufficient signalto-noise ratio and pointing errors.

6749-50, Session 12
3D geo-positioning assessment and accuracy improvement of QuickBird across-track stereo imagery
G. Qiao, W. Wang, Tongji Univ. (China) DigitalGlobes QuickBird satellite, which serves as the highestresolution commercial satellite in orbit, can provide stereo imagery with accurate three-dimensional (3D) mapping products. Based on Rational Function Model, this paper investigated the 3D geo-positioning accuracy of QuickBird across-track stereo imagery. Two QuickBird basic images were collected in Feb and May 2004 by DigitalGlobe in Shanghai area, making a pair of across-track stereo imagery. The study area was very planar within 3 meters in relief. A total number of 169 points collected by GPS were used for the study, of which 16 served as Ground Control Points (GCP), others Tie Points (TP) and Check Points (CKP). First, 3D geo-positioning results were computed by the Rational Polynomial Coefficient (RPC) provided by QuickBird without GCP available. The results showed that the accuracy was about 24 meters in plane and 15 meters in height direction, indicating a systematic error in basic products thus GCP was necessary for accuracy improvement. Then four accuracy improvement models in both object space and image space were deduced and implied to the imagery, namely translation model, scale and translation model, affine model and second-order affine model. Different GCP numbers and distributions were tested for the models in both object and image spaces. Results for the tests were carefully examined and analyzed with detailed discussion. Finally the conclusion was given that the models in both object and image space were comparable with 0.6 meter accuracy in plane direction and 0.5 meter in height direction for the best when 16 points available.

6749-55, Session 13
Change detection using multi-pass and multidate data at P band
X. Dupuis, P. Dreuillet, ONERA (France); L. M. H. Ulander, A. Gustavsson, Swedish Defence Research Agency (Sweden) SAR SENSOR The ONERA RAMSES system is a flexible SAR system in constant evolution, developed mainly as a test bench for new technologies and to provide specific data for TDRI (Target Detection, Recognition and Identification) algorithm evaluation. It is flown on a Transall C160 platform operated by the CEV (Centre dEssais en Vol). This paper gives an overview of the system, lists briefly the latest upgrading (electronic and SAR processor) to acquire huge data without deramp-on-receive mod and then presents the recent campaign which, took place in Sweden, with simultaneous acquisition at P and L bands. CAMPAIGN The acquisition campaign took place in Sweden during about one month between the middle of September and the middle of October 2004. It was a joined campaign with the FOI (the Swedish Defense Research Laboratory). The Data were acquired in multi-pass and multi-date mode for several flight tacks in order to study the change of coherence, the change detection and compare them for the 2 bands used by RAMSES. The Sweden laboratory FOI acquired data at the same time in UHF and VHF bands thanks to their tow sensors: CARABAS-II and LORA-SAR. The main goal of this cooperation is to compare these sensors and their capability in many fields as, for example, change detection and MTI and for the different bands used over a same area at the same time. STUDIES In this paper we describe the analysis to access the data, then we study the multi-pass and multi-date coherences. Indeed, during this campaign the moisture changed drastically and the influence of this geophysical parameter on the interferometric and polarimetric analysis is analyzed over the forests and the bare surfaces to determine how change the polarimetric mechanism height. For this part the DEM were calibrated thanks to three corner reflectors, knowing their altitude with a very good accuracy, in order to compare the multi-dates DEM for the different polarizations. In a second part, we propose to evaluate the capability of target detection and change detection on bare surface and hidden under vegetation. The forest under which the targets were concealed is a natural forest. So we firstly analyze the geophysical parameters of this kind of forest as the backscattering coefficient depending on the observation angle and the attenuation. Then we study several techniques based on interferometry, radiometry, polarimetry or on a combination of these techniques to detect the targets and their changes depending on their location (open area or under foliage).

6749-53, Session 13
SRTM X-SAR and RAMSES DEM comparison
X. Dupuis, ONERA (France) The ONERA RAMSES system is a flexible SAR system in constant evolution, developed mainly as a test bench for new technologies and to provide specific data for TDRI (Target Detection, Recognition and Identification) algorithm evaluation. It is flown on a Transall C160 platform operated by the CEV (Centre dEssais en Vol). In this article we compare the SRTM X-SAR DEM and the ONERAs RAMSES SAR sensor DEM over the same area: the St Mandrier peninsula. Our study is composed of two parts: firstly a global assessment over the peninsula and secondly we focus on the large buildings to evaluate the detection and measurement capabilities of the SRTM products. The SRTM data were acquired from 11th to 22nd February 2000 and the RAMSES data were acquired the 1st December 1999. These data were acquired with a quite similar geometry, the incidence angle for X-SAR data is 54 and for the RAMSES data, it is 60. So, as the relief is not very high, the layover is similar. More over, all the interferograms are computed in the SAR geometry because it was a single pass acquisition mode as SRTM acquisition mode. The X-SAR data are provided in the DTED level 2 format with a pixel spacing of 1 arc second (about 30 m in both ground directions). The ground resolution for the Geocoded Terrain Corrected (GTC) is 25 m. But for the Single Look Complex (SLC) images the resolutions are 17 m along the range axis and 8-12 m along the azimuth axis. The RAMSES data are computed for three different resolutions to study the evolution of the building detection from the full RAMSES resolution to the X-SAR resolution.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

159

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology

6749-56, Session 13
Atmospheric correction of HRSC-AX line scanner data using the dark-pixel-profile (DPP) method: a summary after two years
T. U. Bucher, Deutsches Zentrum fr Luft- und Raumfahrt e.V. (Germany) The HRSC-AX (High Resolution Stereo Camera - Airborne eXtended) systems are multiple line pushbroom instruments developed by the German Aerospace Center (DLR). The dppprofile is a simple correction method based on view-angle dependent histogram statistics; it neither requires an absolute calibration nor a co-registration of the image data. It was developed at the DLR in Berlin-Adlershof and has been applied to several thousand km2 of HRSC-AX data for 2 years. The results are very encouraging: the dpp-method was applied to about 90% of the examined targets (cities and rural areas), in about 10% of the targets the method was not used due to the ambiguity of the statistics and the risk of tampering the data. In all cases the blue and the green channel were corrected, the red channel was only corrected for selected targets. A degradation of the image quality could never be observed. Depending on the amount of atmospheric effects the visual differences between corrected and uncorrected image date can be described from almost not visible, same quality to strongly improved. Also the statistics of the improved image mosaics are more conclusive and show less systematic fluctuations. The dark pixels, which are used in the statistics of the dppmethod, resulted to be in almost all examined targets dark shaded vegetation. In areas clear of vegetation other wide spread dark objects (like water bodies) have to be identified as suitable dark objects, which is not always possible.

groundwater resources. In Greece, the main geological cartographic scale was until now the 1/50.000. One of the recent projects of IGME is called Urban Geology. In the frame of that project the new cartographic scale will be the 1/5.000. Four pilots areas with different geological environment, of fifty square km each were selected. Geological, geochemical, geophysical, geotechnical, hydrogeological and other data are collected for the areas of Drama (North Greece) Nafplion & Sparti (Peloponnesus) and Thrakomakedones (Athens). All the previous analogical studies for these areas are collected and digitized. Drillings, geological and geotectonic mapping and other in situ measurements take place. The contours of the 1/5.000 topographic maps are digitized and a high detail DEM is created. The DEM and ground control points collected with GPS are used for the orthorectification of very high-resolution satellite data. Then, the orthorectified satellite data are used for the land use classification and the urban area mapping. All the analogical and digital data are imported in a geodata base specially designed for geological data. After the necessary topological control and corrections the data will be unified and processed in order to create the final layout at 1/5.000 scale.

6749-59, Session 14
Updating the 1:50.000 geological maps of Rhodes Island using remote sensing data and GIS techniques
K. G. Nikolakopoulos, P. I. Tsombos, A. Photiades, K. Psonis, Institute of Geology & Mineral Exploration (Greece) The island of Rhodes represents an uplifted easternmost segment of the Hellenic forearc extending between Greece and Turkey, which is associated with the subduction of the African plate below the Aegean. Middle Miocene-Pleistocene sedimentary basins, are separated by a stack of Alpine nappes of the Hellenide orogen exposed in uplifted fault blocks such as Plattenkalk series (Attaviros Group), Gavrovo-Tripolitza series (Lindos Group), PindosOlonos series (Profitis Ilias Group) and several outcrops of Pelagonian series with ophiolitic mlanges. These pre-Neogene formations are dominated by low-grade metamorphic sediments, which were folded and faulted during several phases of the alpine orogeny. The Rhodes region was uplifted in the Middle Miocene after which subsidence and deposition of sediments took place in the Upper Pliocene and Lower Pleistocene. In this paper we present the combined use of remote sensing and GIS techniques for the geological mapping of Rhodes Island at a 1/50.000 scale. The geological formations, geotectonic units and the tectonic structure were recognized in situ and mapped. Interpretation of medium resolution satellite images (Landsat 7 ETM and Terra ASTER) has been carried out in order to detect the linear or not structures of the study area. ASTER and SRTM DEM were used for the orthorectification of the satellite data and detection of possible faults. Thus, the in situ mapping was enhanced with data from the digital processing of the satellite data. All the data were imported in a GIS in order to be corrected and unified.

6749-57, Session 14
Logic tree-based GIS inference of geologic structure
C. C. Ryerson, T. S. Anderson, U.S. Army Corps of Engineers (USA) We describe the concept for a geographical information system (GIS) that can infer subsurface geology and material properties using geoinformatics concepts. A proof-of-concept system was devised and tested that integrates the capabilities of traditional terrain- and image-analysis procedures with a GIS to manipulate geospatial data. Structured logic trees were developed to guide an analyst through an interactive, geologic analysis based on querying and mentoring heuristic logic primarily using imagery and map data as input. The hypotheses were that a GIS can be programmed to 1) follow the fundamental logic sequence developed for traditional terrain- and image analysis procedures to infer geologic materials; 2) augment that sequence with correlative geospatial data from a variety of sources; and 3) integrate the inferences and data to develop best-guess estimates. We also developed a surface-projection method to estimate depth to bedrock, and significantly expanded an existing method to determine depth to the water table. Blind evaluations conducted at a cool, temperate site and two hot, arid sites indicate that an analyst can infer the correct geologic conditions 70-80% of the time using this method. This geologic analysis technique has multiple applications, and can be applied wherever an estimate of subsurface geology is needed and site-specific data are not available. We apply the results of our geological analysis to the prediction of local site specific seismic propagation. Comparisons are made with synthetic seismograms computed from a limited set of geological vignettes.

6749-60, Session 14
Mechanical monolithic accelerometer for low frequency seismic noise measurement
F. Barone, F. Acernese, Univ. degli Studi di Salerno (Italy); R. De Rosa, G. Giordano, Univ. degli Studi di Napoli Federico II (Italy); R. Romano, Univ. degli Studi di Salerno (Italy) This paper describes a mechanical monolithic sensor for geophysical applications developed at the University of Salerno. The instrument is basically a monolithic tunable folded pendulum, shaped with precision machining and electricdischarge-machining, that can be used both as seismometer and, in a force-feedback configuration, as accelerometer. The monolithic mechanical design and the introduction of laser interferometric techniques for the readout implementation make it a very compact instrument, very sensitive in the low-frequency seismic noise band, with a very good immunity to environmental noises. Large improvement have been produced since last version (2006), that are mainly concerned with the improvement

6749-58, Session 14
Urban geology: documentation of geo-thematic information for urban areas in Greece, the case of Nafplion, Greece
P. I. Tsombos, K. G. Nikolakopoulos, Institute of Geology & Mineral Exploration (Greece) IGME (Greek Institute of Geology and Mineral Exploration), founded in 1951, is a scientific research institute and, by legislation, the States technical adviser in geoscientific matters. Its fundamental aim is the geological study of the country and the exploration-evaluation of the mineral raw materials and

160

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology
of the mechanics and of the electronics of the instrument. In fact, to improve the mechanical performances, a prototype with elliptical hinges have been developed and tested in comparison with a prototype with circular hinges. Moreover, the data acquisition systems has been integrated with the a new control system, specifically developed for the force-feedback configuration. The theoretical sensitivity curve, calculated on the basis of suitable theoretical models, shows a very good agreement with the experimental measurements. Interesting result is that the measured resonant frequency of the instrument is below 100 mHz, demonstrating the feasibility of a resonant frequency of the order of 5 mHz with a refined tuning. For this task, the transfer function of the folded pendulum in open loop configuration is calculated measuring the resonant frequency and the quality factor for several step responses. Finally, preliminary designs and tests of wireless data transmission necessary for the sensor integration in large geographic networks are in progress, discussed in the paper together with foreseen further improvements. landslide activity in this region. In order to rapidly assess the landslide hazsards and to effectively manage this phenomenon, micro-level hazard zonation has been attempted using GIS, Remote Sensing and Information Theory in data analysis. In all 17 variables derived from various field and collateral data (DEMs, remote sensing imagery and field inventory) are identified as conditioning and triggering factors and accordingly probabilistic prediction map is prepared by this method. On the basis of histogram distribution, the polygon elements are classified into five hazard classes viz. Very Low (Ij <=-0.02), Low (-0.02 <Ij<0.103), Moderate (0.10<Ij<0.23), High (0.23<Ij<0.40) and Very High (Ij<0.40) landslide hazard prone zones. Further this probabilistic hazard map is compared with the actual landslide map generated from recent satellite data (IRS ID LISS-III + PAN, December 2002) for the accuracy of prediction. The generated hazard maps comply with the observed landslide incidences and the derived accuracy of 82% from experimental probability equation adequately proves its efficacy in deriving a reliable Landslide Hazard Zonation.

