Professional Documents
Culture Documents
Geomatica Version 10.1 2007 PCI Geomatics Enterprises Inc.. All rights reserved. COPYRIGHT NOTICE Software copyrighted by PCI Geomatics, 50 West Wilmot St., Suite 200, Richmond Hill, ON CANADA L4B 1M5 Telephone number: (905) 764-0614 RESTRICTED RIGHTS Canadian Government Use, duplication, or disclosure by the Government is subject to restrictions as set forth in DSS 9400-18 General Conditions Short Form Licensed Software. U.S. Government Use, duplication, or disclosure by the Government is subject to restrictions set forth in subparagraph (b)(3) of the Rights in Technical Data and Computer Software clause of DFARS 252.227-7013 or subparagraph (c)(1) and (2) of the Commercial Computer Software-Restricted Rights clause at 48 CFR 52.227-19 as amended, or any successor regulations thereto. PCI, PCI Geomatics, PCI and design (logo), Geomatica, Committed to Image-Centric Excellence, GeoGateway, FLY!, OrthoEngine, RADARSOFT, EASI/PACE, ImageWorks, GCPWorks, PCI Author, PCI Visual Modeler, and SPANS are registered trademarks of PCI Geomatics Enterprises, Inc. All other trademarks and registered trademarks are the property of their respective owners.
Table of Contents
Introduction: Geomatica OrthoEngine .............................................. 1 Module 1: Project Setup ..................................................................... 7 Lesson 1.1 Setting up Satellite Projects .............................................. 10 Lesson 1.2 Setting up Aerial Photograph Projects .............................. 17 Module 2: Computing the Math Model ............................................ 29 Lesson 2.1 Collecting Ground Control Points ..................................... 31 Lesson 2.2 Collecting Tie Points ......................................................... 46 Lesson 2.3 Calculating the Math Model .............................................. 55 Module 3: DEM Operations .............................................................. 65 Lesson 3.1 Creating Epipolar Images ................................................. 66 Lesson 3.2 Extracting and Geocoding the DEM ................................. 71 Lesson 3.3 Building a DEM ................................................................. 79 Module 4: 3-D Feature Extraction .................................................... 85 Lesson 4.1 Viewing Stereo Images in 3-D .......................................... 87 Lesson 4.2 Extracting Features in 3-D ................................................ 94 Module 5: Orthorectification .......................................................... 103 Lesson 5.1 Generating the Orthorectified Images ............................ 105 Module 6: Mosaicking ..................................................................... 111 Lesson 6.1 Defining a Mosaic Area .................................................. 114 Lesson 6.2 Manual Mosaicking ......................................................... 117 Lesson 6.3 Automatic Mosaicking ..................................................... 125
PCI Geomatics
Table of Contents
Module 7: OrthoEngine Componentization .................................. 133 Lesson 7.1 Data Input and GCP Collection ...................................... 134 Lesson 7.2 Project Creation and Tie Point Collection ....................... 145 Lesson 7.3 Automatic GCP Collection and Mosaicking .................... 153 Appendix A: Minimum GCP Requirements .................................. 161
ii
PCI Geomatics
Geomatica OrthoEngine
Introduction
Welcome to the Geomatica OrthoEngine training course. This course is designed for new and experienced users of remote sensing and digital photogrammetry software. In this course you will master the basics of GCP addition, tie point collection, DEM extraction, 3D feature extraction, orthorectification, and mosaicking. In addition, you will learn about some of the newer features such as: OrthoEngine componentization, running components in batch mode, automatic image-toimage registration, and more. There are seven modules in this training manual. Each module contains lessons that you work through in class with the help of your instructor. The lessons are built on basic tasks that you are likely to perform in your daily work. They provide instruction for using the software to carry out essential processes while sampling key OrthoEngine applications and features.
The Geomatica OrthoEngine training manual is for use in a classroom setting with a qualified PCI instructor. Your instructor is available throughout the course to answer any questions you may have. Each module in this book contains a series of hands-on lessons that let you work with the software and a set of sample data. Lessons have brief introductions followed by tasks and procedures in numbered steps. The scope of this guide is confined to the core tools available in Geomatica OrthoEngine. However, some remote sensing concepts are reviewed in the modules and lessons. The following modules are included in this course: Module 1 - Project Setup Module 2 - Computing the Sensor Model Module 3 - DEM Operations Module 4 - 3-D Feature Extraction Module 5 - Orthorectification Module 6 - Mosaicking Module 7 - OrthoEngine Componentization 1
PCI Geomatics
Geomatica OrthoEngine Modules 1 and 2 deal with the Data Preparation stage and include exercises for setting up a project, loading images, adding ground control points (GCPs), collecting tie points (TPs), applying sensor models, and examining reports. Modules 3 and 4 are concerned with Data Extraction, either 2-D Digital Elevation Model (DEM) extraction, or 3-D feature extraction. The extraction of elevation data is necessary so you can either view and edit features in 3-D, or correct your images for distortions by performing an orthorectification using a digital elevation model. Modules 5 and 6 deal with the Data Correction stage and include lessons for setting up and generating orthos, defining a mosaic area, manual mosaicking, and automatic mosaicking. Module 7 examines how to set up and run OrthoEngine components in batch mode using Modeler. The data you will use in this course can be found in the OE Data folder, under the Geomatica directory on your system hard disk.
Note
This training manual can be used to set up any kind of image or airphoto project. Modules 2 through 6 apply to any kind of data. Substitute the file names in the manual with your own data file names. The OrthoEngine Workbook will allow you to explore other math models using a variety of datasets.
PCI Geomatics
The next three sections of the project file list information about: Cutlines created in the mosaicking step Look-up tables generated to match the images radiometrically during the mosaicking step Preferences set up for the appearance of the cursors and labels for such things as GCPs and TPs associated with the working photo or image
The final section of the project file lists the parameters that were set up during the ortho generation step. The information includes the following: Elevation units Resampling method used for the orthorectification Amount of memory allocated for the orthorectification Output resolution Output mosaic file name Upper left bounds of the final mosaic Lower right bounds of the final mosaic
PCI Geomatics
Geomatica OrthoEngine
PCI Geomatics
Figure 1
GeoGateway in Geomatica
With GeoGateway technology you can work through a mapping project by assembling raster and vector data from different sources and different file formats without the need to preprocess or reformat the data. Together, GeoGateway and Geomatica read, view, and process distribution formats, and read, edit, and write exchange formats.
PCI Geomatics
Geomatica OrthoEngine
Figure 2
Conventional files and PCIDSK files
PCIDSK Files
Saved as a single file using the file name extension .pix
Conventional Files
Saved Separately using different file name extensions
Image channels
Image Files
Histogram segments
Histogram files
Starting OrthoEngine
Windows Systems To start OrthoEngine on Windows systems: 1. Click the Start button, click Programs, click PCI Geomatics, click Geomatica Vx.x and then click OrthoEngine. Alternatively, if the Geomatica Toolbar is running, click the OrthoEngine button. The OrthoEngine window opens. Unix Systems To start OrthoEngine on Unix systems: 1. Enter the Unix environment. 2. At the command prompt, type orthoeng. Alternatively, click the OrthoEngine icon on the Geomatica Toolbar.
Figure 3
OrthoEngine window
PCI Geomatics
Project Setup
Module
1
Module 1 has Two Lessons:
Lesson 1.1 Lesson 1.2
The Data Preparation stage consists of the following modules: Module 1: Project Setup Module 2: Computing the Math Model
In Module 1, you learn how to set up your project by: Selecting the math model Specifying the projection information Adding your images to the project Saving your project
Starting a Project
To start a new project you need to select a math model. A math model is a mathematical relationship used to correlate the pixels of an image to correct locations on the ground accounting for known distortions. The math model that you choose directly impacts the outcome of your project. To achieve the results that you are looking for, you need to understand what the math models do, what the math models require to produce an acceptable solution, and which math model to use with your project. You can use one of six math models: Aerial Photography Satellite Orbital Rational Functions Polynomial Thin Plate Spline Mosaic Only
The OrthoEngine Workbook includes exercises and detailed descriptions of the options available for the different math models.
PCI Geomatics
In this manual, an Aerial Photography Modelling project is used to examine the Data Preparation, Data Extraction and Data Correction stages of an OrthoEngine project.
Table 1: Math Modelling Methods available in OrthoEngine Project Stage Aerial Photography
Digital and analog photos supported Camera calibration information is required Exterior orientation can be calculated from GCPs/TPs, GCPs/TPs and GPS/INS, or with GPS/INS only Sensor model is calculated for the block of air photos in the project
Data Preparation
Import and build DEM options are available A DEM can be extracted from stereo photos 3D viewing and 3D feature extraction possible with stereo photos A raster DEM is required for orthorectification Manual and automatic mosaicking
Data Extraction
Data Correction
PCI Geomatics
Polynomial
Mosaic Only
Any digital image in a GDB supported format can be input into a project GCP coordinate contains only x and y values Tie points cannot be collected Sensor model is calculated for each individual image in the project
Any digital image in a GDB supported format can be input into a project GCP coordinate must have x, y and z values Tie points cannot be collected Sensor model is calculated for each individual image in the project
Any digital image in a GDB supported format can be input into a project
Import and build DEM options are available A DEM can be extracted from stereo images 3D viewing and 3D feature extraction possible with stereo images Extracted elevations referenced to ellipsoidal heights
PCI Geomatics
Lesson 1.1
This lesson describes how to set up a pair of satellite images as part of the Data Preparation stage. To set up your images, you require: Optical images in their raw data format, or Radar images in their raw data format Map projection information
Note
The processing steps for optical and radar projects are the same.
10
PCI Geomatics
Data Preparation
Data Extraction
3D Viewing
3D Feature Extraction
Data Correction
Orthorectification Mosaicking
Manual
Automatic
PCI Geomatics
11
Creating a Project
OrthoEngine works on a project-by-project basis. Therefore, you need to open an existing project or create a new project before you gain access to the functions within OrthoEngine.
In this lesson, you will set up a new project using optical data. The procedures are the same for working with radar data. To create a new project: 1. On the OrthoEngine window in the File menu, click New. The Project Information dialog box opens. 2. Click Browse. The File Selector dialog box opens. 3. Locate the SPOT folder. 4. In the File name box, enter spot.prj and click Open. The path and filename appear in the File name box in the Project Information dialog box. 5. In the Name box, enter SPOT Project. 6. In the Description box, enter SPOT ortho project for Irvine, CA. 7. For the Math Modelling Method, select Satellite Orbital Modelling. 8. Under Options, select Toutins Model (Low Res). Note that SPOT is listed in this category. 9. Click OK. The Project Information dialog box closes and the Set Projection dialog box opens.
Figure 1.1
Project Information dialog box
12
PCI Geomatics
The projection information needs to be set at the beginning of each project. In the Set Projection dialog box, enter the projection information for the Irvine area. Output Projection To enter the Output Projection parameters: 1. From the list to the left of the Earth Model button, select UTM. The Earth Models dialog box opens. 2. Click the Ellipsoids tab. 3. Select E000 and click Accept. The UTM Zones dialog box opens. 4. Select Zone 11 and click Accept. The UTM Rows dialog box opens. 5. Select Row S and click Accept. 6. In the Output pixel spacing box, type 10. 7. In the Output line spacing box, type 10.
PCI Geomatics
13
GCP Projection
To enter the GCP Projection parameters: 1. Under GCP Projection, click Set GCP Projection based on Output Projection. The GCP Projection adopts the same settings used for the Output Projection. 2. Click OK. The Set Projection dialog box closes.
Tip
If you wish to modify the projection information, reopen the Set Projection dialog box.
Warning
Changes to the projection mid-project will make any existing orthophotos invalid.
Caution
If you save satellite data from the CD onto a hard disk before reading it to a PCIDSK file, it is important that you maintain the naming structure of the folders as they appeared on the CD. If the structure or folder names are changed, you may encounter errors.