6749-61, Session 14
Comparison of multi resolution SRTM data for morphometric features identification using neural network self-organizing map: case study, Eastern Carpathians
A. Ehsani, F. Quiel, Kungliga Tekniska Hgskolan (Sweden) The Shuttle Radar Topography Mission (SRTM) was launched on 11 February 2000 and 3 arc second data were publicly released in July 2004. Easy availability of SRTM 3 arc second data ( almost 80% of earth surface) has resulted in great advances in morphometric studies and numerical description of landscape features. In this study we introduce a new numerical procedure using Neural Network- Self Organizing Map- to characterize morphometric features of landscape in multi resolution scale. We also investigate the effect of two resolutions for morphometric feature identification. Specifically we investigate how the SRTM 3arc second data projected to UTM grid with 90 meter respectively 28.5 m grid corresponding Landsat TM data resolution affect the results for morphometric characterization. Morphometric parameters such as slope, maximum curvature, minimum curvature, cross-sectional curvature are derived by fitting a bivariate quadratic surface to a window size 5*5 for the 90 m data (450 m on the ground) and 9*9 window for the 28.5 m data (about 250 m) . Kohonen Self Organizing Map as an unsupervised neural network algorithm is employed for classification of these morphometric parameters into 10 exclusive and exhaustive morphometric features. These classes were analyzed and interpreted based on spectral signature, feature space and 3D inspection of area to morphometric forms such as ridge, channel, crest line, planar and valley bottom for both data sets. The final maps were evaluated by superimposing contour lines. The result showed that the introduced procedure is a very useful method for identification of morphometric features. Resampling the SRTM data to 28.5 m resolution significantly improves the result specifically for transition landscape forms between valley bottoms to ridges. But overall the identified classes are the same.

6749-63, Session 15
ASI-Sistema Rischio Vulcanico SRV: a pilot project to develop EO data processing modules and products for volcanic activity monitoring based on Italian Civil Protection Department requirements and needs
M. F. Buongiorno, Istituto Nazionale di Geofisica (Italy); S. Amici, M. Musacchio, M. Silvestri, C. Spinetti, S. Corradini, V. Lombardo, Istituto Nazionale di Geofisica e Vulcanologia (Italy); S. Zoffoli, Italian Space Agency (Italy) The SRV project is funded by Italian Space Agency (ASI) in the frame of Earth Observations activities for natural risks management included in the National Space Plan 2003-2005. The SRV Project started in January 2007 and is coordinated by INGV which is responsible at national level for the volcanic monitoring by means of geophysical ground networks. The objective of the project is aimed to the implementation of pre-operative version for a system based on EO data analysis and ground measurements integration to support the volcanic risk monitoring activities of Italian Civil Protection Department. The project philosophy is to realize a prototype already analyzed in previous feasibility study in 2004. A very important issue regarding the product implementation of SRV is the achievement of End-User requirements, which will guarantee the compliance between the system modules and out-puts with the operative request of the End-User during the different volcanic risk phases: 1)Knowledge and Prevention, 2) Crisis 3)Post Crisis The different products will be developed on three test sites (Etna, Vesuvio and Campi Flegrei) and they will be delivered via WEBGIS interface.

6749-64, Session 15
Volcanic ash retrieval at Mt. Etna using Avhrr and Modis data
C. Spinetti, S. Corradini, Istituto Nazionale di Geofisica e Vulcanologia (Italy); M. F. Buongiorno, Istituto Nazionale di Geofisica (Italy) The volcanic ash detection procedures are based on Brightness Temperature Difference (BTD) algorithm using the thermal infrared channels centred around 11 and 12 microns of a multispectral satellite sensor. The Mie code has been is included in the retrieval procedure to compute the ash optical properties from the ash microphysical characteristics. The simulations has been carried out using a radiative transfer model. The NasaModis data and the Noaa-Avhrr of Mt. Etna eruptive plume occurred in November 2006 has been analyzing in retrieving the plume optical thickness, the particle effective radius and the size distribution.

6749-62, Session 14
Efficacy of Gis-probabilistic technique in landslide hazard zonation: a case study of Askot-Kanalichhina Route Corridor, Pithoragarh, Uttaranchal State, India
A. H. Mehta, K. C. Tiwari, S. Ganapathi, S. K. Sharma, Maharaja Sayajirao Univ. of Baroda (India); D. Ramakrishnan, Indian Institute of Technology (India) The strategic and industrially important Askot-Kanalichhina route corridor bordering Indo-Nepal border along Kali-River Valley in the north-eastern Kumaun Himalayas is characterized by lofty hill ranges, steep valleys, cliffs and huge accumulation of talus and scree deposits owing their origin to complex physical, geologic and tectonic processes of active Himalayan orogenic belt. High intensity rainfall, active seismicity coupled with deforestation and road cutting activities has increased the

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

161

Conference 6749: Remote Sensing for Environmental Monitoring, GIS Applications, and Geology

6749-65, Session 15
Volcanic activity monitoring using geostationary satellite
M. Musacchio, C. Spinetti, S. Corradini, M. Silvestri, S. Amici, V. Lombardo, L. Merucci, Istituto Nazionale di Geofisica e Vulcanologia (Italy); A. Bartoloni, A. Canestro, A. Fiorani, Consultant (Italy); M. F. Buongiorno, Istituto Nazionale di Geofisica (Italy) The objective of the study is to investigate how MSG data can be used to study and monitor High Thermal Events (HTE) and volcanic plume content. Time series of MSG data can provide quantitative information on phenomena evolution through a time. This study approach the Etna volcanic activity in July and November 2006. A new pre-processing chain has been developed in order to extract and co-register HRIT data and store them in a rational database. Preliminary results showed that the SWIR and MIR channels are the most suitable in order to retrieve lava flow parameters. The MIR channel has the further advantage of being less sensitive to cloud presence. The high revisit time of MSG represents an important parameter for the monitoring of active volcanoes especially for those situated near populated areas.

162

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6750: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing
Monday-Wednesday 17-19 September 2007 Part of Proceedings of SPIE Vol. 6750 Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing III

6750-01, Session 1
A fluorescence imaging lidar for the control of cultural heritage
L. Palombi, G. Cecchi, D. Lognoli, V. Raimondi, Istituto di Fisica Applicata Nello Carrara (Italy); L. F. Masotti, Univ. degli Studi di Firenze (Italy) The fluorescence lidar imaging technique turns particularly useful for the control of monuments. The investigated topics range from the detection of biodeteriogens to the characterisation of stones and other masonry or restoration materials, such as protective treatments. In addition, the fluorescence lidar imaging is a non-destructive technique and shows the advantage that it can be carried out in situ without the use of scaffolding that, beside being costly, limits the access to the monument and its use. This paper presents the main technical features of a new fluorescence imaging lidar system specifically developed for the diagnostics on the cultural heritage, whose operative conditions include outdoor and indoor environments, and the possibility of monitoring vaults and ceilings. This fluorescence lidar prototype is mainly composed of a Qswitched, tripled frequency Nd:YAG laser (\@355 nm), a 1 m focal length Newtonian telescope and a 300 mm focal length spectrometer coupled to an intensified, gated 512x512 CCD detector. Imaging is carried out via a scanning system realised with a computer controlled mirror. The lidar prototype includes also a target pointing system for referencing the acquired fluorescence images on the target.

tuned BBO crystals produce second harmonic and sum frequency of 1.052 and 1.319 micrometer pulses, producing visible wavelengths at 0.526, 0.585, and 0.660 micrometers; green, orange, and red, reminiscent of a STOP light. Second harmonic generation of pulses at 0.585 micrometers produce ultraviolet radiation at 0.293 micrometers. An attractive feature of this laser design is that it is line tunable. With a properly designed resonator, a potential combination of 44 possible fundamental laser transitions can be operated, which, when used with non-linear processes, results in a line tunable UV laser operating in the range 0.263 to 0.339 micrometers.

6750-04, Session 1
Highly efficient 1.9 m Tm3+/Yb3+-doped tellurite fibre laser
B. D. O. Richards, Univ. of Leeds (United Kingdom); Y. H. Tsang, D. J. Binks, The Univ. of Manchester (United Kingdom); J. Lousteau, A. Jha, Univ. of Leeds (United Kingdom) We report, for the first time to our knowledge, a Tm3+/Yb3+doped tellurite fibre laser operating at wavelengths in the range 1.910 - 1.994 m. Two different pump schemes have been demonstrated for this laser: a 1.088 m Yb3+-doped silica fibre laser simultaneously pumping the Tm3+: 3H5 and Yb3+: 2F5/2 levels, and a 1.610 m Er3+/Yb3+-doped silica fibre laser directly pumping the Tm3+: 3F4 upper laser level. For the 1.610 m pumping, a slope efficiency of 75% has been achieved in a 32 cm long fibre which is very close to the Stoke efficiency limit of ~80%. An output power of ~280 mW has also been achieved, but with no signs of saturation or fibre damage suggesting that higher output powers should be possible. For the 1.088 m pumping there is very strong pump ESA resulting in bright blue (480 nm) and near-IR (800 nm) emission and this limits the achievable slope efficiency, which in this case was a maximum of 10% for a 16 cm long fibre. With this pump scheme, the highest observed output power was 67 mW, and further power scaling was limited due to the intense ESA and thermal damage to the pump end of the fibre. Lasing has been achieved in <10 cm lengths of this fibre making this material a promising candidate for ultra compact medium power mid-IR laser sources for range-finding, medical and atmospheric monitoring and sensing applications.

6750-02, Session 1
Sea floor classification from airborne lidar data
M. Tulldahl, C. Vahlberg, Swedish Defence Research Agency (Sweden); A. Axelsson, H. Karlsson, Airborne Hydrography AB (Sweden); P. Jonsson, Lunds Tekniska Hgskola (Sweden) Airborne depth sounding lidar has proven to be a valuable sensor for rapid and accurate sounding of shallow areas. The received lidar pulse echo contains information of the sea floor depth, but also other data can be extracted. We currently perform work on bottom classification and water turbidity estimation based on lidar data. In this paper we present theoretical background and experimental results on bottom classification. The algorithms are developed from simulations and then tested on experimental data from the operational airborne lidar system Hawk Eye II. We compare the results to field data taken from underwater video recordings. Our results indicate that bottom classification from airborne lidar data can be made with high accuracy.

6750-05, Session 1
High-power single-frequency waveguide laser for detection applications
S. Taccheo, G. Della Valle, Politecnico di Milano (Italy); K. M. Ennser, Univ. of Wales Swansea (United Kingdom); D. Milanese, Politecnico di Torino (Italy); D. Barbier, Teem Photonics SA (France) Infrared lasers operating in single-frequency at 1.5 micron are key elements for applications to several fields such as spectroscopy and optical sensing. However device specifications may become very demanding in terms of power consumption, compactness, insensitivity to environmental disturbance and high temporal coherence. In particular airborne and spaceborbne applications are very sensitive to technical noise, including vibrations, and a monolithic structure is therefore preferred to bulk devices. With this respect waveguide laser offers the highest immunity with respect external perturbation and therefore is a promising candidate for LIDAR application and optical radar. However, looking at the literature up to now only few mW output level robust single-frequency waveguide laser were demonstrated. In this paper we demonstrate that an optimised configuration may generate over 20 mW in robust single-frequency operation. We used an Er:Ybdoped phosphate glass waveguide only 9-mm long. The waveguide was fabricated by two steps ion-exchange techniques using a commercial Schott IOG-1 phosphate glass base doped with Er and Yb. A power of 23 mW was obtained for 280 mW pump power. More details will be reported at the conference. We also evaluate limits of power extraction from

6750-03, Session 1
A novel UV-laser for lidar applications
B. M. Walsh, N. P. Barnes, NASA Langley Research Ctr. (USA) Lasers operating in the ultraviolet are useful for remote sensing on the Earth. Range resolved measurements of ozone (O3) and sulfur dioxide (SO2) can be obtained using lidar techniques and line tunable lasers operating near 0.3 micrometers. Ozone is of interest because of its importance in solar ultraviolet shielding in the stratosphere and smog contributions in the troposphere. Sulfur dioxide is also of interest because of smog pollution in the troposphere, and is a precursor to acid rain. A novel ultraviolet laser is demonstrated using a dual wavelength Nd:YAG oscillator, sum frequency and second harmonic process. Synchronous pulses at 1.052 and 1.319 micrometers are amplified, mixed and subsequently doubled, producing pulses at 0.293 micrometers. The dual wavelength laser oscillator, referred to as the STOP light laser, produces Synchronous Tunable Optical Pulses in Q-switched mode from a single Nd:YAG oscillator operating on transitions between the 4F3/2 to 4I11/2 and 4I13/2 manifolds simultaneously. Angle

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

163

Conference 6750: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing
waveguide laser in robust single-frequency operation and we obtained with in a preliminary experiment over 80 mW with less than 300 mW pump power. We believe that further optimisation of cavity length and Bragg gratings design will allow to reach 100-mW output power in single frequency operation and narrow-linewidth (below 10 kHz). Allowing long operation range and high spatial resolution in detection applications. this method will be presented. As a perspective we will also discuss a possible extension of the present system, using a second Michelson interferometer, that in principle enables additional measurements of air temperature, density, and scattering ratio.

6750-08, Session 1
Automated polarization-discrimination technique to minimize lidar-detected skylight background noise
Y. Y. Hassebo, City College/CUNY (USA) Recently, there have been numerous significant interests in lidar signal-to-noise ratio (SNR) improvements, particularly for lidar day-time operations. Previously, we devised in the remote sensing laboratory at the City College of New York a polarization discrimination technique to maximize lidar detected SNR using of the advantage of naturally polarization properties of scattered skylight radiations to track and minimize the detected sky background signal (BGS). This tracking technique has been achieved by rotating, manually, a conjunction of polarizer and analyzer on the lidar transmitter and receiver subsystems, respectively. The polarization orientation at which the minimum BGS occurs equals the solar azimuth angle has been confirmed in our recent work, theoretically and experimentally, assuming the single scattering theory. In this paper, a design to automate the polarization discrimination technique by real time tracking of the azimuth angle to achieve the minimum BGS is presented. A feedback control system to track the minimum BGS by rotating the detector analyzer and the transmission polarizer simultaneously to maximize the SNR and attainable lidar ranges, achieving the same results as would be done manually, is introduced. Analytical results of this design are investigated.