For this lesson, the data have already been read to PCIDSK format. In this case, you will use the Read PCIDSK file option from the Data Input toolbar. To import satellite data from a PCIDSK file: 1. On the OrthoEngine window in the Processing step list, select Data Input. A new toolbar with four icons appears on the OrthoEngine window. With the Data Input toolbar, you can input data from either CD-ROM, PCIDSK file, or a generic image file.
14
PCI Geomatics
2. On the Data Input toolbar, click Read PCIDSK file. The Open Image dialog box opens. 3. Make sure the Uncorrected images option is selected. 4. Click New Image. The File Selector dialog box opens. 5. Locate the SPOT folder. 6. Hold down the CTRL key, select spotleft.pix and spotright.pix and click Open. The Multiple File selector message window opens. This window indicates the total number of files that are detected, and the total number to be loaded into the project. 7. Click OK. The File Selector dialog box closes. Both SPOT images are now part of your project.
Tip
If you need to revert to the backup file, rename the backup file so that it uses the .prj extension. OrthoEngine can load this project file in the normal way.
To change the settings of the backup option: 1. On the OrthoEngine window, click the Options menu and select Auto Backup. 2. Type the number of minutes that you want between backups. 3. Click Close.
PCI Geomatics 15
16
PCI Geomatics
Lesson 1.2
This lesson describes how to prepare four airphotos as part of the Data Preparation stage. To set up your airphotos, you require: Aerial photographs with camera calibration data Map projection information
Term
Airphoto is short for aerial photograph. Aerial photograph, in the broadest sense, means a photograph taken from an airborne platform.
Term
In this lesson, you deal with strip photographs. Strip photography refers to a number of consecutive overlapping photos taken along a flight line, usually at a constant altitude.
PCI Geomatics
17
Input Data
Data Extraction
3D Viewing
3D Feature Extraction
Data Correction
Orthorectification
Mosaicking
Manual
Automatic
18
PCI Geomatics
Creating a Project
OrthoEngine works on a project-by-project basis. Therefore, you need to open an existing project or create a new project before you gain access to the functions within OrthoEngine.
In this lesson, you will set up a new project using four aerial photographs data. To create a new project: 1. On the OrthoEngine window in the File menu, click New. The Project Information dialog box opens. 2. Click Browse. The File Selector dialog box opens. 3. Locate the AIRPHOTO folder. 4. In the File name box, enter airphoto.prj and click Open. The path and filename appear in the File name box in the Project Information dialog box. 5. In the Name box, enter Airphoto Project. 6. In the Description box, enter Airphoto ortho project for Richmond Hill, ON. 7. For the Math Modelling Method, select Aerial Photography. Camera Type Select the type of camera in the Options area of the Project Information dialog box. The options are: A Standard Aerial camera A Digital/Video camera
The photos used in this lesson were taken with a Standard Aerial camera.
Tip
If your photos were taken with a Digital/Video camera, refer to the Digital Airphoto exercise in the OrthoEngine Workbook for further details.
PCI Geomatics
19
Exterior Orientation
The exterior orientation is: Computed from ground control points and tie points, or Provided by the user
Many aircraft are equipped with onboard Global Positioning Systems (GPS), and sometimes with Inertial Navigation Systems (INS). These systems collect the exterior orientation of the camera directly on the aircraft. Select User Input to use the GPS and INS readings alone and accept them as correct. Select Compute from GCPs and Tie Points to use ground control points and/or tie points to refine the GPS and INS results.
Term
Whereas the interior orientation defines the relationship between the camera and the image, the exterior orientation defines the relationship between the camera and Earth. Specifically, the exterior orientation defines the spatial position and angular orientation of a photo.
9. In the Exterior Orientation list, select Compute From GCPs & Tie Points. 10. Click OK. The Project Information dialog box closes and the Set Projection dialog box opens.
Figure 1.4
Project Information dialog box after the information is entered
Tip
If you wish to modify your project information at any time, reopen the Project Information dialog box. However, you can not change the math model once it has been set.
20
PCI Geomatics
The projection information needs to be set at the beginning of each project. In the Set Projection dialog box, enter the projection information for the Richmond Hill, Ontario area. Output Projection To enter the Output Projection parameters: 1. From the list to the left of the Earth Model button, select UTM. The Earth Models dialog box opens. 2. Click the Ellipsoids tab. 3. Select E012 and click Accept. The UTM Zones dialog box opens. 4. Select Zone 17 and click Accept. The UTM Rows dialog box opens. 5. Select Row T and click Accept. 6. In the Output pixel spacing box, type 0.4. 7. In the Output line spacing box, type 0.4.
PCI Geomatics
21
GCP Projection
To enter the GCP Projection parameters: 1. Under GCP Projection, click Set GCP Projection based on Output Projection. The GCP Projection adopts the same settings used for the Output Projection. 2. Click OK. The Set Projection dialog box closes and the Standard Aerial Camera Calibration Information dialog box opens.
Tip
If you wish to modify the projection information, reopen the Set Projection dialog box.
Warning
Changes to the projection mid-project will make any existing orthophotos invalid.
Figure 1.6
Standard Aerial Camera Calibration Information dialog box
22
PCI Geomatics
Focal Length
The Focal Length is the distance between the focal point of the lens and the film. Entering an incorrect focal length may introduce unwanted distortions in your project. This is a compulsory parameter. Radial Lens Distortion is the symmetric distortion caused by the lens due to imperfections in curvature when the lens was ground. In most cases, the errors introduced by radial lens distortion (around 1 to 2 um) are much smaller than the scanning resolution of the image (around 25um). Entering the values may significantly increase the processing time while contributing very little value to the final product. The values for the Radial Lens Distortion may be provided to you as R0 through R7 coefficients or in tabular format. These parameters are optional and the coefficients may or may not appear in the camera calibration report. If you are using a USGS camera calibration report, the coefficients are given as K0, K1, K2, K3 and K4, which correspond to R1, R3, R5, and R7. K4 is discarded since it is usually zero.
Fiducial Marks
Fiducial marks are small crosses or small V-shaped indents located precisely on each of the four corners and/or exactly midway along the four sides of a standard aerial photograph. After you identify the fiducial marks in your scanned image, OrthoEngine uses the fiducial marks entered from the camera calibration report to establish an image coordinate frame. The fiducial mark coordinates are a compulsory parameter for standard aerial photographs. Image Scale is the ratio of the size of the objects in the image to the size of the objects on the ground. This parameter is optional, except when you want to import GPS/INS observations and use them during the automatic tie point measurements. Entering the incorrect Image Scale may cause the computation of the math model (the bundle adjustment) to fail.
Photo Scale
Earth Radius
The Earth Radius is the radius of curvature of the earth at the location of the project. This parameter is optional since aerial photographs usually use a large scale (for example, 1:8,000) and the error due to the earth's radius is negligible. You only need earth radius correction for images with a scale over 1:20,000.
PCI Geomatics
23
To enter the Camera Calibration data: 1. In the Standard Aerial Camera Calibration Information dialog box, enter the information shown in Table 2.
Table 2: Camera Calibration Data
Focal Length Corner Fiducial Marks Top Left Top Right Bottom Right Bottom Left Image Scale 152.856 X -106.000 106.000 105.996 -105.996 1:8000 Y 106.000 106.000 -106.000 -106.000
Tip
You can modify the camera calibration information at any time by reopening the Standard Aerial Camera Calibration Information dialog box.
2. On the Data Input toolbar, click Open a new or existing image. The Open Image dialog box opens. 3. Click New Image. The File Selector dialog box opens.
24 PCI Geomatics
4. Locate the AIRPHOTO folder. 5. Press the CTRL key and select files S129.pix, S130.pix, S188.pix and S189.pix and click Open. The Multiple File selection message window opens indicating the total number of files that are detected and are to be loaded into the project. 6. Click OK. The four photos are listed in the Open Image dialog box. To open the first photo: 1. In the Open Image dialog box, select S129.pix and click Open. A viewer opens displaying photo S129.pix. In addition, the Fiducial Mark Collection dialog box opens for this photo.
Figure 1.8
Viewer showing photo S129.pix
Figure 1.9
Fiducial Mark Collection dialog box
PCI Geomatics
25
Tip
If you are working in a project with a large volume of images, it is recommended that you enter the fiducial marks and ground control points for a limited number of images (up to five), complete the calculation of the math model, and then check for errors before continuing. It is easier to locate bad points on a few images than over the entire project.
Manual Collection
To manually collect fiducial marks: 1. Click the approximate location of the fiducial mark in the top left corner, using the zoom tools as necessary. A red crosshair appears in the viewer. 2. Click precisely in the center of the fiducial mark. 3. In the Fiducial Mark Collection dialog box, click Set beside the Top left pixel and line boxes. The Pixel and Line coordinates for the fiducial mark appear in the dialog box. 4. Repeat steps 1 to 3 to collect fiducial marks in the Top right, Bottom right and Bottom left corners. 5. For the Calibration Edge, select Left. This is the position of the data strip as it appears in the image on the screen.
Errors
Under Errors, OrthoEngine compares the computed fiducial mark positions based on the measurements taken from the screen with the fiducial information that you entered from the camera calibration report. Click Clear beside any fiducial marks where the error is not acceptable and repeat the collection process. The error should be less than one pixel, unless the image is scanned at a very high resolution. Large errors may indicate that either the coordinates from the camera calibration report were entered incorrectly or the fiducial mark was collected incorrectly from the scanned image.
26
PCI Geomatics
Automatic Collection
After collecting the fiducial marks manually for one of your images, OrthoEngine can use automated pattern matching to automatically collect the fiducial marks for the rest of your images in the project. To automatically collect fiducial marks for the remaining photos: 1. After manually collecting fiducials for the first photo, click Auto Fiducial Collection. A window opens, that asks Do you want to overwrite photos with fiducial marks? 2. Click No. This will use the pattern matching only on images without measured fiducial marks. 3. After the Progress Monitor closes, click OK. 4. To accept the fiducial marks, click OK. The Fiducial Mark Collection dialog box closes. You can verify the accuracy of the fiducial mark collection by viewing the fiducial.rpt report in the folder where the project is saved.
Tip
This is a good time to save your project file.
To save your project file: From the File menu in the OrthoEngine window, click Save. The airphoto.prj file is saved in the AIRPHOTO folder.
PCI Geomatics
27
Checkpoint
You are now ready to proceed to Module 2 Computing the Math Model. In Module 2, you collect ground control points and tie points for your project, and then calculate the math model by way of a bundle adjustment.
28
PCI Geomatics
Module
2
Module 2 has Three Lessons:
Lesson 2.1 Lesson 2.2 Lesson 2.3
Collecting Ground Control Points Collecting Tie Points Calculating the Math Model
The Data Preparation stage consists of the following modules: Module 1: Setting Up Images and Photos Module 2: Computing the Math Model
In Module 2, you learn how to: Collect ground control points Collect tie points Calculate the math model
Term
The computation of a rigorous math model is often referred to as a bundle adjustment. The math model solution calculates the position and orientation of the sensor - the aerial camera or satellite - at the time the image was taken. Once the position and orientation of the sensor is identified, it can be used to accurately account for known distortions in the image.
PCI Geomatics
29
Data Extraction
3D Viewing
3D Feature Extraction
Data Correction
Orthorectification M osaicking
M anual
Automatic
30
PCI Geomatics
Lesson 2.1
To add GCPs, you require: The airphoto.prj project file from Module 1 for the aerial photographs S129.pix, S130.pix, S188.px and S189.pix. The air_mos.pix mosaic file contained in the AIRPHOTO folder, which serves as the georeferenced image.
Term
A ground control point (GCP) is a feature that you can clearly identify in the raw image for which you have a known ground coordinate.
Ground coordinates can come from a variety of sources such as the Global Positioning System (GPS), ground surveys, geocoded images, vectors, Geographic Information Systems (GIS), topographic maps, chip databases, or by using photogrammetric processes to extend the number of GCPs in your images. A GCP determines the relationship between the raw image and the ground by associating the pixel (P) and line (L) image coordinates to the x, y, and z coordinates on the ground. Although the media, formats, and methods used to collect the coordinates are different depending on the source, the idea is the same. You have to match a point in the raw image to a set of coordinates. Since some sources of ground control only offer dispersed points, it may be more efficient to select a point in the source first and then locate it in the raw image. For example, a vector file may have a limited number of features available as ground control compared with a geocoded image.
PCI Geomatics
31
Select features that can be identified accurately at the resolution of the raw image. Select features that are close to the ground. Because elevated features in the image will appear to lean, selecting features on the ground will ensure that the point is not displaced from the actual ground coordinate. Avoid picking shadows. These are easy to see in the image, but they are not permanent features and can move from one image to another. Avoid repetitive features such as parking lots and lines on a highway, since it is easy to select the wrong one. When collecting GCP coordinates in the field (via GPS or survey), try to identify good targets in the raw image before arbitrarily collecting coordinates in the field. GCPs should be collected in a wide distribution over the image and the project. Ensure that the GCPs are collected from a variety of ground elevations. A GCP may be selected on a single image, or may be selected in an area of overlap between 2 or more images. GCPs selected in multiple images help to produce a more accurate model.
The minimum number of GCPs you need to collect depends on the type of data you are correcting, the processing level of that data, and which math model you are using. For more information, please refer to Appendix A. For the Aerial Photography model, the minimum requirement is that you have at least two GCPs on at least one photo in the project, to ensure scale. However, there should be a few photos with 3 GCPs in the project. This provides correct levelling and scale for the math model. Tie points can hold the rest of the project together.
Tip
If you are working in a project with a large volume of images, it is recommended that you enter the ground control points for a limited number of images (up to five), complete the calculation of the math model, and then check for errors before continuing. It is easier to locate bad points on a few images than over the entire project.
32
PCI Geomatics
You will now open S129.pix as the Working Image. To open the Working Image: 1. On the OrthoEngine window in the Processing step list, select GCP/TP Collection. A new toolbar with seven icons appears. The icons on the toolbar are shortcuts to all the functions you need during GCP and TP collection.
Figure 2.1
GCP/TP Collection toolbar
2. On the GCP/TP Collection toolbar, click Open a new or existing image. The Open Image dialog box opens listing the four photos in this project. 3. Select S129.pix and click Open. A viewer opens containing the S129.pix photo as the Working Image.
Figure 2.2
Viewer containing S129 as the Working Image
PCI Geomatics
33
2. From the Ground control source list, select Geocoded image. A File Selector dialog box opens automatically. 3. From the AIRPHOTO folder, select air_mos.pix and click Open. The air_mos.pix file is loaded in a viewer as the Geocoded Image and is listed at the top of the GCP Collection dialog box.
34
PCI Geomatics
Lesson 2.1 Collecting Ground Control Points If you chose the Aerial Photography, Satellite Orbital, Rational Functions, or Thin Plate Spline math models, you can use a digital elevation model (DEM) to determine the elevation of your GCPs. The DEM does not have to be in the same projection as the source of the GCPs.
To load the DEM to set elevation: 1. Beside the DEM box, click Browse. 2. From the AIRPHOTO folder, select ap_dem.pix and click Open. The DEM File dialog box opens where you select the channel containing the DEM information. This will be your source of elevation for your ground control points. 3. Enter a Background elevation of -150 and click OK. Background Elevation Background elevation represents those areas inside the DEM for which there is no data provided. For DEMs generated by OrthoEngine, the background elevation defaults to -150. Other DEMs have different background elevation values that you must know before they can be used.
Tip
If you do not know the background value, click DEM Info in the DEM File dialog box. The window displays the three lowest and three heights values in the DEM.
To collect GCPs from a geocoded image: 1. In the air_mos.pix viewer, place the crosshairs near the right edge of the image at the position shown in the figure below.
Figure 2.4
Location of first GCP (circled in black)
PCI Geomatics
35
2. Place the cursor on the location shown in the figure below, zooming in as necessary.
Figure 2.5
Location of crosshairs for G0001 (circled in black)
3. In the Geocoded Image viewer, click Use Point. The georeferenced coordinates for this location are transferred to the GCP Collection window. They should be approximately:
627260 X 4857514 Y
4. In the GCP Collection dialog box, click Extract Elevation. 5. Place the crosshairs on the same feature in the uncorrected S129.pix photo. 6. When you are satisfied with the position of the crosshairs, click Use Point. The image coordinates for G0001 are transferred to the GCP Collection dialog box. They should be approximately:
320 Pixel 2257 Line
7. In the GCP Collection dialog box, click Accept. The GCP information is transferred to the Accepted Points list for the GCP with Point ID G0001.
36
PCI Geomatics
Term
The Point ID is a label automatically assigned to each GCP. You can type a new label in the Point ID box, however, all points (ground control points, independent check points, tie points, and elevation match points) in the image must have unique labels. When collecting stereo GCPs (the same GCP in the overlap areas of different images), use the same Point ID in each image.
Note
The same workflow is used when collecting GCPs from geocoded vectors.
Tip
You can edit the error estimate in the +/- boxes to correspond to your ability to precisely identify a feature in the image. For example, if you use coarse imagery, you can probably only measure to the closest pixel. If you use imagery that was compressed or poorly scanned, you may only be able to measure to the closest two pixels. Even if you identify a GCP to the closest pixel, the coordinate may only be accurate to a certain number of meters.
To collect the second GCP: Follow steps 1 to 7 above to collect a GCP at the location shown in the figure below.
Figure 2.6
Location of G0002 on air_mos.pix
PCI Geomatics
37
To collect the third GCP: Follow steps 1 to 7 above to collect a GCP at the location shown in the figure below.
Figure 2.7
Location of G0003 on air_mos.pix
To import GCPs from a file: 1. From the Ground control source list in the GCP Collection dialog box, select PIX/Text file. The Read GCP From PIX/Text File dialog box opens. 2. Click Select. 3. From the AIRPHOTO folder, select S129.GCP and click Open. 4. From the Sample formats list, select IXYE. IXYE Format The format of the S129.GCP file is IXYE, which means that each row of text contains the following information from left to right: The GCP ID (I) The georeferenced East/West coordinate (X) The georeferenced North/South coordinate (Y) The elevation (E) The Pixel and Line positions will be taken from the uncorrected S129 photo.
38 PCI Geomatics
5. Click Apply Format. The information for each point is listed in the GCPs extracted from file area.
Figure 2.8
Read GCP From Text File dialog box with GCPs in IXYE format
To transfer the coordinates: 1. In the GCP Text File dialog box, select G0014 and click Transfer to GCP collection panel. The IXYE coordinate is transferred to the GCP Collection dialog box. If Auto locate is enabled in the GCP Collection dialog box, OrthoEngine will estimate the position of the GCP in the uncorrected image. In the viewer for S129, you will notice that the crosshairs have automatically moved near the correct position for G0014. 2. Use the figure below to more accurately position the crosshairs for G0014.
PCI Geomatics
39
3. Adjust the position of the crosshairs and click Use Point. The image coordinates for G0014 are transferred to the GCP Collection dialog box. They should be approximately:
2665 Pixel 2392 Line
4. In the GCP Collection dialog box, click Accept. The GCP information is transferred to the Accepted Points list for the GCP with Point ID G0014. 5. In the GCP Text File dialog box, select G0015 and click Transfer to GCP collection panel. 6. Use the figure below to more accurately position the crosshairs for G0015.
40
PCI Geomatics
7. Adjust the position of the crosshairs and click Use Point. The image coordinates for G0015 are transferred to the GCP Collection dialog box. They should be approximately:
4131 Pixel 4159 Line
8. In the GCP Collection dialog box, click Accept. The GCP information is transferred to the Accepted Points list for the GCP with Point ID G0015.
Tip
This is a good time to save your project file.
To save time, the GCPs for S130.pix, S188.pix and S189.pix will be imported from a text file in IPLXYE format. IPLXYE Format The format of the S130.GCP file is IPLXYE, which means that each row of text contains the following information from left to right: The GCP ID (I) The image pixel position (P) The image line position (L) The georeferenced East/West coordinate (X) The georeferenced North/South coordinate (Y) The elevation (E)
PCI Geomatics 41
To open S130.pix as the Working Image: 1. On the GCP/TP Collection toolbar, click Open a new or existing image. The Open Image dialog box opens. 2. Select S130 and click Quick Open & Close. The Open Image dialog box closes and a viewer opens containing the S130 photo as the Working Image.
Note
One image is always the Working image, while the other image is the Reference image. Click Reference on the viewer toolbar to set an image to Working.
To import the GCPs for S130.pix: 1. On the GCP Collection dialog box, click Select PIX/Text File. The Read GCP from Text File dialog box opens. 2. Click Select. 3. From the AIRPHOTO folder, select S130.GCP and click Open. 4. From the Sample formats list, select IPLXYE. 5. Click Apply Format. The information for each point is listed in the GCPs extracted from file area.
Figure 2.12
Read GCP From Text File dialog box for GCPs in IPLXYE format
6. Check that the GCPs are listed correctly and click OK. After the GCPs are read in, the GCP ID number for each point appears in the viewer in red. Check to see that the GCP positions are correct.
42 PCI Geomatics
To import the GCPs for S188.pix and S189.pix: 1. Open S188.pix and S189.pix. 2. Make S188 the Working image. 3. Import the GCPs for S188 using the S188.GCP text file. 4. Make S189 the Working image. 5. Import the GCPs for S189 using the S189.GCP text file.
Note
When a photo is selected as the Working image, the GCPs collected for this image are listed in the GCP Collection dialog box.
To collect Stereo GCPs: 1. Ensure photos S129 and S130 are open, as well as the GCP Collection dialog box. 2. Make S130.pix the Working image. 3. In the GCP Collection dialog box, locate and select Point ID G0016. This loads the GCP image and georeferenced position information based on the S130.pix photo. If Auto Locate is selected, this GCP is loaded in each of the viewing windows. 4. Make the S129.pix photo the Working image by clicking Reference on the viewer toolbar. A list of GCPs collected for the Working image now appears in the GCP Collection dialog box. The identification of the desired stereo point, G0016, is listed in the GCP Collection dialog box for S129. The georeferenced position information for this point is shown in the window. However, the image pixel and line information for the Working image is not listed, since you have not yet associated the point G0016 with S129.pix. 5. Place your cursor on the feature that corresponds with point G0016 on S129.
PCI Geomatics 43
6. Click Use Point. Note that the Image pixel and line information updates in the GCP Collection dialog box. 7. Click Accept. This point is now registered to the same georeferenced location on the earth for both overlapping images, making it a Stereo GCP. 8. Collect another stereo GCP based on Point ID G0003.
To enable Compute model: At the top of the GCP Collection dialog box, enable the Compute model option. Residual errors are calculated for each GCP and are listed in the Residual column. Residual errors are the difference between the coordinates that you entered for the ground control points (GCPs) or tie points and where those points are according to the computed math model. Computing the math model will be discussed further in Lesson 2.3.
Check Points
Whereas GCPs are used in computing the math model, Check Points are used to check the accuracy of the math model and are not used to compute the math model. OrthoEngine calculates the difference between their position and the position determined by the model; therefore, the Check Points provide an independent accuracy assessment of the math model.
Next, you will take an existing GCP and turn it into a Check Point. To create a Check Point: 1. Make S130.pix the Working image. 2. In the GCP Collection dialog box, select G0014 or a GCP with a high residual value. 3. From the list beside the Point ID box, select Check. 4. Click Accept. This point is now a Check point and is not included in the bundle adjustment.
44
PCI Geomatics
To convert a Check Point to a GCP: 1. In the GCP Collection dialog box, select G0014. 2. From the list beside the Point ID box, select GCP. 3. Click Accept. This point is once again a GCP and is included in the bundle adjustment. 4. Click Close.
Tip
This is a good time to save your project file.
PCI Geomatics
45
Lesson 2.2
To add tie points, you require: The project file airphoto.prj from Lesson 2.1 for photos S129.pix, S130.pix, S188.pix and S189.pix.