6750-06, Session 1
Fiber lasers: new effective sources for coherent lidars
J. Cariou, Leosphere (France); G. Canat, M. Valla, ONERA (France) Originally developed for telecommunications, fiber lasers are now becoming new effective sources for coherent lidars allowing new instruments to be designed. The advent of the double clad fiber, along with advances in semiconductor pump diode sources, have allowed rapid power scaling of both pulsed and CW fiber sources. The unique capabilities of fiber sources, coupled with significant commercial and academic progress in implementation, have driven fiber technology to enter active remote sensing markets as signal sources and amplification stages for direct detection lidars and coherent lidars as well. Some interesting fiber lasers benefit from a good transmission in the near infrared spectral band: Ytterbium lasers (1.0-1.06m), Erbium lasers (1.48-1.62m) and Thulium lasers (1.8-2.1m). However, useful wavelengths have to be tuned between absorption H20 and CO2 lines. Eye safety may be an issue for atmospheric lidars. Above 1.4 m, the maximum permitted exposure (MPE) is 3 orders of magnitude higher than for shorter wavelengths and an eye-safe operation is possible even with multi-watt lasers. Low power fiber lasers using single mode fibers have a good spatial quality. However, higher power lasers and amplifiers need larger fiber cores, to store enough energy and to avoid non linear effects. The single mode operation is not obvious, and large core low aperture number fibers have to be developed. Spectral quality is an issue for coherent lidars and non linear effects in fibers must be corrected. Moreover, low intensity noise and stable polarization are required for lidar sources. The presentation will focus on laser requirements for coherent lidars and give a state of the art of existing fiber sources.

6750-09, Session 1
Multiplicative decomposition as a tool for intercomparison of range-resolved pulsed and modulated CW lidar capacities
R. R. Agishev, Kazan State Technical Univ. (Russia) In this paper, the analytical comparison of key capabibilities for both continues-wave frequency modulated (CWFM) lidar and pulsed incoherent (PIC) lidar is presented. Principles of CW-FM-lidar architectures are described and the CWFM-lidar equation is derived. The classification of continuous-wave and pulsed lidars depending on the required receiver bandwidth (broad-band versus narrow-band) is given. The dimensionless parameterization methodology developed primarily for incoherent pulsed lidars is now extended to both broad- and narrow-band lidars to perform their detailed estimates and a comparison of the operation range and sensitivity with a considerable reduction in the number of lidar system parameters required. When the systems considered provide equal transmitted energies during the repetition period, the narrow-band lidar has higher potential than the broad-band one due to the substantial narrowing of the required transmission band of the lidar receiver for the same range resolution. In particular, quantitative expressions for the bandwidth and operating range for all lidar classes are provided and the performance improvement in both the operating range and sensitivity threshold using NB-lidars above BB-lidars is estimated. It is shown that the degree of improvement between the NB and the BB systems, complementary to the bandwidth gain, is determined by the operating range of the NB lidar and therefore maximized when the pulse energy is low and the atmopheric attenuation and/or background power is high. Finally, different classes of heterodyning techniques including radio-frequency, optical heterodyning (OH) and opto-electronic heterodyning (OEH) are considered and a quantitative sensitivity comparison of lidars with different heterodyning techniques in relation to the reference pulsed incoherent lidar system is conducted. It is shown, that for low-level (<2%) efficiency of the optical heterodyning, the CWFM+OEH-lidar can in fact exceed the sensitivity the OH-lidar, thereby providing the most appropriate methodology. References: 1. R.Agishev, A.Comeron at al. Applied Physics B, vol. 79, No.2,

6750-07, Session 1
Development of a fringe-imaging Michelson interferometer for wind speed measurements using a short-range 355-nm Rayleigh-Mie lidar
N. Cezard, A. Dolfi-Bouteyre, ONERA (France); J. Huignard, Thales Research and Technology (France); P. H. Flamant, Ecole Polytechnique (France) We report the development of a 355-nm lidar system for shortrange (~ 100m) wind speed measurements, using a fringeimaging Michelson interferometer as a receiver. Such an instrument could be used for reliable True Air Speed measurements from aircraft at any altitude. Using a UV wavelength enables to operate efficiently in clear air condition (Rayleigh scattering from molecules), while the fringe imaging technique ensures the speed measurement to remain unbiased even in the presence of strong Mie scattering from particles. We first present the concept and measurement principle, and provide the instrument theoretical accuracy for shot-noise limited measurements. The key technical aspects of the system are then presented, and some recorded fringe patterns are shown and discussed. The signal processing is designed to provide estimates of five fringe parameters that are: mean energy, contrast, periodicity, angular orientation, and phase. The wind speed is retrieved from the phase difference between the lidar signal and a reference signal (the transmitted laser pulse). The signal processing is implemented in two-stages: the first stage uses a Fourier transform analysis providing with 1st order estimates; then, these latter initialize a second stage maximum-likelihood estimator. The preliminary results validate the instrument concept and feasibility. To solve the problem arising from the noticeable shot-to-shot fringe phase variation of the reference signal, a system using a fast CCD camera has been designed. This should provide self-referenced phase measurements for each pulse. The first results obtained with

164

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6750: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing
pp.255-264 (2004). 2. R.Agishev, B.Gross at al. Applied Physics B, vol. 80, No. 6, pp. 765-776 (2005). 3. R.Agishev, B.Gross at al. Optics and Lasers in Engineering, vol. 44, No. 8, pp. 779-796 (2006).

6750-12, Session 2
Volumetric gas monitoring through a DSA laser network for the estimation of gas emission flux by surface sources
F. Cuccoli, R. Lupo, T. Berna, L. Facheris, Univ. degli Studi di Firenze (Italy) Through DSA (Differential Spectroscopy Absorption) based IR portable laser system, gas concentration of molecular species along air optical paths in real time can be determined. In particular we can measure in-situ concentration in the air where soil gas emission phenomena are present (i.e. volcanic, geothermal and industrial sites) just above the emission areas with optical path length up to 1 km. Here we propose a measurement approach for the estimation of the emission flux by a surface-distributed source, based on the use of IR laser measurements and atmospheric diffusion models. An ad hoc disposition of the optical links close to the emission area allows to measure the gas concentration over a closed surface corresponding to an air volume that covers the whole emission area. The real time concentration measurements over this closed surface, associated to suitable diffusion models, allow us to estimate the emission flux of the area under exam. The diffusion model to be applied strictly depends on the current atmospheric conditions, therefore it requires the knowledge of the main atmospheric parameters. In this paper we present some simulations of measurements about a system for the laser monitoring of a parallelepiped. The closed surface is therefore defined by 5 of its sides, while the 6th is the emission surface. We discuss some estimation results using diffusion models where the air diffusion and transportation phenomena are due mainly to the wind strength.

6750-10, Session 1
Finding the mixing layer: algorithms to identify the mixing layer height in lidar signals automatically
J. Mehnert, M. Pesch, Technische Univ. Berlin (Germany) The mixing layer provides an atmospheric barrier. Its height has become an important parameter in meteorology. To detect the mixing layer with Lidar seems to become a requirement to sell Lidar for meteorology standard applications. The detection of the mixing layer is even manually not easy because it is overlaid by noise and may contain more than one layer (e.g. planetary boundary layer). Two known solutions take in advance that most backscattering aerosols are located in the mixing layer. This causes (for wavelengths in the infrared) on the one hand a sharp fall in the backscattering signal and on the other hand, that the signal of the mixing layer should be more significant different from the noise than signals from higher altitude. The latter idea has already been deployed but it includes a variable to distinguish between the signal and the noise (signal-noise-ratio). To avoid this variable we designed numerous algorithms by trying to find correlations of minima and maxima in the derivations. Therefore an intensive analysis of the global and local extremes of numerous Lidar signals was followed by the evaluation of different correlatives and conditionings of the curves. The algorithm with the best results takes the Lidar signal and after conditioning the signal it searches for a maximum followed in 210 meters by a minimum in the derivation. It does not operate total automatically but is a helpful tool for the identification of the mixing layer.

6750-13, Session 2
The development of an optical parametric amplifier for CO2 profiling
J. F. Burris, NASA Goddard Space Flight Ctr. (USA); D. A. Richter, ITT Industries Systems Inc. (USA) An optical parametric amplifier has been developed to generate 1570 nm radiation from a 1064 nm pump at high efficiency. A micropulse Nd:YAG operating with an average power of 2 watts (10 kHz) is used to pump a PPLN crystal injection seeded by two CW distributed feedback lasers: one at 1570.824 nm and the second at 1570.670 nm. A three stage amplification process has demonstrated a conversion efficiency of ~40% from the pump into the signal wavelengths. The 1-nanosecond signal has a measured time-averaged jitter of <40 MHz.

6750-11, Session 2
New concept design of a DIAL based on a pulsed quantum-cascade laser
A. P. Lytkine, B. Lau, W. Jaeger, J. Tulip, Univ. of Alberta (Canada) A new concept design of a differential absorption LIDAR (DIAL) for range-resolved remote chemical sensing in the atmosphere has been developed and validated. The DIAL is based on a pulsed quantum-cascade laser (QCL) with chirped laser pulses utilized to scan target absorption features. Range-resolved gas sensing has been implemented using a set of retroreflectors arranged along an absorption path with a spacing allowing for time multiplexing of measurement sections. The sections can be interrogated with a single laser pulse to retrieve the distribution of gas concentrations along the absorption path. The spatial resolution of the QCL-based DIAL is defined by the length of measurement sections set along the absorption path. In the validation experiments we used a pulsed DFB QC laser operating at room temperature. The laser was excited with 200ns current pulses at 10-kHz repetition rate. A 915.0 - 916.5cm-1 spectral interval containing ammonia and ethylene absorption lines was scanned with chirped laser pulses at laser substrate temperature set to 20.8 0C. We have demonstrated the feasibility of the new measurement method using two retroreflectors with a 25-m spacing between them. The maximum absorption path in the atmosphere was ~ 100 m. We will present the results of a field campaign with a QCL-based DIAL operating near 10.9 m. Novel DIAL systems based on pulsed QCLs can be developed for multi-species rangeresolved trace gas sensing in the atmosphere over the ranges up to ~ 1 km.

6750-14, Session 3
Optimization of lidar data processing: a goal of the EARLINET-ASOS project
A. Amodeo, G. Pappalardo, Consiglio Nazionale delle Ricerche (Italy); J. Bsenberg, Max-Planck-Institut fr Meteorologie (Germany); A. Ansmann, Leibniz-Institut fr Troposphrenforschung e.V. (Germany); A. Apituley, Rijksinstituut voor Volksgezondheid en Milieu (Netherlands); L. Alados-Arboledas, Univ. de Granada (Spain); D. S. Balis, Aristotle Univ. of Thessaloniki (Greece); C. Bckmann, Univ. Potsdam (Germany); A. P. Chaikovsky, Instytut Fizyki (Belarus); A. Comern, Univ. Politecnica de Catalua (Spain); G. DAmico, Consiglio Nazionale delle Ricerche (Italy); V. Freudenthaler, Ludwig-Maximilians-Univ. Mnchen (Germany); G. H. Hansen, Norwegian Institute for Air Research (Norway); I. Mattis, Leibniz-Institut fr Troposphrenforschung e.V. (Germany); V. Mitev, Observatoire Cantonal de Neuchtel (Switzerland); D. Mller, Leibniz-Institut fr Troposphrenforschung e.V. (Germany); D. N. Nicolae, National Institute of Research & Development for Optoelectronics (Romania); L. Osterloh, Univ. Potsdam (Germany); A. D. Papayannis, National Technical Univ. of Athens (Greece); M. R. Perrone, Univ. degli Studi di Lecce (Italy); A. Pietruczuk, Institute of Geophysics (Poland); M. Pujadas, Ctr. de Investigaciones Energticas, Medioambientales y Tecnolgicas (Spain); J. Putaud, Joint Research Ctr. (Italy); F. Ravetta, Univ. Pierre et Marie Curie (France); V. Rizi, Univ. degli Studi dellAquila (Italy); V. B.

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

165

Conference 6750: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing
Simeonov, Ecole Polytechnique Fdrale de Lausanne (Switzerland); N. Spinelli, Univ. degli Studi di Napoli Federico II (Italy); D. V. Stoyanov, Institute of Electronics (Bulgaria); T. Trickl, Forschungszentrum Karlsruhe (Germany); M. Wiegner, Ludwig-Maximilians-Univ. Mnchen (Germany) EARLINET-ASOS (European Aerosol Research Lidar Network Advanced Sustainable Observation System) is a five year EC Project started in 2006. It is based on the EARLINET infrastructure (the first aerosol lidar network, established in 2000, with the main goal to provide a comprehensive, quantitative, and statistically significant database for the aerosol distribution on a continental scale) and will provide appropriate tools to improve the quality and availability of the continuous observations. The EARLINET multi-year continental scale data set is an excellent instrument to assess the impact of aerosols on the European and global environment and to support future satellite missions. The project is organized in several activities addressed in optimizing instruments and algorithms existing within EARLINET-ASOS, exchanging expertise, with the main goal to build a database with high quality aerosol data. In the frame of the project, the optimization of the algorithms for the retrieval of the aerosol optical and microphysical properties is a crucial activity. The main objective is to provide all partners with the possibility to use a common processing chain for the evaluation of their data, from raw signals to final products. Raw signals may come from different types of systems, and final products are profiles of optical properties, like backscatter and extinction, and, if the instrument properties permit, of microphysical properties. This activity will have a strong impact on the scientific community because data with homogeneous well characterized quality will be made available in near real time. Acknowledgments The financial support by the European Commission under grant RICA-025991 EARLINETASOS is gratefully acknowledged.