The objective of this lesson is to tie the four photos together and check the layout to ensure a proper distribution of points.
Term
A tie point is a feature that you can clearly identify in two or more images that you can select as a reference point.
Tie points do not have known ground coordinates, but you can use them to extend ground control over areas where you do not have ground control points (GCPs). Used in rigorous models such as Aerial Photography and Satellite Orbital (high and low resolution) math models, tie points identify how the images in your project relate to each other. In a project using the Rational Functions math model where you have imported the polynomial coefficients distributed with the data, you can collect tie points and ground control points to compute a transformation to improve the fit between the images. For projects using the Aerial Photography math model, you usually collect tie points in a three-by-three pattern over the image. Since the images have a 60 percent overlap between each other and a 20 percent overlap between the strips, you can use the three-by-three pattern to connect six overlapping images. Projects using the Satellite Orbital math model generally have fewer images so you can collect tie points wherever overlap occurs. Since the overlap between satellite images is unpredictable, satellite imagery generally covers a large area containing a lot of ground control. Using the tie points in the calculation of the math model ensures the best fit not only for the individual images, but for all the images united as a whole. Therefore, the images will fit the ground coordinate system, and overlapping images will fit each other.
46
PCI Geomatics
Select features that can be identified accurately at the resolution of the raw image. Select features that are close to the ground. Because elevated features in the image will appear to lean, selecting features on the ground will ensure that the point is not displaced from the actual ground coordinate. Avoid picking shadows. These are easy to see in the image, but they are not permanent features and can move from one image to another. Avoid repetitive features such as parking lots and lines on a highway, since it is easy to select the wrong one. While tie points that join two images together are effective, tie points that join 3 or more together are even better. Tie points that join multiple images together produce a more accurate model. If the elevation value at the tie point location is known, then enter that value in the elevation field in the tie point collection panel. These points help to quantify elevation, improving the accuracy of the geometric model.
PCI Geomatics
47
Notice that the block of photos have an area in common. This area represents the overlap. The figure below depicts the four photos with the area of overlap bounded by a black rectangle.
Figure 2.13
Overlap
48
PCI Geomatics
To collect tie points manually: 1. On the GCP/TP Collection toolbar, click Manually collect tie points. The Tie Point Collection dialog box opens.
Figure 2.14
Tie Point Collection dialog box
2. In the Auxiliary Information section, click Select. 3. From the AIRPHOTO folder, select ap_dem.pix and click Open. The DEM File dialog box opens where you select the channel containing the DEM information. This will be your source of elevation for your tie points. Entering tie point elevation is optional. You can either load a DEM or select the Elevation option and manually enter the elevation. The elevation of the tie point is automatically incorporated into the math model. 4. Enter a Background elevation of -150 and click OK. 5. Make S129.pix the Working image. 6. Find a location in S129.pix that can be clearly seen in the overlap area in S130.pix, zooming in as necessary. 7. Click to place the cursor at this location. A red crosshair appears.
PCI Geomatics
49
8. In the S129 viewer, click Use Point. If Auto locate is enabled in the Tie Point Collection dialog box, OrthoEngine estimates the position of the Tie Point in the overlap area of the other photos. You need to refine the position of the crosshairs on S130.pix before accepting the tie point on this image. 9. Place the cursor at the exact same location in S130.pix, zooming in as necessary. 10. In the S130 viewer, click Use Point. 11. In the Tie Point Collection dialog box, click Accept. The point is listed in the Accepted Tie Points table. 12. Repeat these steps to collect at least four additional tie points for the project. For Auto locate to work, you need to click Use Point on the Working image after placing the crosshairs on a feature that can be seen in both images. It does not matter which image you set as the Working image. Tie points can also be collected between flight lines. 13. Click Close.
Tip
This is a good time to save your project file.
50
PCI Geomatics
Lesson 2.2 Collecting Tie Points The exterior orientation of each image was computed based on ground control points (GCPs) and/or tie points. You collected three tie points between every pair of overlapping images. You used a Global Positioning System (GPS) to obtain the x, y, and z coordinates for each image center, and you estimated the omega, phi, and kappa rotations, or they were supplied by an Inertial Navigation System (INS). You used the Satellite Orbital math model and imported the ephemeris or orbit information with the satellite data. You imported the polynomial coefficients from the satellite data for use with the Rational Functions math model.
Now that you have collected some tie points manually, you will collect tie points automatically. To open the Automatic Tie Point Collection dialog box: On the GCP/TP Collection toolbar, click Automatically collect tie points. The Automatic Tie Point Collection dialog box opens.
Figure 2.15
Automatic Tie Point Collection dialog box
Entire image: To distribute the tie points evenly over the entire image and match
each tie point in all the overlapping images. This is normally used to generate standard tie point distributions for aerial photographs such as the three-by-three pattern.
Overlap area: To distribute the points evenly only in the overlap area between any pair of overlapping images. This is normally used for satellite images or for aerial photographs with less than 60% overlap.
PCI Geomatics
51
The Tie Point Options section contains the following four options:
Tie points per area: You specify the number of Tie Points to generate per
overlap area for each pair of images.
Search radius: This is the number of pixels defining the radius of the search frame. If you are experiencing a low success rate with the Automatic Tie Point Collection, increasing the search radius may improve the results. Approximate Elevation: Entering an approximate elevation allows the matching algorithm to make a better first estimate of the parallax between the images. This improves the success rate of the collection process.
Images to Process Tie Points are automatically collected in one of two ways:
All Images: To collect tie points for all images in the project. Working Image: To collect tie points for the image designated as the working
image. This option is available if there is currently a Working Image in the viewer. Processing Start Time Two options are available for the start time:
Start now: Begins the process after you click the Collect Tie Points button. Start at (hh:mm): To run the process overnight, select the Start at (hh:mm)
option, which lets you start the process at any time within the next 24 hours. To set up and run Automatic Tie Point collection: 1. For the Distribution Pattern, use the default setting of Entire image. 2. For the Tie points per area, enter the value 3. 3. For the Matching threshold, use the default value of 0.75. 4. For the Search radius, use the default value of 100 Pixels. 5. Enter an Approx. elevation value of 200 m. This value is the average elevation of the terrain in your project. 6. For Image to Process, use the default option of All images. 7. Click Collect Tie Points. A progress bar appears at the bottom of the dialog box to monitor the status. After the process is complete, a message box opens indicating the total number of tie points found. 8. Click Close.
52 PCI Geomatics
To open Image Layout: 1. On the GCP/TP Collection toolbar, click Display overall image layout. The Image Layout window opens. The Overview area shows the center of each image in the project. The top of the window points northward. 2. Under Overview, click a crosshair to reveal the images footprint. In the right side of the window, the selected image is displayed with a red frame, while the other images are framed in blue. The GCPs are displayed as red squares, while the TPs are displayed as blue squares. 3. To open an image, double-click the image footprint or click Quick Open. If you are not satisfied with the distribution, edit your GCPs and tie points. 4. Click Close.
PCI Geomatics
53
Tip
This is a good time to save your project file.
54
PCI Geomatics
Lesson 2.3
Geomatica OrthoEngine - Module 2: Computing the Math Model the solution the least. Using many GCPs and tie points provides redundancy in the observations so that a few bad points will not greatly affect your model, and the bad points will be easier to identify. Once the sensor orientation is calculated, it is used to drive all the other processes such as digital elevation model extraction, editing in three-dimensional stereo, and orthorectification. You must obtain an accurate math model solution before continuing with other processes.
The Compute model feature appears on the GCP Collection and the Tie Point Collection dialog box when you are creating a project using a rigorous model. When you select Compute model, OrthoEngine calculates the math model every time you add a point to the project. This can help you determine whether the point that you collected is good enough for your project.
To enable Compute model: At the top of the GCP Collection or Tie Point Collection dialog box, enable the Compute model option.
Note
If Compute model is disabled, the GCPs or TPs are listed with a status of Stale in the Residual column of the corresponding Accepted Points table.
The second way to calculate the math model is found on the Model Calculations toolbar on the OrthoEngine window. To compute the rigorous math model: 1. On the OrthoEngine window in the Processing steps list, select Model Calculations. A new toolbar with one icon appears. 2. Click Compute model.
56
PCI Geomatics
Note
In Rational Functions computed from GCPs, Polynomial, and Thin Plate Spline projects, images are not connected together with tie points. Therefore, the math model and the resulting residual errors are calculated for each image separately. If you selected the Thin Plate Spline math model for your project, the residual errors will always indicate zero. Use Check Points to check its accuracy.
Another way to verify the quality of the model is to collect some GCPs as Check Points. Check Points are not used to compute the math model, but OrthoEngine calculates the difference between their position and the position determined by the model and includes the error in the Residual Errors report. Therefore, the Check Points provide an independent accuracy assessment of the math model. In most projects you should aim for the residual errors to be one pixel or less. However, you should also consider how the resolution of the image, the accuracy of your ground control source, and the compatibility between your ground control source and the images can affect the residual errors. You may want to use a topographic map as a ground control source, however, features on topographic maps may be shifted several meters for aesthetic reasons. This limits the accuracy of the coordinates that you can obtain from the map. Also, the detail visible on a 1:50,000 scale topographic map may not be compatible with the high resolution of an aerial photograph. For example, if you choose a road intersection in a topographic map as your coordinate, the same road intersection in the aerial photograph may consist of several pixels. Therefore, the residual error will likely be larger than a pixel.
PCI Geomatics
57
Geomatica OrthoEngine - Module 2: Computing the Math Model An existing LANDSAT orthorectified image may make a convenient ground control source for registering a new IKONOS image, but the resolution of the LANDSAT image is 30 meters and the resolution of the IKONOS raw image is 1 meter. Therefore, even if you could pick the right pixel in the IKONOS image, your GCP from the LANDSAT image is only accurate to 30 meters. You cannot achieve accuracy of 2 to 4 meters unless your ground control source is equally accurate. At first glance, a residual error of 250 meters in ground distance may appear too high. However, if your raw data has a resolution of 1000 meters, such as AVHRR, you have already achieved sub-pixel accuracy.
Outliers
A ground control point (GCP) or tie point with a very high residual error compared to the others in the Residual Errors report may indicate an error in the original GCP coordinate, a typographical mistake, or an error in the position of the GCP or tie point on the raw image. These points are called outliers. To correct an outlier: Verify that the feature you picked in the raw image corresponds to the one from your source. Verify that the typed ground coordinate matches the coordinate listed in your source. Confirm that the ground coordinate you collected in the raw image is consistent with the coordinate you selected from the vector or the geocoded image. Verify that the projection and datum for the ground coordinate are correct.
If the residual errors for all the GCPs and/or tie points in general are high, it may indicate a poor model solution. Poor model solutions can be the result of inaccurate GCPs, errors in the projection or datum, inadequate distribution of the ground control, or insufficient ground control. If all the residual errors for the GCPs and tie points read zero, it usually indicates that you have collected only the minimum number of ground control points or fewer. Collect more GCPs and tie points.
58
PCI Geomatics
Lesson 2.3 Calculating the Math Model However, if you selected the Thin Plate Spline math model for your project, the residual errors will always indicate zero. Use Check Points to check the accuracy for the Thin Plate Spline math model.
If you have high residual errors in one part of an image or project, it can indicate that you need more ground control in the problem area, or it may indicate that you have one or more bad points in the area that are skewing the math model. Some bad points are difficult to identify since some points may compensate for others.
To open the Display Residuals dialog box: From the Options menu on the OrthoEngine window, select Residual Display. The Display Residuals dialog box opens.
Figure 2.17
Display Residuals dialog box
PCI Geomatics
59
Residuals List
X,Y: to display the x-axis and y-axis residual errors for each point as separate
vectors.
XY: to display the residual error for each point as one vector representing the combined x-axis and y-axis residual errors. Z: (for stereo GCPs and stereo check points only) to display the residual error for the elevation of each point as one vector along the y-axis.
Coordinate System There are two options for the axis on which the residual error will be displayed.
Ground: to display the residual error using Easting and Northing directions
(ground coordinate system).