6750-16, Session 3
Combining OPAC and lidar
D. N. Nicolae, C. L. Talianu, C. M. Radu, National Institute of Research & Development for Optoelectronics (Romania); S. Stefan, Univ. din Bucuresti (Romania) The properties of aerosol particles are highly variable, both in time and space. This reffers to the number density, the microphysical properties (size distribution, refractive index, effective radius), and to the height distribution. In most cases the actual properties are not known. Using lidar data together with models can help improve the knowledge regarding the particulate atmospheric constituents which affect local radiative forcing, the radiation balance of the earth, and thus climate. This paper presents an attempt to integrate elastic backscatter lidar data in OPAC software package in order to find the most realistic aerosol vertical distribution and their optical and microphysical characteristics. The necessity to reduce the variability of naturally occurring aerosols to typical cases, but without neglecting possible fluctuations, is achieved in OPAC by the use of a dataset of typical internally mixed aerosol components. In addition, any mixtures of the basic components can be used to calculate the overall optical parameters. Experimental or modeled meteorological profiles (temperature, pressure, relative humidity) in complementary to experimental lidar data are used to calculate the solutions of lidar equation that fits, in an iterative manner, to the output of the model. Two type of uncertinties are diminished in this way: first, the modeled profiles of lidar ratio are used in lidar data processing instead of a constant value; second, aerosol height profiles are no longer being assumed in the model, but directly measured. This procedure was applied to synthetic lidar signals in order to test its advantages and limitations.

6750-17, Session 3
AGLITE: a multiwavelength lidar for measuring particle sizes, concentrations, and fluxes emitted from agricultural operations
T. D. Wilkerson, G. E. Bingham, V. V. Zavyalov, C. Marchant, J. M. Anderson, L. P. Andrew, Space Dynamics Lab. (USA) AGLITE is a new multiwavelength lidar developed for the Agricultural Research Service of the United States Department of Agriculture and its program on particle emissions from animal production facilities. The lidar transmission system is a pulsed Nd:YAG laser at 355, 532 and 1064 nm operating at a pulse rate of 10 kHz. We analyze and model lidar backscatter and extinction to extract the aerosol physical properties. All wavelength channels are used simultaneously, day or night, using photon counting and high speed data acquisition. The housing for the lidar is a transportable trailer suitable for allweather operation at any accessible site. We direct the laser and telescope FOVs to targets of interest in both azimuth and elevation. The lidar has been applied in multidisciplinary atmospheric studies at a swine production farm in Iowa, a dairy in Utah, and almond and cotton facilities in California. Prominent aerosol plumes emitted from these facilities were measured to study variations with temperature, turbulence, stability and farm operations. Arrays of particle samplers and turbulence detectors were also used by colleagues specializing in those fields and are compared with the lidar data. Lidar measurements also focused on air motion as seen by long duration scans of the farm region. The value of multiwavelength, eyesafe lidars for agricultural aerosol measurements has been confirmed by the successful operation of AGLITE. In this paper, we demonstrate the ability to measure particulate fluxes generated by a specific facility and its components. Many types of particulates and gases from agricultural operations can have major consequences for air quality. Accounting for terrain and the presence of buildings requires large scale, whole facility monitoring systems for understanding the overall emission source and strength characteristics. The ability to use standoff lidar methods to determine the movement, concentrations and fluxes of emissions on a whole facility basis opens new and important avenues for model development and verification. Rational monitoring and control of emitted fluxes will be greatly aided and abetted through validation, adoption, and increased use of optical remote sensing.

6750-15, Session 3
Characterization of atmospheric aerosols with multiwavelength Raman lidar
D. Mller, I. Mattis, A. Ansmann, U. Wandinger, D. Althausen, Leibniz-Institut fr Troposphrenforschung e.V. (Germany); A. Kolgotin, Physics Instrumentation Ctr. (Russia) Multiwavelength Raman lidars are a powerful tool for the vertically resolved characterization of optical and microphysical properties of atmospheric aerosols. The stationary Raman lidar at the Leibniz Institute for Tropospheric Research has been collecting data on aerosol properties since 1997. Several other research groups of the European Aerosol Research Lidar Network (EARLINET) are currently installing such Raman lidar systems. Additional multiwavelength lidar stations became operational in Russia, Japan, South Korea, and Spitsbergen in recent years. Standard output of these systems are particle backscatter coefficients at 355, 532, and 1064 nm, and particle extinction coefficients at 355 and 532 nm. From that follow profiles of Angstrm exponents and extinction-to-backscatter ratios, which contain qualitative information on mean particle size and light absorption properties. On the basis of such data one may separate between different aerosol types. Sophisticated data inversion algorithms, which have been developed in the past ten years, use the optical information and provide microphysical aerosol parameters, e.g., particle effective radius and complex refractive index, from which single-scattering albedo is derived. In this contribution we present measurement examples that show the capabilities of multiwavelength Raman lidar systems. Results are taken from measurements at Central Europe. Further examples obtained during field campaigns in Africa, and South and East Asia show the need for a worldwide network of such lidar stations, which would greatly enhance knowledge on the global four-dimensional distribution of aerosols. The EARLINET-ASOS project is funded by the EC under grant RICA-025991

166

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6750: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing

6750-18, Session 3
Characterization of the atmospheric aerosol by combination of lidar and sunphotometry
L. Alados-Arboledas, J. L. Guerrero-Rascado, H. Lyamani, F. J. Olmo-Reyes, Univ. de Granada (Spain) During the summer 2006 an extended observational campaign of atmospheric aerosol has been developed in the area of Granada, South-eastern Spain. From July to the end of September two Cimel CE-318 radiometers have been operated continuously, one at Andalusian Centre for Environmental Studies (CEAMA), located in the urban area of Granada, a nonindustrialised medium size city (37.16N, 3.61W and 680 m a.m.s.l.), and the second one at the Astronomical Observatory of Sierra Nevada (37.06N, 3.38W and 2896 m a.m.s.l.), with a short horizontal separation between stations that allows us to consider both instrument located in approximately the same vertical column. The CIMEL CE-318 measurements have been used to retrieve the aerosol columnar properties, including the columnar volume size distribution, volume scattering phase function, asymmetry factor and single scattering albedo, by means of appropriate inversion procedures. Additionally, at the CEAMA a Raman Lidar system based on a Nd:YAG laser source operating at 1064, 532 y 355 nm and including elastic, polarized and Raman shifted detection has been used to derive profiles of several atmospheric aerosol properties. In this paper we present analyses of the changes and temporal evolution detected in atmospheric aerosol vertical profile. Several long range transport episodes have been detected and backtrajectories analyses and synoptic information has been used in the discussion of results. Acknowledgements: This work was supported by CICYT from the Spanish Ministry of Science and Technology through project CGL2004-05984-C07-03 and by Andalusian Regional Government project P06-RNM-01503.

6750-20, Session 3
An automatic, low-cost telemetric lidar for the continuous monitoring of urban aerosols in the surface Layer
M. Del Guasta, F. Castagnoli, V. Venturi, Consiglio Nazionale delle Ricerche (Italy) An old-fashioned technology such as the searchlight profiler, which was used in the 1960s for the pioneristic profiling of atmospheric backscatter has been revisited in our telemetricLIDAR. The instrument is intended for the low-cost monitoring of aerosols, with a limited useful range of 100-200 meters. A modulated CW laser beam (532 nm) is used to probe the atmosphere. An 8-cm refractive telescope is used to collect the backscattered light, as in any ordinary LIDAR. A photodiode array placed in the focal plane of the lens is used as a detector. As in a telemetric system, the distance between the laser beam and the telescope determines the distance of the measurement volume for each photodiode pixel. The instrument is mainly suitable for the monitoring of urban aerosol pollution, because the range resolution is adequate and similar to that of an ordinary LIDAR in the surface layer. In the case of urban aerosols, light backscatter can be converted into aerosol mass concentration once the relative humidity is known. For an interpretation of the results in terms of meteorological conditions, the instrument is equipped with ordinary meteorological sensors. The instrument was run for a first field campaign in Prato (Italy), together with a Beta-attenuation PM10 instrument. The telemetric-LIDAR provided PM10 data every 5 minutes, while hourly data were supplied by the Beta-attenuation instrument. The instruments reliability as a monitoring tool, at least when Relative Humidity was <90%, was confirmed.

6750-21, Session 3
Characterization of super-cooled water aerosols by dual polarization lidar at 1.574 m
G. J. Fochesatto, K. Sassen, R. L. Collins, Univ. of Alaska/ Fairbanks (USA) Ice fog is a particular pollution mechanism during winter time in Arctic urban settings. Aerosols emitted from local production (e.g. car exhaust, open waters, living species, power plants) when suspended in the atmospheric boundary layer at lower temperatures than the water freezing point (273 K) form supercooled fog and ice-fog depending upon the thermodynamic environmental equilibrium regime. Lidar experiments have been conducted in Fairbanks, AK in horizontal path deployment to characterize Arctic fog under different thermodynamic conditions. In this paper the optical polarization Lidar signature of Arctic urban fog is obtained at the ambient temperature regime from 253 K to 240 K.

6750-19, Session 3
Calibrating a backscatter lidar for continuous measurements of PM2.5
M. Pesch, D. C. Oderbolz, Technische Univ. Berlin (Germany) One of the main issues of atmospheric research and air quality control is the reduction of harmful particulate matter (PM) in the atmosphere. Small particles can enter the human airways and cause serious health problems such as COPD (Chronic Obstructive Pulmonary Disease), asthma or even lung cancer. Recently, interest has shifted from PM10 to finer fractions of particulate matter, e.g. PM2.5, because the health impact of finer particles is considered to be more severe. Up to now measurements of particulate matter were carried out mainly at ground level. However important atmospheric processes - particle formation, transport and - take place predominantly at higher altitudes in the planetary boundary layer. A Lidar system provides the ability to observe these processes where they occur. The new method outlined in this paper demonstrates the use of a small sized and quite inexpensive lidar in stand-alone operation to determine transport processes of particulate matter, and PM2.5 in particular. Continuous measurements of PM2.5 as a reference are gained with a conventional in-situ monitor, installed on a tower at an altitude of 325 m in the North of Berlin (Frohnauer Turm). These PM2.5 measurements will be compared with backscatter Lidar data (1064 nm) taken from approx. 100 m over ground up to an altitude of 15 km with a spatial resolution of 15 m. The vertical backscatter profiles at 325 m will be correlated to the concentrations obtained by the PM2,5 monitor on the tower. Both measurements have a time resolution of 180 s to observe also processes that take place at short time scales. The objective is to gain correlation functions for estimating PM2.5 concentrations from backscatter Lidar data. Such a calibrated Lidar system is a valuable instrument for environmental agencies and atmospheric research groups to observe and investigate causes of high level PM concentrations. First results show a reasonably good linear correlation depending on the level of the relative humidity.

6750-22, Session 4
Standoff determination of bioaerosol size based on double scattering measurement with MFOV lidar: concept and experimental validation
G. A. Roy, N. Roy, Defence Research and Development Canada (Canada) Recently, with the Multiple field-of-view (MFOV) lidar developed at DRDC Valcartier we have demonstrated that the size and the concentration of relatively dense water clouds can be measured. These innovative results combined with recent developments with an ICCD camera have lead us to do a preliminary investigation on the characterization of bioaerosol clouds. MFOV lidars measure the multiple scattering of a laser beam propagating in the atmosphere. It is the multiple scattering signal that contains information on the particle size and the concentration of the aerosol. The difficulty in dealing with bioaerosol clouds is that they are sub-visible. Their low concentrations do not produce strong multiple scattering signals; in addition they are mixed with background aerosols. The basic principle of the technique we developed to characterize bioaerosol clouds is based on second order scattering: a first forward scattering from the bioaerosol followed by a backscattering on background aerosols at a given distance at the back of the bioaerosol. This last scattering is collected

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

167

Conference 6750: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing
by the collecting optic We present MFOV lidar measurements on low concentration aerosol with particle size ranging from a few microns to 100 m. The bioaerosols were disseminated into a 22 m long aerosol chamber. The open square side of the chamber is 2.4 m long. The lidar was located 100 m from the aerosol chamber and the scatter light was measured to distance ranging from 10 m to 90 m away the back of the aerosol chamber. Analysis of data shows that the size of the aerosol is recovered with in an accuracy of 10% while the bioaerosol cloud concentration seems to be achievable within a factor of 2 or 3 so far. for 3d-wind profiles or in nadir-viewing mode for high resolution vertical wind measurements that can be combined with the DIAL profiles using eddy correlation to obtain profiles of the latent heat flux in the convective boundary layer. The wind lidar spatial resolution is 100 m in the vertical and 150 m (vertical wind, boundary layer) to 12 km (3d-wind profiles, whole troposphere) in the horizontal. The DIAL horizontal and vertical resolution ranges from 150 m in the boundary layer to 500 m in the upper troposphere. Examples of first results covering the different measurement possibilities will be presented.

6750-25, Session 5 6750-23, Session 4


Fluorescent lidar technologies for organic aerosol remote analysis
G. G. Matvienko, A. I. Grishin, G. M. Krekov, Institute of Atmospheric Optics (Russia); V. I. Timofeev, M.V. Lomonosov Moscow State Univ. (Russia) The paper describes the fluorescent lidar created for monitoring of the atmosphere and for estimating the content of fluorescent components of organic aerosol. The lidar operation is based on the use of ultraviolet radiation of harmonics of Nd:YAG solid state laser for exciting the atmospheric fluorescence and the spectral analysis of the atmospheric fluorescence is used in the near ultraviolet and blue spectral range with the resolution of 2 nm. The lidar was found to be efficient for remote analysis of organic aerosol occurring as a result of vegetation emission of secondary metabolites to the atmosphere. Fluorescence spectra processing allows us to select some organic compounds, which molecules contain 7 and more carbon atoms. In order to better investigate the link between laser-induced fluorescence spectra and plant physiology, numerical experiments under controlled conditions were carried out for different kinds of samples and vegetation stresses. In numerical experiments some new algorithms of Monte Carlo method were used. The results of comparison with real lidar measurements are dicussed. Neural Network (NN) technology also were used for some organic components (Indole) identification by fluorescence spectra. The methodology are shown acceptable results and availability of application. The advantage of this methodic under another is stability work in cases of data noisiness and possibility to take into account different input factors. The main concept of NN is to obtain some optimal coefficients, that are define functionality of NN, by learning on special examples.

Scanning temperature measurements in the lower troposphere by rotational Raman lidar


M. Radlach, A. Behrendt, S. Pal, A. Riede, V. Wulfmeyer, Univ. Hohenheim (Germany) For a comprehensive investigation of atmospheric processes, scanning lidar systems are highly beneficial because they provide volume data of atmospheric key parameters with high accuracy and resolution. At University of Hohenheim we have developed the first scanning rotational Raman lidar. This instrument transmits laser radiation at 355 nm. Measured parameters are temperature, particle extinction coefficient, and particle backscatter coefficient. The system is optimized for temperature measurements in the planetary boundary layer and lower free troposphere but shows also high performance up to the middle stratosphere. It will be deployed from June to August in the field campaign COPS (Convective and OrographicallyInduced Precipitation Study) on top of Hornisgrinde, the highest peak in the Northern Black Forest, at an elevation of 1161 m above sea level. First results will be presented at the conference. Due to a number of technical improvements, much better performance of the measurements can be expected for this campaign compared to previous deployments.

6750-26, Session 5
GPS calibrated multiwavelength water vapor Raman lidar measurements to assess urban aerosol hygroscopicity
B. M. Gross, D. V. Vladutescu, Y. Wu, F. Moshary, S. Ahmed, City College/CUNY (USA) Aerosols have a major influence on the climate, human health and chemical processes in the atmosphere and their interaction with water vapor is critical to the understanding of CCN formation. Quantifying the formation of CCN nuclei requires a fundamental understanding of the hygroscopic properties of aerosols. In most traditional studies, ground based nephelometer readings are compared directly to RH at the surface but unfortunately, the RH measurements at the surface do not reach the saturation regime where the largest hygroscopic affects are observed. On the other hand, observations using simultaneous backscatter and water vapor raman lidar offers the possibility of probing the simultaneous aerosol backscatter and RH profiles over a much larger dynamic range of RH. This is accomplished by probing stable homogeneous air masses below low altitude stratiform clouds. Besides the homogeneous nature of the airmass, determination of the aerosol structure to just below cloud base allows us to efficiently probe the high RH humidity regime. In support of our hygroscopic measurements, we discuss the need of careful calibration procedures for both the aerosol and water vapor profiles. The calibration algorithm we ultimately use for the water vapor measurements are twilight measurements where water vapor radiosonde data, are combined with total water vapor obtained from a GPS MET station. These radiosondes are then time correlated with independent near surface RH measurements to address any bias issues that may occur due to imperfect calibration due to lidar overlap issues and limitations in seeing the water vapor at high altitudes. The results of our calibrations are then presented to illustrate both short term and long term stability performance Finally, it should be noted that in previous lidar based hygroscopic studies, the simultaneous profiles were obtained at a single wavelength. Unfortunately, the complexity of the hygroscopic properties of aerosols makes it very difficult to pick between hygroscopic models with a single wavelength

6750-24, Session 5
Water vapour fluxes measured by collocated airborne DIAL and wind lidar during COPS 2007
C. Kiemle, S. Rahm, M. Wirth, A. Schaefler, A. Fix, M. Weissmann, G. Ehret, Deutsches Zentrum fr Luft- und Raumfahrt e.V. (Germany) Profiles of tropospheric water vapour and wind, as well as horizontal and vertical water vapour fluxes over Southwest Germany in summer 2007 will be presented. They are measured with differential absorption lidar (DIAL) and heterodyne detection wind lidar collocated onboard the DLR Falcon research aircraft during the Convective and Orographically-induced Precipitation Study (COPS). This international field campaign aims at refining observational and modelling efforts to improve the forecast skill of convective precipitation over complex terrain in the summer season. Both horizontal (through advection) and vertical (latent heat) fluxes of water vapour play an eminent role in this context. Three main objectives and mission scenarios for the deployment of this unique lidar payload were defined: (1) map the preconvective mesoscale wind and water vapour heterogeneity in a situation when deep convection and heavy precipitation is forecasted, by flying a grid or box pattern; (2) measure latent heat fluxes over the Rhine valley and the Black Forest mountains when convection initiation is predicted to be triggered by the orography and the surface humidity; (3) perform targeted upstream measurements over SW-Europe for the quantification of humidity advection into the COPS area. While the DIAL is permanently nadir-viewing, the wind lidar can either be operated in scanning mode at 20 degrees off-nadir

168

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6750: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing
measurement. To improve on this situation, we explore, in this paper, the nature and variability of urban aerosol hygroscopic properties using multi-wavelength Raman lidar measurements at 355nm, as well as backscatter measurements at 532nm and 1064nm. The addition of these longer wavelength channels allow us to more accurately validate the homogeneity of the aerosol layer as well as provide additional multiwavelength information that can be used to validate and modify the aerosol models underlying the hygroscopic trends observed in the Raman Channel. The lidar measurements are complemented by nephelometer measurements where the lidar overlap function is not reliable. In particular, we show that aerosols backtracked to different sources (bio-mass, maritime due to sea breeze) have different hygroscopic behavior which we compare to possible theoretical models. mapping. Lastly, we will discuss instruments and satellites for meteorology and earth science missions.

6750-29, Session 6
A laser sounder for measuring atmospheric trace gases from space
H. Riris, J. B. Abshire, NASA Goddard Space Flight Ctr. (USA); G. R. Allan, Sigma Space Corp. (USA); J. F. Burris, J. R. Chen, NASA Goddard Space Flight Ctr. (USA); P. Jian, Science Systems and Applications, Inc. (USA); S. R. Kawa, NASA Goddard Space Flight Ctr. (USA); J. Mao, Science Systems and Applications, Inc. (USA); M. A. Krainak, M. A. Stephen, X. Sun, E. L. Wilson, NASA Goddard Space Flight Ctr. (USA) We report progress in developing a laser technique for the remote measurement of the trace gases from space. Our initial goal is to demonstrate a lidar technique and instrument technology that will permit measurements of CO2 column abundance in the lower troposphere at 1572 nm. Our approach uses the differential absorption LIDAR technique and uses several tunable fiber laser transmitters to permit simultaneous measurement of the absorption from CO2 and O2 extinction. The energy of the laser echoes reflected from land and water surfaces The lasers are tuned on- and off- a selected CO2 line near 1572 nm and a selected O2 line near 768 nm in the Oxygen A band at kHz rates. The receiver uses a 1-m diameter telescope and photon counting detectors for both channels and measures the energies of the laser echoes from the surface along with scattering from any clouds and aerosols in the path. The extinction and column densities for the CO2 and O2 gases are estimated from the ratio of the on and off line signals via the differential optical absorption technique. The lasers are pulsed, which allows the receiver to gate out signals that were scattered by aerosols in the path. Our final goal is to extend the technique into the mid-infrared wavelengths and develop a practical instrument for active trace gas measurements from space.

6750-27, Session 5
Errors estimation in water vapor mixing ratio from a Raman lidar
M. Adam, Howard Univ. (USA); D. N. Whiteman, NASA Goddard Space Flight Ctr. (USA); D. D. Venable, Howard Univ. (USA) 5. The custom procedure to estimate the errors in calculating the Water Vapor Mixing Ratio from the water vapor Raman channel assumes only the random errors (following Poisson statistics). However, in general, it is known that any system is susceptible to systematic errors. The purpose of this study is to estimate other errors besides random ones and to account for them during error propagation. Particularly, we refer to the systematic errors due to the following processes: the gluing, the incomplete overlap correction and the calibration. The data analyzed are taken with Howard University Raman Lidar (HURL) during WAVES 2006 field campaign. The HURL system operates at the third harmonic of an Nd:YAG laser and acquires data within three channels, at 354.7 nm (elastic backscatter), 386.7 nm and 407.5 nm (Raman scattering from nitrogen molecules and water vapor molecules). The laser beam and telescope divergences are 50 rad and 250 rad respectively. The data acquisition is achieved with Licel Optical Transient Recorders which allow both photon counting and analog acquisition. The WAVES (Water Vapor Variability - Satellite/Sondes) 2006 field campaign took place at the Howard University Research Campus in Beltsville, MD, USA from July 7 to August 10. The field campaign was intended to provide quality measurements of water vapor and ozone for comparison with AURA satellite retrievals and to quantify the air quality. The operations include intensive observations by multiple radiosonde/ozonesonde sensors and several lidar systems during overpasses of the AURA satellite. Continuous measurements are taken by a 31m instrumented tower, various broad-band and spectral radiometers, microwave radiometer, Doppler C-band radar, wind profiler, sun photometer, and GPS system.

6750-30, Session 6
Measurement of carbon dioxide column via space-borne laser absorption
W. S. Heaps, NASA Goddard Space Flight Ctr. (USA) In order to better understand the budget of carbon dioxide in the Earths atmosphere it is necessary to develop a high precision understanding of the globalcarbon dioxide column. In order to uncover the missing sink that is responsible for the large discrepancies in the budget as we presently understand it calculation has indicated that measurement accuracy on the order of 1 ppm is necessary. This represents a precision on the order of .25% for these column measurements. No species has ever been measured from space at such a precision. In recognition of the importance of understanding the CO2 budget in order to evaluate its impact on global warming the National Research Council in its decadal survey report to NASA recommended planning for a laser based total CO2 mapping mission in the near future. The extreme measurement accuracy requirements on this mission places very strong requirements on the laser system used for the measurement. This work presents an analysis of the characteristics necessary in a laser system used to make this measurement. Consideration is given to the temperature dependence, pressure broadening, and pressure shift of the CO2 lines themselves and how these impact the laser linewidth, tuning, and tuning stability of the measurement system. Several systems for meeting these requirements that are under investigation at various institutions in the US as well as Europe and Japan will be discussed.

6750-28, Session 6
Recent advances in technology and instruments for Earth, planetary and space exploration
T. J. Carrig, Lockheed Martin Coherent Technologies (USA); J. C. Petheram, R. S. Price, Lockheed Martin Corp. (USA); S. W. Henderson, Lockheed Martin Coherent Technologies (USA); J. H. Crocker, Lockheed Martin Corp. (USA) Lockheed Martin Space Systems Company has recently formed a new entity, Sensing and Exploration Systems (S&ES), to focus on the development of optical payloads and spacecraft for remote sensing and space exploration. This initiative seeks to foster the development of payloads that combine passive and active systems and their integration into spacecraft. S&ES will also continue to support the development of next-generation technology and sensors for ground and airborne laser radar sensors. In this talk we will describe current efforts to develop key enabling technology such as space-qualified pulsed diode arrays, lasers, heat pipes, signal processors and energy systems. We will also discuss recent progress in the development of innovative high-bandwidth ladar sensors for high-resolution three-dimensional imaging, multi-wavelength differential absorption spectroscopy systems for discrimination of atmospheric gases such as carbon dioxide and methane, polarization sensitive laser radar systems for aerosol and hard target discrimination, and coherent Doppler lidars for wind-field

6750-31, Session 6
Contamination control of future lidar remote sensing satellites: a recipe for mission success
C. R. Maag, T&M Engineering (USA); C. S. Weimer, Ball

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

169

Conference 6750: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing
Aerospace & Technologies Corp. (USA); M. S. Cisewski, NASA Langley Research Ctr. (USA); F. E. Hovis, Fibertek, Inc. (USA) Past experience has demonstrated that contamination can be a significant factor in whether a spacecraft or instrument can successfully meet its mission objectives. Contamination will be a continuing problem with future payloads, especially those employing sophisticated laser optics, those cooled to cryogenic temperatures or those scheduled for long lifetimes in space. Contamination can degrade the power output of lasers, distort images due to altered transmission of optical components, and degrade the properties of optical surfaces and coatings. The major sources of contamination for any space asset are ground testing and operations; payload launch, ascent, and deployment; outgassing of materials in vacuum, impingement of thruster effluents, venting of internal gases, and interaction with the space radiation environment. The key issues in contamination control are material selection and pre-flight conditioning of these materials, proper configuration and subsystem design, and a contamination budget, which is used during the assembly, integration and testing phases. In order to establish a degree of confidence that a spacecraft or instrument will not suffer serious contamination, many safeguards and procedures need to be instituted. The Cloud -Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) satellite and scientific payload underwent an intense contamination control program in order to eliminate the possibility of a contamination-related laser failure or degraded SNR due to contamination related transmittance losses. Successfully launched on April 28, 2006 on-board a Boeing Delta II rocket from the Vandenberg Air Force Base, CALIPSO has been functioning well with no signs of unexpected contamination related degradation after over six months of near continuous operation. This paper will discuss the CALIPSO contamination control program with emphasis on the system level contamination budget, unique material conditioning processes and testing, laser transmitter contamination control, and integration and testing contamination and control regimes. Based on the teams experience, the authors will address improved contamination control procedures for future Lidar missions and discuss the special challenges associated with active measurements at 355 nm and lower wavelengths. instantaneously the atmospheric turbulence effect. For space application the disturbances characteristic time is really lower, therefore a control systems of about ten hertz bandpass is sufficient; the on orbit error budget is composed by the mirror manufacturing and deployment errors and the thermo-elastic deformations. These decoupled phenomena can be recovered by few low power actuators. The overall preliminary telescope design is presented with a special attention to the optical performances analysis. The mechanical layout with the explanation of the deployment and baffling technique is shown; the complete satellite thermoelastic analysis, mapping the primary mirror deformation due to the thermal loads, is presented; the control system architecture is explained and the optical design including the angular and spatial resolution, effective optical aperture and radiometric transmission, optical alignment tolerances, straylight and baffling is deeply discussed. Finally an overview of different mission profiles that this technology can satisfy is presented; the imaging performances can be achieved using the shown technology tuning the surface control to higher performances.