Image: to display the residual error using the x and y axis directions (image
coordinate system). To open a photo: 1. From the Processing step list, select GCP/TP Collection. 2. Click Open a new or existing image. 3. Select S129.pix and click Quick Open and Close. 4. On the GCP/TP Collection toolbar, click Collect GCPs Manually. The GCP Collection dialog box opens. You will now plot the residuals on the photo you opened. To plot the residuals on a photo: 1. In the Display Residuals dialog box, use the default of X,Y for the Residuals option. 2. For the Coordinate system, select Image. 3. Under the Display column, click beside GCPs, Check Points and Tie Points. A red check mark appears in each cell. 4. Click Apply. To view the vector residuals: 1. In the GCP Collection dialog box, select G0001 from the Point ID list. The viewer is updated to show this point at a resolution of 1:1. The vector residuals are plotted as red lines.
60
PCI Geomatics
2. On the Display Residuals dialog box, change the Coordinate system to Ground and click Apply. The Residuals are displayed using the ground coordinate system instead of the image coordinate system. 3. Examine how the residuals are plotted when you select the NE and H options.
Note
By default, the magnification factor for the residual plots is set to 5. The magnification factor exaggerates the appearance of the residual error so it becomes easier to observe systematic patterns. If the residual error is one pixel and the Magnify value is five, the residual error is displayed as five pixels long in the viewer.
To highlight points with high residual values: 1. Change the Coordinate system back to Image. 2. For the Residuals option, select X,Y. 3. In the Highlight residual greater than box, enter 1 pixels. 4. Click Apply. 5. In the GCP Collection dialog box, select a GCP that has a Res X or Res Y residual greater than 1. Residual errors over the 1 pixel threshold are highlighted by an increase in the thickness of the vector lines. 6. Click OK.
PCI Geomatics 61
To open the Residual Errors dialog box: 1. On the GCP/TP Collection toolbar, select Residual report. The Residual Errors dialog box opens. The Residual report can also be opened by selecting Reports from the Processing Step list.
Figure 2.19
Residual Errors dialog box
Defines whether residual errors are reported in Ground units or Image pixels. Lets you select which types of points to display.
All images: To display all the images in the project. Selected image: To display one image in the project, click an image in the table under Image ID and click Selected image, or you can type the image's identification in the Selected image ID box and press ENTER.
62
PCI Geomatics
Sort By
Residual: To order the residual errors from the highest to the lowest value. Data snooping: To order the normalized residual errors from highest to lowest
probability of error that is not noise. Because the residuals are highly correlated to one another, a blunder in one point may cause all points to have higher residuals. Because of this, it may be difficult to isolate the source of the problem. Data snooping tries to uncorrelate the errors in order to isolate the bad point by calculating this statistical value.
Residual Errors
For each point in the project, the Residual Errors dialog box lists the information shown in the following table. Table 1: Residual Errors Heading Point ID Res Res X Res Y Type Image ID Image X Image Y Comp X Comp Y Description The points identification number The combined residual error The points X-axis residual error The points Y-axis residual error Type of point, GCP or TP Image to which the point belongs X-coordinate - # of pixels from left Y-coordinate - # of pixels from top Computer adjusted X-coordinate Computer adjusted Y-coordinate
Caution
If you make any edits to your model in the Residual Errors window, make sure you recalculate the model by clicking Compute Model.
Tip
This would be a good time to save your project file.
Checkpoint
Module 3
After the preparation stage, you may need to extract a Digital Elevation Model or build a DEM from existing data. If so, proceed to the data extraction stage which begins with Module 3: DEM Operations. Here, you learn how to extract a DEM from four airphoto scenes. However, the procedures can also be used to extract a DEM from any stereo images. On the other hand, after the preparation stage you may need to extract features in 3-D. If so, go directly to Module 4: 3-D Feature Extraction where you learn about 3-D operations, such as viewing and extracting features in 3-D from a pair of stereo images. If you have an existing raster DEM to use for orthorectification, go directly to the correction stage, which begins with Module 5: Orthorectification. In Module 5, you learn how to set up, select your DEM, and then orthorectify your satellite images or aerial photographs.
Module 4
Module 5
64
PCI Geomatics
DEM Operations
Module
3
Module 3 has Three Lessons:
Lesson 3.1 Lesson 3.2 Lesson 3.3
Creating Epipolar Images Extracting and Geocoding the DEM Building a DEM
The Data Extraction stage consists of the following modules: Module 3: DEM Operations Module 4: 3-D Feature Extraction
In Module 3, you learn how to extract a Digital Elevation Model (DEM) from four airphoto scenes. OrthoEngine needs a raster DEM to create an orthorectified image or photo. The objectives of this module are two-fold: To extract and geocode a DEM from pairs of stereo images To build a raster DEM from contour data
PCI Geomatics
65
Lesson 3.1
The objective of this lesson is to set up and convert four airphoto scenes into their epipolar projections. OrthoEngine is able to create epipolar images for multiple stereopairs in batch mode. You are able to add your stereopairs to a list and process them in one step. To complete this lesson you require: The airphoto scenes S129.pix, S130.pix, S188.pix, and S189.pix. The project file airphoto_model.prj, which contains the four airphotos from the Richmond Hill dataset. This project contains all the required GCPs and tie points and an up-to-date model.
Epipolar images are stereo pairs that are reprojected so that the left and right images have a common orientation, and matching features between the images appear along a common x axis. Using epipolar images increases the speed of the correlation process and reduces the possibility of incorrect matches.
Figure 3.1
Comparing raw images to epipolar images
Epipolar images are used in both extracting DEMs from stereo images and for three-dimensional stereo editing. In Module 4, you will edit features in 3-D stereo.
66
PCI Geomatics
Data Extraction
3D Viewing
3D Feature Extraction
Data Correction
Orthorectification
Mosaicking
Manual
Automatic
PCI Geomatics
67
2. On the DEM From Stereo toolbar, click Create Epipolar Image. The Generate Epipolar Images dialog box opens.
Figure 3.3
Create Epipolar Image dialog box
68
PCI Geomatics
You need to select the scenes to be used for the generation of the epipolar images. Left Image This area displays the candidates for the left-looking image of the stereo pair. All four airphotos are listed in the Left Image area when the dialog box opens. The Right Image area lists the candidates for the right-looking image. Because this project contains airphotos, the idea of Left and Right Images is not important for DEM extraction. You simply need to select the left and right images in order to create pairs of epipolar images. If you decide to proceed to 3-D feature extraction, which also uses epipolar pairs, you will save time if you generate pairs that can be use in 3-D viewing. If your photos are scanned north up, the photo that is geographically on the left is the left image, and the photo that is geographically on the right is the right image. To create the epipolar images: 1. From the Epipolar Selection list, select Maximum overlapping pairs. 2. For the Minimum Percentage Overlap, enter 60. This will automatically select pairs with a minimum of 60 percent overlap. 3. Click Add Epipolar Pairs To Table. The List of Epipolar Pairs lists two sets of epipolar pairs: S129 with S130 and S188 with S189.
Figure 3.4
Completed List of Epipolar Pairs
Right Image
PCI Geomatics
69
Note
It is recommended that you NOT exceed 50% of the available RAM. Specifying more than half may significantly reduce performance since the operating system needs RAM for its own operations.
5. Select a Down sampling factor of 2. This is the number of image pixels and lines that will be used to calculate one epipolar image pixel. For example, typing 2 means that two adjoining pixels and two adjoining lines will form one pixel in the epipolar image. The spatial detail of your resulting epipolar images will not be as high as your original imagery. You can adjust this value if you see noisy features in your image that you do not want to see in your DEM. 6. For the Down sample filter, select Average. 7. Click Save Setup. Save Setup saves the options chosen for batch processing with Automatic DEM Extraction. Generate Pairs begins the process based on the time set under Processing Start Time. Use this option if you are using the epipolar pairs for 3-D Feature Extraction. If you are using the epipolar pairs for Automatic DEM Extraction, you can either use this option or Save Setup. 8. Click Close.
Tip
This would be a good time to save your project file.
Now that you have setup to generate the epipolar pairs, you will set up for DEM extraction in the next lesson. The epipolar pairs and the DEM will be generated all at once.
70
PCI Geomatics
Lesson 3.2
This lesson describes how to extract, geocode, and edit a DEM. To complete this lesson you require the airphoto epipolar pairs generated in the previous lesson. A digital elevation model (DEM) is a digital file of terrain elevations for ground positions. It is a raster layer representing the elevation of the ground and objects, such as buildings and trees, with pixel values in the images. You can extract a digital elevation model (DEM) from stereo pairs of images, which are two or more images of the same area taken from different view points. This method can be very useful for creating a DEM for inaccessible areas. You can obtain stereo pairs from aerial photographs, digital or video images, and these sensors: ASAR, ASTER, IRS, IKONOS, SPOT, QUICKBIRD, and RADARSAT. OrthoEngine uses image correlation to extract matching pixels in the two images and then uses the sensor geometry from the computed math model to calculate x, y, and z positions.
Figure 3.5
Creating a DEM from stereo images
PCI Geomatics
71
To open the Automatic DEM Extraction dialog box: On the DEM From Stereo toolbar, click Extract DEM automatically. The Automatic DEM Extraction dialog box opens.
72
PCI Geomatics
The epipolar pairs you set up in the last lesson are listed in the Stereo Pair Selection table. To select the stereo pairs: In the Select column, click to select both stereo pairs. You can also click Select All to select all pairs that appear in the list. If the epipolar pairs do not exist or are not available, OrthoEngine will automatically generate the epipolar pairs using the options that you saved in the Generate Epipolar Images window.
Epipolar DEMs
The Epipolar DEM column specifies the output name for the extracted DEMs. Two epipolar DEMs will be created from the two sets of epipolar pairs. A text report is generated during the DEM extraction process. The report indicates the parameters used to extract the DEM as well as the correlation success. The names of the output DEM reports are specified in the DEM Report column.
DEM Report
PCI Geomatics
73
Extraction Options
The Extraction Options section contains the following options that govern the quality and resolution of the DEM extraction:
Failure value: This value is assigned to failed pixels within the extracted DEM.
Specifying a value assists the manual editing process. The default Failure Value is -100.
Background value: This value is used to represent No Data pixels in the DEM. The "No Data" or background identifies the pixels that lie outside the extracted DEM overlap area so they are not mistaken for elevation values. For DEMs generated by OrthoEngine, the background elevation defaults to -150. DEM detail: DEM Detail determines how precisely you want to represent the
terrain in the DEM. Selecting High, Medium or Low determines at which point in correlation process you want to stop. Low means that the process stops during the coarse correlation phase on aggregated pixels so the level of detail in the DEM will be quite low. High means the process continues until correlation is performed on images at full resolution.
Output DEM channel type: This option allows you to save the DEM in either a 16-bit signed channel or 32-bit real channel. Pixel sampling interval: The Pixel Sampling Interval sets the frequency of
samples taken in the processing of the DEM extraction. It also controls the size of the pixel in the final DEM relative to the input images. The higher the number you choose, the larger the DEM pixel will be, and the faster the DEM is processed.
Use clip region: This option allows you to process only the area determined in
the Define Clip Region window, which results in smaller DEMs and faster processing.
Fill holes and filter: This option will enhance the output quality of the DEM by
interpolating the failed areas and filtering the elevation values automatically.
Delete Epipolar Pairs after use: This option deletes the epipolar pairs from the disk to save space after the DEM is generated.
74
PCI Geomatics
To set the extraction and geocoding options: 1. For the Minimum elevation, enter 150. 2. For the Maximum elevation, enter 250. 3. In the DEM detail list, select Medium. 4. Set the Pixel sampling interval to 2. Every second pixel is sampled and processing time is reduced. 5. Select the Fill holes and filter option. 6. Select the Create score channel option. 7. Select the option to Create Geocoded DEM. This will geocode and merge the extract epipolar DEMs in one step. 8. For the Output File name, enter GeocodedDEM.pix. 9. For the DEM Bounds, select All Images. This will use the extent of all the images in the Stereo pairs table as the extents for the DEM. 10. For the Output Options, select Highest Score. Output Option As the epipolar DEMs are extracted and geocoded, they are added to the geocoded DEM file. When a new geocoded DEM is added to the file and it overlaps an existing geocoded DEM, you must choose a method to determine which pixel value will be used. There are three methods:
Use last value: To replace the pixel values in the overlap area in the existing
geocoded DEM by the pixel values of the geocoded DEM being added to the file.