6750-33, Session 6
Lightweight active controlled primary mirror technology demonstrator
M. Olivier, A. Bursi, Carlo Gavazzi Space SpA (Italy); P. Mazzinghi, D. Ferruzzi, V. Bratina, Istituto Nazionale di Ottica Applicata (Italy); P. Salinari, F. Lisi, Osservatorio Astrofisico di Arcetri (Italy); D. Gallieni, ADS International s.r.l. (Italy); R. Biasi, Micro Photon Devices s.r.l. (Italy); J. P. N. Pereira do Carmo, European Space Agency (Netherlands) This paper describes the design, manufacturing and test of a ground demonstrator of an innovative technology able to realize lightweight active controlled space-borne telescope mirror. In particular this analysis is derived from a large aperture space telescope for advanced Lidar application, but it can be applicable to every lightweight mirror. For a space-borne telescope the mirror weight is a fundamental parameter to be minimized (less than 15 Kg/m2), maximizing the optical performances (more than &#61548;/3 optical quality). In order to guarantee this results the best solution, selected by a detailed trade off analysis, is a thin glass primary mirror coupled to a stiffening CFRP (Carbon Fiber Reinforced Plastic) panel with a surface active control system. A preliminary design of this lightweight structure has highlighted the critical areas that have been deeply analyzed by the ground demonstrator: the 1 mm thick mirror survivability on launch and the actuator functional performances with low power consumption. To preserve the mirror glass the Electrostatic Locking technique has been developed; in particular the electrostatic force has been used for attracting the glass (suitably coated with a conductive coating on its back face) toward the back plane, coated with a conductive layer and with a thin insulating layer of high dielectric constant. Appling 300 V tension the mirror was practically glued to the CFRP back-plane, withstanding a vibration test simulating the launch environment. The active optics is a technique already widely been used since several years in ground based telescopes to correct optical errors due to slowly varying effects (orientation changes in the gravity field, thermal effects). It consists of : a metrology system, normally a wave front sensor (WFS), a control algorithm and a system of actuators to slightly deform the primary mirror and/ or displace the secondary mirror. Such configuration is a typical closed-loop control system that computes and applies the corrections by means of the actuators to maintain the optical errors within the required range, starting from a reference. These actuators types are designed and tested in order to guarantee satisfactory performances in terms of stroke, force and power consumption. The realized and tested ground demonstrator is a squared CFRP structure with a flat mirror on the upper face and with an active actuator installed. The test campaign has demonstrated the technology feasibility and robustness, supporting the next step toward the large and flat surface with several actuators.

6750-32, Session 6
Deployable, lightweight and large-aperture spaceborne telescope for lidar-based Earth observations
P. Mazzinghi, V. Bratina, D. Ferruzzi, L. Gambicorti, F. Simonetti, A. Zuccaro, Istituto Nazionale di Ottica Applicata (Italy); P. Salinari, F. Lisi, Osservatorio Astrofisico di Arcetri (Italy); M. Olivier, A. Bursi, Carlo Gavazzi Space SpA (Italy); D. Gallieni, ADS International s.r.l. (Italy); R. Biasi, Micro Photon Devices s.r.l. (Italy); J. P. N. Pereira do Carmo, European Space Agency (Netherlands) Abstract This paper describes innovative approach and technology for advanced Lidar applications using large aperture telescopes. The telescope collecting area is a key parameter for a Lidar instrument because the transmitter laser power is inversely proportional to the receiver aperture; therefore a large aperture telescopes allows to decrease the laser power necessary for a given retrieved signal. The study aim is to monitor the atmospheric water vapor distribution at 935.5 nm wavelength using a DIAL (Different Absorption Lidar) technique. This mission profile is well adapted to an ESA Earth Observation LEO (Low Earth Orbit) small satellite scenario, therefore the considered launcher vehicle can be the ROCKOT, a medium class launcher. Using such launch vehicle class, in order to have a large primary mirror size, more than 7 m2 , with an optical quality in operation at least of lambda/ 3, a deployable thin primary mirror coupled with an active control system was selected by a deep technological trade-off analysis. This technology is used for ground-based telescope for the adaptive thin-glass secondary mirrors; these mirrors are actuated by a large number of electromechanical actuators that use high frequency control (several KHz) to correct

170

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6750: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing

6750-44, Poster Session


Simulation studies of the Raman measurements of NOx and H2O using a space-borne lidar
M. V. Satyanarayana, S. R. Radhakrishnan, B. Presennakumar, Vikram Sarabhai Space Ctr. (India) The estimation of the relative contribution of the green house gases and pollutants in the Earth atmosphere on global climate requires appropriate experimental investigations and continuous monitoring. The quantification of the human activities on the radiative budget of the earths atmosphere is of crucial importance in the present day context. NOX is one of the important constituents of the atmosphere due to its role in the formation of nitrate aerosols and also its reaction with water vapor leading to acid rain. The role of NOX on the radiative budget depends up on the altitude of its presents. It will have both cooling and warming effects depending up on its location and it directly participates in the formation and depletion of Ozone. Due to its unique physical and chemical properties, H2O is crucial for the earth system both on weather and global climate related processes. The concentration of NOX and H2O are very important in climate modeling and short time weather forecasting. Ground based measurements of NOX and H2O are well established using the powerful lidar laser remote sensing methods. A Space based lidar consisting of Raman channels can provide measurement of green house gases and pollutants globally. Space Physics Laboratory, VSSC, Trivandrum programmed the design, development and launching of a Space Borne Lidar (SBL) for the measurement of the vertical profiles of NOX and H2O is currently in progress. The Space Borne lidar system will be launched in to a polar orbit by the ISROs Polar Satellite Launch Vehicle (PSLV). The lidar instrument consist Nd: YAG laser transmitting at three wavelengths, 700mm telescope, nine optical channels with PMT, APD etc will be housed in a satellite at a nominal altitude of 600Km with a period of approximately 90 minutes providing global coverage. Raman scattered signals from NO2 (571.1nm) and H2O (407nm) correspond to the exciting wavelength at 532nm and 355nm are monitored by the respective Raman channels. In this paper we present the simulation studies of water vapor, NOX measurements using the Space Borne lidar.

C. Boitel, M. Haeffelin, Ecole Polytechnique (France) The SIRTA (Site instrumental de Recherche par Tldtection Atmosphrique) observatory is located at Palaiseau, France (25km south of Paris). The SIRTA was created in 1999 by the Institut Pierre Simon Laplace and cole Polytechnique to conduct research programs lying on complementarity between passive radiometric instruments and active sounding by lidar. The SIRTA benefits from state-of-art lidars developed in research labs or manufactured radiometers in order to improve the knowledge of radiative and dynamic processes in the atmosphere as well as complex interactions between clouds and aerosols. The objective is to provide significant contribution to climate change studies, impact of urban pollution, aerosols and clouds interactions. The SIRTA contributes to many international of instrument networks such as the European Aerosol Research Lidar Network (EARLINET), the Baseline Surface Radiation Network (BSRN), the European pilot network for observing cloud profiles (Cloudnet) and the Aerosol Robotic network (AERONET). Two ground platforms, a wooden tower, a roof platform and a building (from where the lidar operates) are the main facilities of SIRTA. The project team is composed of six persons to ensure the station operations from instrument deployment, maintenance, to data transfer and preliminary data analysis. The SIRTA infrastructure enables 1) to conduct many research activities that involve the cloud and aerosol lidar. Some of them will be discussed: the development of the STRAT (Structure of the Atmosphere) algorithm dedicated to automatically discriminate atmospheric layers and retrieve geophysical parameters from lidar profiles, 2) the CALIPSO validation, 3) the support in the EARLINET network, 4) the instrumental synergy to improve the lidar profiles inversion.

6750-47, Poster Session


Improved methodology for the retrieval of the particulate extinction coefficient and backscatter-to-extinction ratio from lidar multiangle measurements
V. A. Kovalev, C. E. Wold, J. O. Newton, W. M. Hao, USDA Forest Service (USA) A new practical methodology, which utilizes the Kano-Hamilton multiangle technique for processing elastic-lidar data, is considered. The goal of this study is to improve measurement accuracy when determining vertical profiles of the particulate extinction coefficient and the backscatter-to-extinction ratio from elastic lidar data. The improvement is achieved by the implementation of new data processing procedures. First, optimal minimal and maximal ranges for the lidar signal are determined such that the remaining signal multiplicative and additive distortions do not produce large errors in the measurement result. This is achieved by using new criteria that allow minimizing influence of these systematic distortions. Second, a special procedure is applied to obtain the extinction coefficient profile while eliminating the conventional procedure of numerical differentiation. The new inversion procedure is based on determining the best values of the backscatter-toextinction ratio over extended ranges; the lengths and boundaries of these are determined through the preliminary estimated uncertainty boundary in the retrieved optical depth. The inversion method was used to process the experimental data obtained in the vicinity of large wildfires with the Fire Sciences Laboratory lidar. Examples of the inversion results are presented which illustrate the specifics and prospects of the new data-processing methodology.

6750-45, Poster Session


Retrieval of aerosol microphysical parameter by inversion of multi-wavelength Raman lidar data collected at Gwangju and Anmyeon Island, Korea
Y. Noh, Y. J. Kim, Gwangju Institute of Science and Technology (South Korea); D. Mller, Leibniz-Institut fr Troposphrenforschung e.V. (Germany) Information on vertical distribution and optical properties of atmospheric aerosols is quite important for estimating the aerosols effects on atmospheric environment and radiative forcing. Vertical profiles of microphysical properties of atmospheric aerosol were obtained using inversion algorithm from optical data of GIST/ADEMRC multi-wavelength Raman lidar system. Inversion algorithm was applied for data collected on 14 June 2004 at Gwangju (35.10oN, 126.53oE) and 27 May 2005 at Anmyeon island (36.32oN, 126.19oE). According to NOAA/HYSPLIT backstrajectory analysis of air mass, Siberian forest fire smoke and local haze particles were dominant above and inside the planetary boundary layer (PBL), respectively on 14 June 2004. Inversion of lidar optical data resulted in particle effective radii ~ 0.32 m and single scattering albedo ~ 0.97 inside PBL, and ~ 0.27 m and ~ 0.92 at 532 nm above PBL. In the case of 27 May 2005, biomass burning from East China greatly affected the aerosol plume, resulting in particle effective radii ~ 0.23 m, single scattering albedo ~ 0.92 at 532 nm. Additionally, the lidar inversion results of AOD, effective radius, and single-scattering albedo were compared with those of colocated Sun/sky radiometer.

6750-48, Poster Session


Saharan dust event over Bucharest observed by an elastic backscatter lidar
C. L. Talianu, A. V. Nemuc, E. D. Carstea, L. Belegante, National Institute of Research & Development for Optoelectronics (Romania) A lidar measurements campaign took place in northwestern part of Bucharest, ROMEXPO site (44.25 N, 26.05 E) during a week in April 2006 and was intended for aerosol loading characteristics over the urban area. An event of long-range Saharan dust transport to Eastern Europe (Romania) observed

6750-46, Poster Session


SIRTA, a multi-sensor station for clouds and aerosols characterization in the atmosphere: infrastructure, objective and prospective
C. M. Pietras, Y. Morille, T. G. Elias, B. Romand, F. Lapouge,

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

171

Conference 6750: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing
during this time is presented in here. We have used an elastic backscattering lidar, based on a Nd:YAG laser, working at two wavelengths (1064 and 532 nm). It can detect in real time aerosols density profiles up to 10 Km high with a spatial resolution of 6 m. Origin of lidar sampled air masses arriving at various heights over Bucharest have been determined by the analytical backtrajectories from NOAA Hysplit model. Saharan dust layers reached the southern part of Romania predominantly by cyclonic circulation due to the strong trough observed at all the levels from a cyclonic system located in northwestern part of Africa. Analysis of cloud cover and dust load was estimated by the Dust Regional Atmospheric Modeling (Dream model). The dust event presented highlights how the synergy of Lidar data together with 3-D back trajectories analysis and model calculations can improve our ability to determine accurately the source of high aerosol loading.

6750-51, Poster Session


Development of the 1.6 m carbon dioxide DIAL for tropospheric CO2 study
D. Sakaizawa, C. Nagasawa, Tokyo Metropolitan Univ. (Japan); T. Nagai, Meteorological Research Institute (Japan); M. Abo, Y. Shibata, Tokyo Metropolitan Univ. (Japan); M. Nakazato, Meteorological Research Institute (Japan) The balance of carbon dioxide budget has been perturbed by human activities since the Industrial Revolution. Carbon dioxide vertical profile has been required to understand the fundamental processes and improve the future rate of its increase in the atmosphere. A differential absorption lidar (DIAL) system has being developed for toropospheric CO2 studies. Transmitter wavelength is around 1.572 m where there is one of CO2 absorption bans of 30012-00001. The needed uncertainty of CO2 measurements using a differential absorption lidar is better than 1% up to 10 km. It is especially significant that the laser transmitter is capable high energy and repetition rate around 1.572 m CO2 absorption band. The output energy of our optical parametric oscillator (OPO) using a KTP crystal is around 20mJ because of the small nonlinear coefficient and phase matching method. The OPO using a large aperture QPM device is improved higher efficiency and energy than the KTP-OPO. We present the integrated laser system in this conference.

6750-49, Poster Session


Highly compact backscatter lidar for unattended measurements in the planetary boundary layer and lower troposphere
R. Matthey, V. Mitev, Observatoire Cantonal de Neuchtel (Switzerland) Increased needs to monitor aerosol load and stratification in the atmosphere lead to organisation of aerosol lidar networks. One attractive option to arrange measurements at remote sites with under-developed infrastructure in such networks relies on compact lidars with minimum requirements for maintenance and capable to operates in unattended way. A highly compact backscatter lidar is put in operation in Observatory of Neuchatel in the frame of the European EARLINET network to test and verify the practical problems with such instrument, as well as the achievable performances. The instrument application is expected to be optimal for measurements of the planetary boundary layer (PBL) and lower troposphere. The lidar was previously deployed during airborne field campaigns and is assembled in a single protected box, transportable in a small car. We will describe the lidar and discuss results from the first long-term test operation, carried out in our premises. The measurements will cover PBL diurnal cycles in various seasons. Time development of the aerosol mixed layer height as well as selected aerosol backscatter coefficient profiles will be reported.

6750-52, Poster Session


Bright onyx: a tunable, mid-IR, DIAL lidar for chemical and biological sensing
E. J. Burlbaw, A. R. Geiger, J. Ding, C. Ting, Y. Huang, C. J. Urbina, M. Gutierrez, Akamai Physics, Inc. (USA); R. D. Richmond, Air Force Research Lab. (USA) Akamai Physics, Inc. has developed a tunable, pulsed, mid-IR, DIAL LIDAR for chemical and biological detection. The Bright Onyx laser is pulsed and tunable at 4-kHz over the range of 3.1 to 3.7 microns. The receiver has an 8-inch primary focused onto a 2-mm diameter InSb cryogenic detector. This paper will describe the system, provide performance data, and show comparisons against several chemicals of environmental, industrial, and defense interest. (AFRL clearance pending; expected within 3 weeks)

6750-53, Poster Session


Sounding of the environment by means of the un-impulse of the low power continuous source
J. Polkanov, B.I. Stepanov Institute of Physics (Belarus) At the moment laser systems of sounding of an atmosphere in the majority assume use of powerful pulse lasers, and registration of an absent-minded signal from environment occurs to the certain step of digitization corresponding to the required spatial sanction. And the increase in range of sounding demands essentially disproportionate increase in capacity of radiation and causes difficultly taken into account effects of repeated dispersion and demands complex methods of expansion of a dynamic range of the reception devices. Qualitatively other approach which allows to avoid many byeffects is offered. He is based on use a low-power source of radiation (for example, white light), at creation of a un-radiation impulse, interruption for a while commensurable with duration of the used pulse sources of radiation. Transformation of an absent-minded signal registered back to a digital kind can be spent by usual digital systems, as discrete readout of a signal of identical duration. However for increase of accuracy of restoration of characteristics of the environment on a registered signal it is offered to spend restoration of average values of such characteristics on sites commensurable with all length of a line of sounding, and about characteristics of environment with the required sanction to judge on average values of two sites differing on the required step of digitization. According to such algorithm creation of recording system with the corresponding organization processing a coming signal is offered. The carried out ratings [1] have shown that to become possible to spend measurements with accuracy ~ 1...10 % in a zone of unitary dispersion of a signal at a wide range of atmospheric

6750-50, Poster Session


Selection of an averaging technique by simulation study of a DIAL system for toxic agents monitoring
J. P. Dudeja, M. K. Jindal, S. Veerabuthiran, Missile Facilities Laser Science and Technology Ctr. (India) Differential Absorption Lidar (DIAL) is a very effective technique for standoff detection of various toxic agents in the atmosphere. The Lidar backscattered signal received usually has poor signal to noise (SNR) ratio. In order to improve the SNR, statistical averaging over a number of laser pulses is employed. The aim of the present work is to select a particular statistical averaging technique, which is most suitable in removing the noise in Lidar return signals. The DIAL system considered here uses laser transmitters based on OPO based (2-5 mm) and TEA CO2 (911mm) lasers. Eight commonly used chemical warfare agents including five nerve agents and three blister agents have been considered here as examples of toxic agents. A Graphical User Interface (GUI) software has been developed in LabVIEW to simulate return signals mixed with the expected noise levels. A toxic agent cloud with a given thickness and concentration has been assumed to be detected in the ambient atmospheric conditions at various ranges up to 5 Km. Data for 200 pulses per agent was stored in the computer memory. Various known statistical averaging techniques were used and number concentrations of particular agent have been computed and compared with ideal Lidar return signal values. This exercise was repeated for all the eight agents and based on the results obtained; the most suitable averaging technique has been selected.

172

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6750: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing
conditions ( = 10~2....1,0 km - 1) and capacities of radiation of the source, not surpassing 10-20 BT. And difficulties of maintenance of a linear operating mode reception of the photoelectronic receiver are excluded and there is an opportunity to lead measurement with the specified accuracy in day time conditions. Such approach allows to increase accuracy of measurements at high and application in various areas can find, beginning from lidar, radar systems and finishing biological and medical devices. The literature: Polkanov, Iu. A. The possibility of improving the accuracy of the measurement of a light-scattering signal. Radiotekhnika i Elektronika (ISSN 0033-8494), vol. 33, Dec. 1988, p. 2599-2603. http://adsabs.harvard.edu/abs/1988RaEl...33.2599P data to be used for operational evaluation and validation of weather and climate numerical prediction models and satellite products. CNR-IMAA advanced ground-based observational facility, located in Tito Scalo, Italy, (760 m a.s.l., 40.60 N, 15.72 E) is an operational site for weather and climate studies. The facility is equipped with a comprehensive set of remote-sensing and insitu equipment to operationally characterize key variables in the study of the weather and climate, such as aerosol, water vapor, clouds, temperature and radiation. The facility includes a multi-wavelength lidar, a water vapor Raman lidar, a DIAL for pollutant monitoring, a laser ceilometer, a microwave profiler, a sun-photometer, a surface radiation station and radiosounding systems. The retrieved information are steadily compared and cross-validated. The synergism of all the instruments operational at the facility allows a detailed description of the optical and microphysical properties of aerosol and clouds. The presented paper provides an overview on several integration approaches developed and implemented at the CNR-IMAA for the optimization of aerosol and cloud profiling.

6750-54, Poster Session


Optimization of dispersion models used to locate gas sources
L. C. Thomson, Univ. of Glasgow (United Kingdom); W. Hirst, Shell International Exploration and Production BV (Netherlands); M. J. Padgett, Univ. of Glasgow (United Kingdom) Several novel oil and gas prospecting techniques based on new measurement technologies have been developed in recent years. The technique we describe here is based on the fact that all oil and gas reservoirs gradually leak some of their contents to the atmosphere. We use an ultra-sensitive absorption spectroscopy system to measure ethane concentrations in the atmosphere at sub- parts per billion levels. We combine concentration measurements from numerous locations with associated wind velocity data to provide a data set that allows us to locate and quantify the flux sources responsible. Many dispersion models exist for describing how gas disperses from known sources. However, locating those sources from measured concentrations is an ambiguous problem - a small source nearby cannot be distinguished from a larger source further away. This is an inverse problem: we can calculate the concentrations resulting from a known pattern of sources, but not the pattern of sources responsible for the measured concentrations. We use a random search algorithm with annealing methods in order to solve this problem. Although many dispersion models exist to describe how gas disperses we have found that they do not accurately describe real data. Having used a random search to successfully locate the unknown source from the known measurements we now present details of how we use a similar algorithm to optimise a chosen dispersion model based on real data. Such models are used in many areas of remote sensing and optimisation in this way improves their description of real data.

6750-57, Poster Session


Scanning water vapor DIAL based on a highpower Ti:sapphire laser: first experimental results
M. Schiller, V. Wulfmeyer, A. Riede, A. Behrendt, H. Bauer, G. Wagner, Univ. Hohenheim (Germany); U. Wandinger, D. Althausen, Leibniz-Institut fr Troposphrenforschung e.V. (Germany); A. Fix, M. Wirth, G. Ehret, Deutsches Zentrum fr Luft- und Raumfahrt e.V. (Germany) We developed a high-power Ti:Sapphire (TISA) laser transmitter for a 3D scanning ground-based water-vapor differential absorption lidar (DIAL). A water-vapor DIAL transmitter must fulfill very stringent requirements with respect to spectral purity, frequency stability, bandwidth, and beam quality. Our approach is an injection-seeded, gain-switched TISA ring laser pumped from both sides with 532-nm radiation of a new high-power, diode-pumped frequency-doubled Nd:YAG laser. The Nd:YAG operates at a repetition rate of 250 Hz with 45 W laser power at 532 nm. A dynamically stable TISA resonator design, based on a unidirectional ring cavity, was selected for single-frequency operation to avoid spatial hole burning. To fulfill the spectral and stability requirements, a stabilized single-mode external cavity diode laser as injection seeder is used. The TISA laser is optimized with respect to average power and frequency stability by in-cavity astigmatism correction and a Pound-Drever-Hall stabilization technique, respectively. Single-frequency operation with a frequency stability of 50 MHz, a beam quality of M2 < 2, and an average power of more than 7 W have been achieved, simultaneously. The TISA operates in the 820-nm region where a number of absorption lines are present, which are suitable for groundbased water-vapor measurements. This TISA is part of the scanning DIAL with a telescope diameter of 80 cm, which will participate in the WWRP RDP COPS field campaign in Germany this summer. Detailed specifications of the system as well as first measurement results will be presented at the conference.

6750-55, Poster Session


MicrodsPIC-based design for a lidar flexible unit
A. Lay-Ekuakille, Univ. degli Studi di Lecce (Italy) Flexible and programmable components have been playing a key role in many electronic systems. Lidar systems can be improved by means of component and electronic progress, especially in data collection, acquisition and processing. The objective of present paper is to show the design of an electronic unit capable of substituting Raman lidar - based discriminator and multichannel scaler. For this experiment, a Hamamatsu H7732P-01 photomultiplier has been used.

6750-58, Poster Session


Mid IR-fiber spectroscopy in 2-17m range (II)
V. G. Artiouchenko, A. Bocharnikov, G. Colquhoun, Fibre Photonics Ltd. (United Kingdom); C. A. Leach, Fibre Photonics Ltd (United Kingdom); V. Lobachov, T. Sakharova, D. Savitskij, General Physics Institute (Russia) The latest development in IR-fiber optics enables to expand spectral range of processspectroscopy from 2m up to 17m (5.000 to 600cm-1). Up to now fiber systems are already produced for broad variety of process-control applications using transmission, reflection, fluorescence and Ramanspectroscopy, but their spectral range is limited to transmission of silica fibers from 0.2 to 2.4m. Nowadays the longer Mid-IR wavelength range from 2 to 6-10m may be covered by Chalcogenide IR-glass CIR-fibers, while Polycrystalline PIRfibers from Silver Halides can cover the whole finger-print part of spectrum - the most informative on absorption bands

6750-56, Poster Session


CNR-IMAA integrated measurements for the characterization of the atmosphere
F. Madonna, A. Amodeo, A. Boselli, C. Cornacchia, G. DAmico, A. Giunta, L. Mona, G. Pappalardo, Consiglio Nazionale delle Ricerche (Italy) Integration of measurements performed using active and passive ground-based techniques is a powerful solution for the achievement of a synergic analysis of the atmospheric processes and to provide continuous high-resolution data. These data form the basis for the supply of sustained reference

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

173

Conference 6750: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing
of specific molecular vibrations (Fig.1). These specific bands are 100-1000 times more intensive and more narrow than the 2nd & 3rd harmonic bands at the shorter wavelengths <2m. Remote spectroscopy of chemical reaction, monitoring of exhaust or exhalation gases or invivo tissue analysis on molecular composition could be done by PIR-fiber probes coupling with different type of spectrometers - based on diffraction grating and detector arrays, FTIRinterferometers, IRLED, optical filters or using Tunable Diode or Quantum Cascade lasers. Design of immersion probes for liquids and touch probes for tissues is based on PIR-fiber coupling with a sealed ATRelements made from Diamond, ZnSe or other IR-materials. Choice of PIR-fibers in comparison with a brittle and toxic CIRfibers transmitting in 2-6m range is done they are very flexible, non-toxic, stable from 5 to 400K and transmit in much broader range from 3 to 17m. Transmission/Reflection T/R-probes for gases have been designed with IRfibers coupling of dual-pass optical cell with QCL or TDL-spectrometer. In addition to the review on IR-fiber coupled FTIR- and QCLspectroscopy systems the first FlexISpec spectrometer test data will be presented on the innovative prototype of (Fig.2) based on use of diffraction grating with IR-pyrodetector array and PIR-fiber probes. In difference with expensive FTIR- & QCLsystems FlexISpec developed is directed to more cost effective solution for real time process-control in chemical, petrochemical and pharmaceutical industries. Fig. 1. Attenuation spectra of IR-fibers Fig. 2. Spectrum of sun-flower oil measured with FlexISpec spectrometer in 5-7mm range with ATR-PIR-probe At CNR-IMAA, an aerosol lidar system is operative since May 2000 in the framework of EARLINET (European Aerosol Research Lidar Network) ), the first lidar network for tropospheric aerosol study on continental scale. Since August 2005, PEARL (Potenza EARlinet Lidar) system provides backscatter coefficient profiles at 1064 nm, and independent measurements of extinction and backscatter coefficient profiles at 355 and 532 nm. In addition, measurements of the vertical profiles of aerosol and cloud depolarization ratio at 532 nm are obtained by means of the detection of components of backscattered light polarized perpendicular and parallel to the direction of the linearly polarized transmitted laser beam. High quality multiwavelength measurements (3 backscatter + 2 extinction) plus depolarization measurements make PEARL a reference point for the validation of CALIPSO data products. A direct comparison with CALIPSO data can be carried out for depolarization ratio and aerosol backscatter at 532 and 1064 nm measurements. Furthermore PEARL aerosol extinction measurements at 532 nm and 355 nm and backscatter measurements at 355 nm can be used to improve the retrieval of aerosol backscatter coefficient from pure backscatter lidar. Since 14 June 2006, devoted measurements are performed at CNR-IMAA in coincidence of CALIPSO overpasses (maximum 80 km and 2 hours as spatial and temporal distance). First results of comparison between PEARL and CALIPSO observations will be shown. Acknowledgment The financial support of this work by the European Commission under grant RICA-025991 is gratefully acknowledged.