Average: To replace the pixel values in the overlap area by the average pixel
values between the existing geocoded DEM and the one being added to the file.
Highest Score: To replace the pixel values in the overlap area by the pixel value
with the highest correlation score between the existing geocoded DEM and the one being added to the file. This option is only useful if you select Create Score Channel.
PCI Geomatics
75
To extract the DEM: 1. Click Extract DEM. A Progress Monitor opens which shows the status of the extraction process. After the extraction is complete, the Progress Monitor closes. A message box opens with the message DEM Extracted successfully. 2. Click Close. To view a report of the resulting DEM file: 1. From the AIRPHOTO folder, open S129_S130_dem.rpt and S188_S189_dem.rpt in a Text Editor. 2. View the report to see how the extracted DEM elevations compare with the elevations that were entered with your GCPs.
Note
Even though you chose to interpolate failed areas during DEM extraction by selecting the Fill Holes & Filter field in the Extraction Options section, you may still need to manually edit the DEM.
76
PCI Geomatics
To open the 2D DEM Editing dialog box: 1. From the DEM From Stereo toolbar, click Manually edit generated DEM. A Focus window and the DEM Editing dialog box open.
Figure 3.8
DEM Editing dialog box
2. Click Browse. 3. From the AIRPHOTO folder, select GeocodedDEM.pix and click Open. 4. From the Layer list, select channel 2. The DEM is loaded in the Focus view area. 5. For the current project, click Close at the bottom of the 2D DEM Editing dialog box.
Note
DEM editing is described in more detail in the ASTER exercise of the OrthoEngine Workbook.
PCI Geomatics
77
78
PCI Geomatics
Lesson 3.3
Building a DEM
This lesson describes how to build a raster DEM from contour data. To complete this lesson you require: The project file airphoto_model.prj, which contains the four airphotos from the Richmond Hill dataset. This project contains all the required GCPs and tie points and an up-to-date model. The file ap_contours.pix that contains contour information in a vector layer
OrthoEngine can calculate the elevations from vector layers to generate a raster digital elevation model (DEM), which is saved as a PCIDSK file (.pix). OrthoEngine uses raster DEMs to orthorectify images. If your elevation data is stored as vectors such as contours, points, TIN, or even a text file containing coordinates, you can convert them into a raster DEM as long as the vectors are in any of the supported formats.
Note
You can combine vectors from different layers and files to generate a DEM.
Vector layers can contain: Points: A point is a single coordinate (x, y, and z). Lines: A line is a start and end coordinate with points in between to define the shape. Polygons: A polygon is a line with the same start and end coordinate forming an area with numerous points along the line to define its size and shape. Contours: A contour is a line formed by a set of points representing the same value of a selected attribute. Contours are usually used to represent connecting points on the ground with the same elevation. TIN: A Triangulated Irregular Network (TIN) is a digital model of adjoining triangles formed from points selected on the terrain to represent an accurate model of the surface. The TIN model can contain coordinates and other geographical data.
PCI Geomatics
79
Data Extraction
3D Viewing
3D Feature Extraction
Data Correction
Orthorectification
Mosaicking
Manual
Automatic
80
PCI Geomatics
2. On the Import & Build DEM toolbar, click DEM from contours. The Input Vector Layer Selection dialog box opens.
Figure 3.11
Input Vector Layer Selection dialog box
To import the vector file to generate the DEM: 1. In the Input Vector Layer Selection dialog box, click Select. 2. From the AIRPHOTO folder select ap_contours.pix and click Open. The available vector layers are listed. 3. Under Vector layer available, select 1 [VEC]: Contour and click the arrow. The selected layer appears under Vector layers to interpolate.
PCI Geomatics
81
Note
You can have vector layers containing vectors, points, contours, or TIN from a variety of different files in the list of vectors to be interpolated.
4. In the list of Vector layers to interpolate, select the 1 [VEC]: Contour layer. 5. In the Data type list, select Contours. This is the type of vector contained in this layer. 6. For the Elevation source, select ELEVATION. This is the attribute field that stores the elevation values.
Figure 3.12
Input Vector Layer Selection Window
7. Click OK. The Input Vector Layer Selection dialog box closes and the Define Output DEM File dialog box opens.
82
PCI Geomatics
To define output DEM file: 1. In the Output DEM box, enter BuiltDEM.pix. 2. To generate a DEM that covers the area where elevation data exists, click Elevation Source Area. Mosaic Area will generate a DEM that covers the area defined by the Mosaic Area. Photo Extents will generate a DEM that covers the extents of all the images in the project. This is useful when you want the DEM to cover the images being orthorectified, but extrapolating beyond the elevation source area can cause significant errors in your project. 3. Click Generate DEM. The Iteration Values dialog box opens. 4. Use the default values of 10 Iterations and a Tolerance of 1.00. The No. of Iteration is the maximum number of times that the DEM is smoothed. The Tolerance is the minimum difference in value required during smoothing to warrant another application. The output DEM opens in a viewer.
PCI Geomatics
83
Checkpoint
Module 4
You can now proceed to Module 4: 3-D Feature Extraction. This is an optional step in the Data Extraction stage. The lessons in Module 4 show you how to view stereo image pairs using an anaglyph display, extract features in three dimensions, and edit a DEM using 3-D feature extraction.
Module 5
If you prefer, you can go directly to Module 5: Orthorectification. In this module, you learn how to set up and correct your images or photographs for distortions due to camera tilt and terrain relief.
84
PCI Geomatics
Module
4
Module 4 has Two Lessons:
Lesson 4.1 Lesson 4.2
The Data Extraction stage consists of the following modules: Module 3: DEM Operations Module 4: 3-D Feature Extraction
Images from the ASAR, ASTER, EROS, FORMOSAT 2, IKONOS, IRS, QuickBird, RADARSAT and SPOT satellites normally have viewing geometries that are suitable for the creation of a stereo pair. For this module, continue using the airphoto_model.prj project file and the epipolar images created in Lesson 3.1.
The 3-D tools are used to: Create epipolar pairs for improved stereo viewing Convert an extracted or third-party DEM to points or vectors for interactive editing using a pair of stereo images Update vector data bases using new stereo imagery Develop vector data bases through photo interpretation, in stereo
PCI Geomatics
85
Data Preparation
Create a Project Set Interior Orientation Input Data Collect Photo Fiducial Marks
Data Extraction
3D Viewing
3D Feature Extraction
Data Correction
Orthorectification Mosaicking
Manual
Automatic
86
PCI Geomatics
Lesson 4.1
PCI Geomatics
87
Because you created the epipolar pairs in preparation for DEM extraction, there is no need to recreate them. You simply need to select the stereo pair. To select and load a 3-D stereo pair: 1. On the OrthoEngine window in the Processing step list, select 3-D Operations. 2. Click 3-D Feature Extraction. The 3-D Stereo Pair Selection dialog box opens. Here, you select the left and right stereo images for display in the 3-D viewer. The crosshairs in the Project Pair Overview window represent the centers of the uncorrected images.
Figure 4.1
3-D Stereo Pair Selection dialog box
3. In the 3-D Stereo Pair Selection dialog box under Project Pair Overview, click the crosshairs of the S129 photo. This image will be displayed to your left eye. All the crosshairs of the images that overlap the selected image appear in blue under Stereo Pair Detail. 4. Under Stereo Pair Detail, click the crosshairs of the S130 photo. This image will be displayed to your right eye. 5. Select the Epipolar image option and click Load Pair.
88
PCI Geomatics
If epipolar images exist for the pair that you selected, the option Epipolar image will become available. Using epipolar images while editing in 3-D stereo produces a sharper stereo view over the entire area of the image, and reduces the need to manually align the images.
Figure 4.2
3-D Stereo Pair Selection after the images are selected
Note
Having the stereo images in the wrong order (left image on the right and vice versa) will not affect processing, but it will cause some visually disturbing effects if you try to view the epipolar pair in 3-D. Our brains are trained to interpret the images seen from our left and right eyes in a certain way. If the images are in the wrong order, we cannot compensate for the error so, for example, we may see mountains that look like valleys instead. If this occurs, select the images in the opposite order.
Anaglyph Mode
This mode displays one image in red and the other in blue. To view the images in 3-D, you must wear a pair of 3-D glasses, which has a red lens for the left eye and a blue lens for the right eye. This mode displays in a single window the images alternately in black and white or in color. Special stereo hardware such as specialized graphics cards and shutter glasses or polarizing monitors are used to filter the images so that each image is only displayed to the appropriate eye. The OpenGL display method provides a more natural stereo viewing experience than the Anaglyph method, but it requires more expensive hardware.
OpenGL Mode
PCI Geomatics
89
6. In the Stereo Mode list, select Anaglyph. The Stereo Mode list allows you to select the type of viewing display that you are using. OrthoEngine will automatically search for the OpenGL libraries, but not the required stereo viewing hardware. If the OpenGL libraries are present, both Anaglyph and OpenGL options will be available. 7. Click Load Pair. The 3-D Feature Extraction dialog box opens with the stereo images displayed in the viewer on the right.
Figure 4.3
3-D Feature Extraction dialog box
90
PCI Geomatics
To align the images: 1. Move the mouse over the viewer to make the image active. 2. Use the keyboard controls in the table below to shift the right image to align with the left image:
Table 1: Keyboard controls for parallax adjustment Key I J K M Action Shifts the right image up Shifts the right image left Shifts the right image right Shifts the right image down
To switch between the pointer and the stereo cursor: 1. Move the pointer to the viewer and press ESC. The pointer disappears and the stereo cursor is activated. 2. Move the stereo cursor to a feature in the image and press ESC. The stereo cursor remains on the feature and the pointer reappears.
PCI Geomatics
91
To move the stereo cursor pixel by pixel: 1. Press the keyboard Arrow keys to move the screen cursor screen pixel by screen pixel. 2. Press CTRL+ the Arrow keys to move the stereo cursor 10 screen pixels at a time. To zoom to a location in the stereo viewer: 1. Position the stereo cursor at the desired location. 2. Press the ESC key to activate the pointer. 3. On the viewer toolbar, click Zoom In. 4. Press the ESC key to reactivate the stereo cursor. Alternatively, while using the stereo cursor, click Page Up or Page Down.
To change the elevation of the stereo cursor: 1. To move the cursor up, press the Z key. 2. To move the cursor down, press the X key. 3. If you have a track wheel on your mouse, roll the track wheel away from you to move the cursor up, and towards you to move the cursor down. Moving the cursor up or down is a result of changing the relative offset, or parallax, between the left and right halves of the cursor. Each time a key is pressed on the keyboard or a track wheel is rotated one revolution on the mouse, the left or right half of the cursor shifts by one pixel on the screen. To automatically snap the stereo cursor to the ground: Press the G key. This automatically moves the stereo cursor to the three-dimensional surface (z coordinate) of a feature at the cursor's current position, such as a field, a roof top, or a tree top.
92
PCI Geomatics
Reducing Eyestrain
Many people experience difficulty such as eyestrain and fatigue when they work in a 3-D environment. The following are a few tips to avoid discomfort:
Normally, the cursor and the screen share the same focus. In the 3-D viewer, however, the cursor can move not only in x and y, but also in z. This means that the cursor can appear as if it is floating off the screen or behind the screen.
To refocus and center images in the viewer: Recenter the Images Press F4.
Working in the middle of the 3-D viewer is easier on your eyes. Since the 3-D viewer and the normal screen display information differently, your eyes may have difficulty adjusting to both environments at the same time.
The following table summarizes the controls for the stereo cursor.