6750-36, Session 7
Comparisons with CALIPSO measurements for Thessaloniki EARLINET station
D. S. Balis, E. Giannakaki, Aristotle Univ. of Thessaloniki (Greece); V. Amiridis, National Observatory of Athens (Greece); T. Boubouloudis, Aristotle Univ. of Thessaloniki (Greece) Thessaloniki EARLINET station started correlative measurements for CALIPSO since July 2006. Each observation lasts a minimum of 1 hour centered on the overpass time, while longer records of measurements are performed for special case studies. In this study we present preliminary results of selected case studies such as, Saharan dust layer and forest fires. In order to investigate the optical properties of aerosols, we have used the synergetic use of a ground based and a satellite lidar instrument. A two-wavelength lidar system that combines Raman and elastic-backscatter observations performs measurements above Thessaloniki (40.50N, 22.9 0E, 50 m asl) in coincidence with CALIPSO overpasses. CALIPSO combines an innovative combination of an active lidar instrument with passive infrared and visible images to probe the vertical structure and properties of aerosols. For verification purposes, Saharan dust layers and biomass burning aerosols were related to air mass back-trajectory analysis and model calculations from the DREAM dust model in conjunction with satellite data analysis (ATSR World Fire Atlas, MODIS, SeaWiFS). The main objective of this study is to compare the resulting profiles of attenuated backscatter coefficient derived from CALIPSO with the backscatter profiles derived from our ground-based lidar. Moreover, a comparison of extinction coefficients for night measurements is included in this study.

6750-34, Session 7
EARLINET correlative measurements for CALIPSO
I. Mattis, Leibniz-Institut fr Troposphrenforschung e.V. (Germany); G. Pappalardo, Consiglio Nazionale delle Ricerche (Italy) The European Aerosol Research Lidar Network (EARLINET) was established in 2000 to derive a comprehensive, quantitative, and statistically significant data base for the aerosol distribution on the European scale. At present, EARLINET consists of 24 stations: 18 Raman lidar stations, including 7 multi-wavelength Raman lidar stations (3 backscatter plus 2 extinction coefficients) which are used to retrieve aerosol microphysical properties. EARLINET performs a rigorous quality assurance program for instruments and evaluation algorithms. All stations measure simultaneously on a predefined schedule at three dates per week to obtain unbiased data for climatological studies. Additional measurements are performed to address specifically important events like Saharan dust, forest fires, volcanic eruptions, and photochemical smog. Since June 2006 the first backscatter lidar is operational aboard the CALIPSO satellite. EARLINET represents an excellent tool to validate CALIPSO lidar data on a continental scale. Aerosol extinction and lidar ratio measurements provided by the network will be particularly important for that validation. The measurement strategy of EARLINET is as follows: Measurements are performed at all stations within 80km from the overpasses and additionally at the lidar station which is closest to the actually overpassed site. If a multi-wavelength Raman lidar station is overpassed then also the next closest 3+2 station performes a measurement. We present a statistical overview on the CALIPSO-overpass measurements, provide quantitative comparisons between EARLINET profiles and available CALIPSO products, and discuss synergy effects of both tools. The EARLINET-ASOS project is funded by the EC under grant RICA-025991.

6750-37, Session 7
CALIPSO correlative measurements at Napoli EARLINET station
X. Wang, M. Armenante, M. G. Frontoso, G. Pisani, N. Spinelli, Univ. degli Studi di Napoli Federico II (Italy) The Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) satellite had been launched in April 2006. Its main goal is to probe the vertical structure and to measure the properties of thin clouds and aerosols plume of the Earths atmosphere. In order to validate the CALIOP (CloudAerosol Lidar with Orthogonal Polarization) products, correlative measurements between CALIPSO and EARLINET stations have been planned in the framework of the Quid Pro Quo Validation Plan. At Napoli EARLINET station correlative measurement have been performed starting from the beginning of June. Results

6750-35, Session 7
First comparisons between CNR-IMAA multiwavelength Raman lidar measurements and CALIPSO measurements
L. Mona, A. Amodeo, G. DAmico, G. Pappalardo, Consiglio Nazionale delle Ricerche (Italy)

174

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

Conference 6750: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing
obtained both during daytime with elastic lidar and nighttime with Raman lidar operating at two wavelengths (355nm and 532nm) are shown. Case study of Saharan dust outbreaks and cirrus clouds captured during correlative measurement runs will be discussed in detail. Beside, since the ground track of the spacecraft is often several kilometres far from lidar station, we tried to evaluate the horizontally and/or temporal inhomogeneity of the atmosphere conditions during the overpasses Following a mission overview, this paper presents results of these pre-development activities related to the ATLID instrument that consist of designing, manufacturing and testing a functional representative model of the ATLID receiver critical units and laser source, of developing and assessing high-power pump laser diodes with extended lifetime and improved efficiency, and of demonstrating the performance of electron-multiplication CCDs devices.

6750-38, Session 7
ALADIN doppler wind lidar: recent advances
D. Morancais, F. Fabre, EADS Astrium (France); M. J. Endemann, A. J. F. Culoma, European Space Research and Technology Ctr. (Netherlands) The Atmospheric Laser Doppler Instrument (ALADIN) is the payload of the ADM-AEOLUS mission, which will make direct measurements of global wind fields. It will determine the wind velocity component normal to the satellite velocity vector. The instrument is a direct detection Doppler Lidar operating in the UV, which will be the first of its kind in space. ALADIN comprises a high energy laser and a direct detection receiver operating on aerosol and molecular backscatter signals in parallel. The laser is all solid-state, based on Nd-YAG technology and high power laser diodes. The detector is a silicon CCD whose architecture allows on-chip accumulation of the returns, providing photon counting performance. The 1.5 m diameter telescope is lightweigh, all made of silicon carbide. ALADIN is now in its final construction stage: the integration of the Flight Model is on-going. Most of the subsystems have been integrated; the payload performance and qualification test campaign will commence. This paper briefly describes the instrument design and provides insights on the development status and the results obtained so far. This regards in particular the receiver performance, the telescope development and the challenges of the laser. The Aeolus satellite is developed for the European Space Agency with Astrium Satellites as prime contractor for the satellite and the instrument.

6750-41, Session 7
NASA laser remote sensing technology needs for earth science in the next decade and beyond
D. M. Tratt, J. M. Neff, The Aerospace Corp. (USA); A. Valinia, NASA Goddard Space Flight Ctr. (USA) In late 2005 the NASA Earth Science Technology Office convened a working group to review decadal-term technology needs for Earth science active optical remote sensing objectives. The outcome from this effort will guide future NASA investments in laser remote sensing technologies. This presentation will summarize the working group findings and place them in context with the conclusions of the National Research Council assessment of Earth science needs, completed in 2007.

6750-42, Session 7
Spaceborne lidar data acquisition system and its challenges
M. V. Satyanarayana, B. Presennakumar, D. Rama Krishna Rao, S. R. Radhakrishnan, Vikram Sarabhai Space Ctr. (India) Ground based multi wavelength lidar systems are widely used for measurements on atmospheric aerosols and clouds to study their role in Earths radiation budget. But such measurements are at the best applicable to a small region only. A Space Borne Lidar will provide the required global measurements for climate modeling. A space based multi wavelength lidar system with three wavelengths is under development at Space Physics Laboratory, Vikram Sarabhai Space Centre, ISRO, Trivandrum, India. The lidar receiver electronics and data archival systems are designed for the three transmitting wavelengths of Nd:YAG and the Raman shifted wavelengths corresponding to some of the important trace gases/minor constituents. In order to cover the wide dynamic range of the backscattered signal intensities, a combined mode of data acquisition, which will have both analog and photon counting modes, is designed. The 532nm and 1064nm lines will be used for the altimeter and bathymetry measurements. The analog channel will digitize the 1064 nm/ 532nm signal with 22 bit resolution in addition to the fast photon counting system meant for weak backscattered signals. The SBL payload will be launched into a polar low earth orbit of around 630 KM by ISROs PSLV launch vehicle. Data will be stored in the onboard Solid State Recorder (SSR) and will be down linked to the dedicated ground station for offline analysis. This paper discusses in detail the data acquisition scheme taken up for implementation for the Space Borne Lidar system.

6750-40, Session 7
The ESA EarthCARE mission: results of the ATLID instrument pre-developments
Y. Durand, A. Hlire, J. Bzy, R. Meynart, European Space Research and Technology Ctr. (Netherlands) Due for launch in October 2012, the EarthCARE (Earth Clouds, Aerosols and Radiation Explorer) has been selected as ESAs sixth Earth Explorer Missions within its Living Planet Programme. Explorer missions of ESAs Living Planet programme [1]. Understanding both natural and anthropogenic climate change is of paramount importance in the 21st century. The predictions of future climate rely on global numerical models, all of which have limitations arising from the parameterisation of sub-grid scale processes. Clouds are a key feature of weather and play a crucial role in both the hydrological cycle and the energy budget of the Earth. Despite their importance there are still large deficiencies in the representation of clouds, aerosols and radiative transfer in atmospheric models and these are the source of large uncertainties in the predictions of climate change. Advances in the representation of the radiative effects of clouds and aerosols will be made possible with EarthCARE. Its payload aims at providing measurements, in a radiatively consistent manner, of the global distribution of vertical profiles of clouds and aerosol field characteristics. The EarthCARE payload is composed of four instruments: an Atmospheric backscatter Lidar, a Cloud Profiling Radar, a Multi-Spectral Imager and a Broad Band Radiometer. The EarthCARE mission is a cooperative mission with Japan (JAXA and NiCT), which will provide the Cloud Profiling Radar. ESA will provide the ground segment and the rest of the space segment including the Lidar, the imager and the broadband radiometer. ESA and JAXA have initiated pre-development activities to reduce technical and programmatic risks for the critical elements of the mission.

6750-43, Session 7
Studying regional aerosol events with the CALIPSO lidar and other a-train instruments
L. A. Hunt, M. T. Ferebee, NASA Langley Research Ctr. (USA) The A-Train satellite constellation offers unprecedented opportunities for the synergistic study of several important areas related to Earth and its atmosphere. With the launch of the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite (CALIPSO) and CloudSat in April 2006, five of the six planned satellites are now in place, making nearly coincident measurements of water, aerosols, clouds, and chemical constituents in the Earth/ atmosphere system. Named for its afternoon equator crossing at 1:30 pm local time, the A-Train travels in tight formation with only about fifteen minutes between the leading and trailing satellites. The near-simultaneous measurement of atmospheric parameters globally and over time allows data from these different satellites to be used together to provide new understanding of atmospheric processes. The Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) on CALIPSO, the

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

175

Conference 6750: Lidar Technologies, Techniques, and Measurements for Atmospheric Remote Sensing
Tropospheric Emission Spectrometer (TES) on Aura, the MODerate Resolution Imaging Spectroradiometer (MODIS) on Aqua, and the POLarization and Directionality of the Earths Reflectances (POLDER) instrument on Polarization and Anisotropy of Reflectances for Atmospheric Science coupled with Observations from a Lidar (PARASOL) provide information about aerosol particle properties, their distribution, both horizontal and vertical, and the chemical composition of the atmosphere. These data can be combined to develop a more complete characterization of aerosol phenomena such as dust storms, biomass burning, and pollution events. Data fusion and visualization techniques will demonstrate complementary views of specific aerosol events. The Atmospheric Science Data Center provides data access, services and tools for over 35 projects, including CALIPSO and TES, in the discipline areas of Earths radiation budget, clouds, aerosols and tropospheric chemistry. Additional information is available from our web site, http://eosweb.larc.nasa.gov.

6750-59, Session 7
Lidar measurements within the world weather research project COPS
A. Behrendt, Infineon Technologies Austria AG (Austria); V. Wulfmeyer, Univ. Hohenheim (Germany); C. Kiemle, G. Ehret, O. Reitebuch, Deutsches Zentrum fr Luft- und Raumfahrt e.V. (Germany); C. Flamant, Univ. Pierre et Marie Curie (France); M. Radlach, S. Pal, M. Schiller, A. Riede, Univ. Hohenheim (Germany); A. Wieser, Consultant (USA); P. Di Girolamo, Univ. degli Studi della Basilicata (Italy); F. Davies, Univ. of Salford (United Kingdom); C. Collier, Consultant (USA); D. Althausen, Leibniz-Institut fr Troposphrenforschung e.V. (Germany); R. Engelmann, J. Cuesta, Consultant (USA) The Convective and Orographically-Induced Precipitation Study (COPS) is a Research and Development Project of the World Weather Research Programme (WWRP) with the goal to investigate the relevant processes leading to convective precipitation in low-mountain regions and to improve their model representation. COPS is carried out in summer 2007 in an area of approximately 250 x 150 km^2 covering the Black Forest Mountains and the Voges Mountains in South-Western Germany and Eastern France. The measurement strategies are based heavily on state-of-the-art lidar systems as key instruments to provide accurate high-resolution data of atmospheric humidity, temperature, wind, and aerosols in clear-air, i.e., in the cloudfree pre-convective atmosphere. These data shall then be complemented with other remote and in-situ data to study the full chain of processes leading to convective precipitation. Airborne and ground-based, vertical pointing and scanning lidars are employed during COPS. Two aircrafts carry lidars: both the Falcon aircrafts of SAFIRE and DLR are equipped with water vapor differential absorption lidar (DIAL) while in the DLR Falcon also a scanning Doppler lidar is implemented. The ground-based lidars are situated along a west-east transect of four supersites where cloud radars and radiometers allow for synergetic remote sensing measurements. At the Supersite V near the Voges mountains, aerosol and water vapor Raman lidar are employed. At both Supersite R in the Rhine valley and Supersite M in the Murg Valley at the Atmospheric Radiation Measurement (ARM) Mobile Facility vertical looking Raman lidar and vertical looking Doppler lidar are combined while at a mountain site in the Northern Black Forest on top of the Hornisgrinde peak (Supersite H), three scanning lidars are collocated: A water vapor DIAL, a rotational Raman lidar and a Doppler lidar. In this presentation an overview of the measurements performed by these 12 lidars will be given and first results of COPS will be discussed.

176

spie.org/ers spie.org/esd spieeurope@spieeurope.org Tel: +44 (0)29 2089 4747

You might also like