Table 2: Keyboard controls for the stereo cursor in 3-D Key Z X CTRL + Z CTRL + X G F R CTRL + Left click CTRL + Right click Action Moves the cursor up one pixel on the Z axis Moves the cursor down one pixel on the Z axis Moves the cursor up ten pixels on the Z axis Moves the cursor down ten pixels on the Z axis Moves the stereo cursor to ground level automatically Refocus the stereo view at the current elevation Reloads the imagery centered around the cursor position Zoom tool in stereo cursor mode Zoom tool in stereo cursor mode
PCI Geomatics
93
Lesson 4.2
This lesson describes how to load existing vector layers, create new vectors, and then edit the vectors in 3-D. To complete this lesson you require: The 3Dvector.pix file from the AIRPHOTO folder.
You will begin by loading existing vector layers. To load vector layers: 1. In the Vector Layer Information area, click Load Layer. 2. From the AIRPHOTO folder, select 3Dvector.pix and click Open. 3. From the Database Vector Segment list, select 2(Vector): Schools and click Load. The vector layer is added to the table in the Vector Layer Information section. 4. Select 3(Vector): Roads and click Load & Close. The Database Vector Segments dialog box closes and the vector layer is added to the table. The vector layers represent a major road and school property boundary for the overlap area between S129 and S130. The highlighted row in the table represents the active layer.
94
PCI Geomatics
Active Layer
A layer is active when it is selected in the Vector Layer Information table. Properties such as priority, visibility, color, and description can only be modified if the layer is active. To make the Schools layer active: In the Vector Layer Information table, click the Schools vector layer. The Schools row is now highlighted. You can see the attribute information associated with this selected layer in the Attribute table.
Priority
When two or more vector layer overlap, it may be difficult for you to identify the vectors that you want. The layers in the 3-D viewer are displayed one on top of the other in the same order as they appear in the Vector Layer Information table. The topmost row in the Vector Layer Information table contains the layer with the highest priority, which means that its vectors are not covered by any other layer.
To change the priority of the Schools layer: 1. In the Vector Layer Information table, select to the Schools layer. 2. Click the Change priority arrow to move Schools to the top of the list. Visibility
You can display or conceal vector layers individually or collectively. The check mark in the Visible column in the Vector Layer Information table indicates that a layer is displayed in the viewer.
To change the visibility of the active layer: Right-click in the Visible column for the Schools layer. The vectors are concealed. Color
You can select the color of each vector layer. Most of the colors are compatible with the anaglyph viewer, however, some colors are easier to see than others.
To change the color of the active layer: 1. Right-click in the Color column for the active layer. 2. Click the desired color. The new color is applied to the active layer.
PCI Geomatics
95
To create a new layer: 1. In the Vector Layer Information section, click New Layer. The Create New Layer dialog box opens. 2. In the Name box, enter Layer 1. 3. In the Description box, enter New Vectors. 4. Click Accept. The Create New Layer dialog box closes. In addition, the new layer is loaded into the project, added to the top of the table, and also becomes the active layer.
Note
The default georeferencing is automatically applied based on the stereo pair listed in the Stereopair Selection section. By default, the projection of the new layer is set to the projects output projection, and the bounds of the layer are approximated based on the imagery currently loaded in the viewer.
Figure 4.4
3-D Stereo dialog box after vector layers are added
96
PCI Geomatics
To add lines to the selected layer: 1. Select Layer 1 to make it active. 2. In the Operation section, click New Line. 3. Move the mouse pointer to the 3-D viewer and press ESC to switch to the stereo cursor. 4. Position the stereo cursor where you want to begin the line. 5. Adjust the z position of the stereo cursor by pressing the G key. Your stereo cursor automatically locates at the ground elevation at the position of the stereo cursor. 6. Click to anchor the first vertex. Beneath the 3-D viewer, you can see the x, y, and z coordinates of the stereo cursor's location. 7. Move the stereo cursor to the next position. 8. When you are satisfied with the x and y positions, adjust the z position and click to anchor the vertex. 9. Repeat steps 7 and 8 until you have collected the vertices you need to created the desired line. 10. Click Accept, double-click or press ENTER to complete the line. 11. Click New Line to stop collecting lines.
Snap to Vertex places the cursor on the vertex nearest to where you click. You can use it to quickly position new vertices or to move existing vertices on the same layer.
Tip
When working with the stereo cursor in the 3-D viewer, you can press V to activate and deactivate Snap to Vertex.
To snap to a vertex: 1. In the Operation section, click Snap to Vertex. 2. Position the stereo cursor close to an existing vertex and click. The cursor aligns with the vertex nearest to where you clicked.
PCI Geomatics 97
3. Once you have snapped to the vertex, click Snap to Vertex to turn off the feature. 4. Continue to collect vertices for the line. 5. Click Accept, double-click or press ENTER to confirm the completion of the line. Using Snap to Line
Snap to Line places the cursor on the line nearest to where you click. You can use it to quickly position new vertices or to move existing vertices on the same layer.
Tip
When working with the stereo cursor in the 3-D viewer, you can press C to activate and deactivate Snap to Line.
To snap to a line: 1. Click Snap to Line. 2. Position the stereo cursor close to an existing vector and click. The cursor aligns with the vector nearest to where you clicked. 3. Once you have snapped to the line, click Snap to Line to turn off the feature. 4. Continue to collect vertices for the line. 5. Click Accept, double-click or press ENTER to confirm the completion of the line. Collecting 3-D Points To add points to the selected layer: 1. In the Operation section, click New Point. 2. Position the stereo cursor where you want to collect a point. 3. Adjust the z position of the stereo cursor by pressing the G key. Your stereo cursor automatically locates at the ground elevation at the position of the stereo cursor. 4. Click to create a point. 5. Repeat steps 2 to 4 to collect additional points. 6. Click New Point to stop collecting points.
Note
New points are automatically accepted and are displayed in the color associated with the vector layer.
98
PCI Geomatics
To add polygons to the selected layer: 1. In the Operation section, click New Poly. 2. Position the stereo cursor where you want to begin the polygon. 3. Adjust the z position of the stereo cursor by pressing the G key. Your stereo cursor automatically locates at the ground elevation at the position of the stereo cursor. 4. Click to anchor the first vertex. Beneath the 3-D viewer, you can see the x, y, and z coordinates of the stereo cursor's location. 5. Move the stereo cursor to the next position. 6. When you are satisfied with the x and y positions, adjust the z position and click to anchor the vertex. 7. Repeat steps 5 and 6 until you have collected the vertices you need to created the desired polygon. 8. Click Accept, double-click or press ENTER to confirm the completion of the polygon. Polygons automatically close when they are accepted. 9. Click New Poly to stop collecting polygons.
Selecting a Vector
To select a vector: 1. Move the pointer into the 3-D viewer and press ESC to switch to the stereo cursor. 2. Click the vector that you want to select or click the vector in the Attribute table. The vector is highlighted and the stereo cursor is placed on the nearest vertex. Only one vector can be selected at a time. 3. Right-click in the viewer to clear the selection.
Inserting a Vertex
If you add a vertex to the existing line or polygon, the new vertex is added at the location where you click. It will connect to the last selected vertex and the next vertex following the sequence that the original vector was created. If the last selected vertex is the end point of a line, the vertices that you add will extend the line. If you add a vertex to a point, the point is converted into a line.
PCI Geomatics
99
To insert a vertex: 1. Move the pointer into the 3-D viewer and press ESC to switch to the stereo cursor. 2. Click the vertex that precedes the location where you want to place the new vertex. 3. Click Insert Vertex or press INSERT. 4. Position the stereo cursor where you want to place the new vertex. 5. Adjust the z position of the stereo cursor by pressing the G key. 6. Click to insert one or more vertices. If the vector is not reacting as expected, click Undo and then Accept. Try the process again by selecting the vertex beside the one you selected before. Remember that vertices inserted into the vector are added following the sequence that the original vector was created. Deleting a Vertex To delete a vertex: 1. Select a vertex. 2. Click Delete Vertex or press D.
The vertex is deleted, the vector adjusts automatically, and the action is accepted.
Deleting a Line or Polygon To delete a line or polygon: 1. Select a vector. 2. Click Delete Line or press DELETE. The vector is deleted, and the action is accepted. Moving a Vertex or a Point To move a vertex: 1. Move the pointer into the 3-D viewer and press ESC to switch to the stereo cursor. 2. Select a vertex. 3. Click Move or press B. 4. Position the stereo cursor precisely on the location where you want to place the vertex. 5. Adjust the z position of the stereo cursor. 6. Click to move the selected vertex. 7. Click Accept or press ENTER to accept the move.
Tip
You can also use Move to adjust the elevation of a vertex.
100
PCI Geomatics
Reversing an Action
You cannot cancel actions that delete vectors or vertices. Undo only functions for actions that have not been accepted.
Before you can add attributes to a vector, you must define the columns in the table. To add a new field to the Attribute table: 1. Click Fields. 2. In the Vector Field Definition dialog box, click Add Field. 3. In the New Field dialog box, click in the Name box and enter a name for the attribute column. 4. In the Description box, type a brief sentence describing the column. 5. In the Data type list select: Integer to define the field as containing positive or negative whole numbers Double to define the field as containing double-precision real numbers Float to define the field as containing single-precision real numbers Text to define the field as containing a string of text Integer List to define the field as containing a list of positive or negative whole numbers separated by commas. 6. Click OK. 7. The new column is added to the Attribute table. 8. Repeat steps 3 to 6 to add another column or click Cancel. 9. On the Vector Field Definition window, click Close. Now that you have added a new field to the Attribute table, you will enter attributes for vectors in the selected layer.
PCI Geomatics
101
To enter values in the Attribute table: 1. In the Vector Layer Information table, select the Layer 1. 2. In the Attribute table, select a vector, or select a vector in the viewer. The vector is highlighted in the 3-D viewer and the Attribute table. 3. Click in the Road Name column for the selected shape. 4. Type in the attribute value and press ENTER or click Accept. 5. Repeat steps 2 to 4 to enter additional attribute values.
Saving a Layer
You can add a new layer to an existing file or you can save the layer in a PCIDSK (.pix), an ESRI Shape File (.shp), or an AutoCad (.dxf) format. The layers that appear under Savable Vector Layers are the layers that you have available in the stereo viewer. The layers that appear under Database Vector Segments are the layers that are saved in the file. If you close the stereo viewer, unsaved layers are discarded.
To save the layer: 1. In the Vector Layer Information section, click Save Layer. Unless you specify a new file, the vector layer will be saved to the 3DVECTOR.pix file, which is currently loaded. 2. Click New File. The File Selection dialog box opens. 3. For the File name, enter new3Dvectors.pix. You can save to a PCIDSK file, an ESRI shape file or an AutoCad DXF file. 4. Click Open. A question window opens asking if you want to create the new file. 5. Click Yes. 6. From the list of Savable Vector Layers, select the layer you want to save. 7. Click Save & Close. The edited layer has now been save to a new file.
102
PCI Geomatics
Orthorectification
Module
5
Module 5 has One Lesson:
Lesson 5.1
The Data Correction stage consists of the following modules: Module 5: Orthorectification Module 6: Mosaicking
The objective of this module is to correct a set of airphotos so the resultant orthophotos can be used in the mosaicking process in Module 6: Mosaicking.
PCI Geomatics
103
Data Preparation
Create a Project Set Interior Orientation Input Data Collect Photo Fiducial Marks
Data Extraction
3D Viewing
3D Feature Extraction
Data Correction
Orthorectification Mosaicking
Manual
Automatic
104
PCI Geomatics
Lesson 5.1
This lesson describes how to set up and perform orthorectification using aerial photographs. For this lesson you require: The airphoto_model.prj file from Lesson 3.2, which contains the four airphotos from the Richmond Hill dataset. The project contain all the required GCPs and tie points and an up-to-date model. The ap_dem.pix DEM file.
Orthorectification is the process of using a rigorous math model and a digital elevation model (DEM) to correct distortions in raw images as shown in the figure below. The rigorous math models, such as the Aerial Photography or Satellite Orbital math models, provide a method to calculate the position and orientation of the sensor at the time when the image was taken. The DEM is a raster of terrain elevations. The quality of the orthorectified image is directly related to the quality of the rigorous math model and the DEM. A poorly computed math model, an inaccurate DEM, or a DEM incorrectly georeferenced to the math model will cause errors in the orthorectified images.
Figure 5.1
Using sensor geometry and a DEM to orthorectify imagery
PCI Geomatics
105
2. Click Schedule ortho generation. The Ortho Image Production dialog box opens.
Figure 5.3
Ortho Image Production dialog box
3. Under Available images, use the SHIFT key to select all four photos and click the arrow button to move the images under Images to process. The images are processed in the order that they are listed. 4. Under Images to process, select S129. By default, the ortho image will be named oS129.pix. You could also enter a different filename in the Ortho Image section.
106 PCI Geomatics
To select and load the DEM: 1. Under DEM, click Browse. 2. From the AIRPHOTO folder, select ap_dem.pix and click Open. The Database Channels dialog box opens. 3. For the Background elevation, enter -150 and click OK. This represents No Data pixels in the DEM. For DEMs generated by OrthoEngine, the background elevation defaults to -150. Other DEMs have different background elevation values that you must know before they can be used. If you do not know the background value, click DEM Info in the DEM File dialog box. The window displays the three lowest and three heights values in the DEM.
Note
It is critical that the background elevation be set to the correct value, if there are areas of no data in the DEM.
Elevation Scale: This is used to convert the pixel values in a digital elevation
model (DEM) into their correct elevation value. For example: since an 8-bit channel can only contain integers between 0 and 255, you may have a DEM that was multiplied by 10 to maintain the decimal precision of its elevation values. A DEM pixel may have a value of 102, but the actual elevation that it represents is 10.2. To convert the DEM pixel value from 102 to 10.2 you must multiply it by 0.1. Therefore, you type 0.1 in the Elevation scale box to convert the DEM pixels back to their true values.
Elevation Offset: This is used to add a value to the pixel values in a DEM to
obtain their actual elevation value. Perhaps the DEM pixel with a value of 102 actually represents an elevation value of 1,102. To store the elevation values in an 8-bit channel, 1,000 was subtracted from all the pixel values when the DEM was created. Therefore, you must type 1,000 in the Elevation offset box to restore the true values. You can also use Elevation offset to adjust the elevation reference of a DEM. The elevations in a DEM can be calculated above Mean Sea Level or an ellipsoid. The elevation reference in the DEM must match the elevation reference of the imagery that you want to orthorectify. To compensate for a discrepancy, you can type the difference between the two elevation references in the Elevation offset box.
Tip
If you do not have a DEM, you can use the average elevation of an area to orthorectify the image. However, this will not produce results as accurate as using a DEM. Type the average elevation in the Elevation offset box.
PCI Geomatics
107
Working Cache
This is the maximum amount of RAM that you allocate for the ortho generation process. The limit should not be more than half the available RAM. Specifying more than half may significantly reduce performance. To set up the working cache: Enter an amount of RAM appropriate for your computer.
Sampling Interval
The Sampling Interval controls how the computations are performed when an image is orthorectified or geometrically corrected. When an image is corrected, OrthoEngine selects a pixel from the output file, computes the elevation from the DEM (if available), applies the math model to determine which pixel location it corresponds to in the raw image, and then transfers the data to the pixel in the output file. The Sampling Interval determines how many output pixels are computed following this method. A Sampling Interval of 1 means that the position of every output pixel is processed. To speed up the process, you can increase the Sampling Interval. Note that this sampling interval interpolates the position of the pixels, not the intensity of the pixels. To set up the sampling interval: Enter a value of 4. This means that the correction for every fourth pixel is calculated and the correction for the pixels in between are interpolated.
Resampling extracts and interpolates the gray levels from the original pixel locations to the corrected locations.
Nearest Neighbour: Identifies the gray level of the pixel closest to the specified input coordinates and assigns that value to the output coordinates. Although this method is considered the most efficient in terms of computation time, it introduces small errors in the output image. The output image may be offset spatially by up to half a pixel, which may cause the image to have a jagged appearance. Bilinear Interpolation: Determines the gray level from the weighted average of the four closest pixels to the specified input coordinates and assigns that value to the output coordinates. This method generates an image with a smoother appearance than Nearest Neighbor Interpolation, but the gray level values are altered in the process, which results in blurring or loss of image resolution. Cubic Convolution: Determines the gray level from the weighted average of the
16 closest pixels to the specified input coordinates and assigns that value to the output coordinates. The resulting image is slightly sharper than one produced by Bilinear Interpolation, and it does not have the disjointed appearance produced by Nearest Neighbor Interpolation. To select the resampling method:
108
This option removes a specified percentage of the images outside edge. You can use this option to remove unwanted areas such as the data strip and fiducial marks from aerial photographs or a dark perimeter or distortion along the edge of the image. You have two choices for the ortho generation: You can generate your orthos now You can generate your orthos at a later time
If you intend to automatically mosaic the processed images, you can click Close instead of Generate Orthos. When you set up the Automatic Mosaicking window, select Regenerate offline orthos, and OrthoEngine will process the images and mosaic them in one step. In this lesson, you will generate the orthos now. To generate the ortho photos now: 1. Click Generate Orthos. The Ortho Production Progress monitor opens and shows the status of the orthorectification process for each photo. After all the orthos are generated, the following message appears: All Processing Completed. 2. Click Close. The message Ortho done appears beside each photo in the Available images section, indicating that the original photos are now orthorectified. The files containing the corrected photos are named oS129.pix, oS130.pix, oS188.pix, and oS189.pix.
Tip
This is a good time to save your project file.
PCI Geomatics
109
Checkpoint
Module 6 Automatic Mosaicking Module 6 Manual Mosaicking
In Module 6 you will define an area for mosaicking, and then automatically mosaic the orthorectified photos contained in your project file.
If you would like more control over the mosaicking of your corrected photos, or if you do not have the OrthoEngine Productivity Tools package, proceed to the manual mosaicking lesson in Module 6. In Module 6 you define an area for mosaicking, and then interactively mosaic the orthorectified photos contained in your project file.
Additional Operations
If you need additional operations to be performed on your photographs, proceed to Geomatica for cartographic production and quality assurance applications.
110
PCI Geomatics
Mosaicking
Module
6
Module 6 has Three Lessons:
Lesson 6.1 Lesson 6.2 Lesson 6.3
The Data Correction stage consists of the following modules: Module 5: Orthorectification Module 6: Mosaicking
Mosaicking is joining together several overlapping images to form a uniform image as shown in the figure below. It is similar to creating a jigsaw puzzle with your images, and then making the seams disappear. For the mosaic to look like one image instead of a collage of images, it is important that the images fit well together. You will achieve better results if you orthorectify your images. Using a rigorous math model ensures the best fit not only for the individual images, but for all the images united as a whole. To achieve that seamless look in the mosaic, place the seams, called cutlines, where they will be the least noticeable and select images or portions of images that are not radically different in color.
PCI Geomatics
111
Manual Mosaicking
You can use Manual Mosaicking to create your mosaic one image at a time, to edit the cutlines in an automatically mosaicked project or to replace unsatisfactory areas in the mosaic. For each image that you want to include in the mosaic file, you must complete four steps in sequence; select an image to add, collect the cutline, adjust the color balance and add the image to the mosaic area.
Automatic Mosaicking
Although you can create your mosaic one image at a time by using Manual Mosaicking, most of the time you will use Automatic Mosaicking to do the bulk of the work, and you will use Manual Mosaicking to edit portions of the mosaic file. Some projects may require more editing than others such as those containing large bodies of water or urban areas with buildings leaning in different directions. In addition to reducing your work load, Automatic Mosaicking will often produce a more seamless look than if you had attempted to create the mosaic by hand.
Both manual and automatic mosaicking will be described in detail in this module although as stated above, automatic mosaicking would generally be used to do the bulk of the work. Manual mosaicking could then be used to edit the cutlines.
112
PCI Geomatics
Data Preparation
Data Extraction
3D Viewing
3D Feature Extraction
Data Correction
Orthorectification Mosaicking
Manual
Automatic
PCI Geomatics
113
Lesson 6.1
To define an area for mosaicking, you require: The file airphoto.prj that you used in Module 5 to generate the ortho images. Alternatively, you can open airphoto_ortho.prj from the AIRPHOTO folder. This project contains all the required ground control points and tie points, and the photos are already orthorectified.
The objective of this lesson is to set up an area that includes a part of each of the four orthophotos. The defined area is used in the mosaicking process.
Before you create a mosaic, you need to define an empty mosaic file. To define the Mosaic Area: 1. On the OrthoEngine window in the Processing step list, select Mosaic. A new toolbar with four icons appears. The toolbar contains functions for defining a mosaic area, manual mosaicking, reapply mosaicking, and automatic mosaicking.
Figure 6.2
Mosaic toolbar
2. On the Mosaic toolbar, click Define mosaic area. The Define Mosaic Area dialog box opens.
114
PCI Geomatics
By default the bounds of the Mosaic Area are the maximum extents of the images in the project. The size of the mosaic area can be changed manually using your mouse. 3. Place the cursor over the side or corner of the frame and move it to change its size and shape. Alternatively, you can enter corner coordinates for the mosaic area. Define a broad area that includes part of each photo in the project.
Figure 6.4
Define Mosaic Area dialog box after the area has been adjusted
PCI Geomatics
115
Tip
If you want to start over and define a new mosaic area, click Define New Mosaic Area.
4. Click Create Mosaic File. The File to Create dialog box opens. 5. For the name of the file, enter manual_mosaic.pix and click Create. OrthoEngine creates your new mosaic file and stores it in the folder where your project is saved. 6. Click Close. You are now ready to mosaic your photos. With OrthoEngine there are two methods for generating a mosaic: Manual Mosaicking Manual mosaicking Automatic mosaicking
First, use the manual process in Lesson 6.2: Manual Mosaicking to create a seamless final product from the four orthophotos. Here, you use a set of manual tools, which affords more precise control. After this process is complete, proceed to Lesson 6.3: Automatic Mosaicking and mosaic the same set of orthophotos using the automatic tools.
Automatic Mosaicking
116
PCI Geomatics
Lesson 6.2
Manual Mosaicking
To create the mosaic from the orthophotos, you require: The project file from Lesson 6.1 An empty mosaic file
PCI Geomatics
117
Mosaicking Steps
Four Mosaicking Steps are listed in the Manual Mosaicking dialog box and must be completed in sequence. They are: Select image to add Collect cutline Color balance Add image to mosaic
Images within the current project are listed in the Project Image Files table. The ID and Status for each image is shown. Images are identified by a crosshair within the viewer. The crosshairs are color-coded to represent the centers of images that are: Available - blue crosshair Mosaicked - blue crosshair Off-line - gray crosshair
118
PCI Geomatics
Mosaic Preview
To activate the mosaic preview: Select Show mosaic preview. Changes to the Blend Width update the Mosaic Preview window. The image to be mosaicked is displayed with the specified blend width.
PCI Geomatics
119
To collect cutlines: 1. Under Mosaicking Steps, click Collect cutline. The Cutline information section appears. 2. Using the zoom tools, zoom to a road that runs through the area of overlap and position your cursor where you want to begin collecting the cutline. 3. Under Cutline information, click Add. You are now in Add mode. 4. Click in the viewer to outline a polygon around the area that you want to include in the mosaic. Draw the cutline in the overlapping area displayed in gray and the area of the input image that appears in cyan. 5. When you are finished the cutline, click Add. You are no longer in Add mode. 6. Click Finish. The cutline polygon is automatically closed. To edit a cutline: 1. Select a vertex in the viewer or in the Cutline information table. 2. The following editing tool are available: Click Delete to deleted the selected vertex. Click Move and click in the viewer to reposition the selected vertex. Click Insert and click to place a new vertex after the selected vertex Click New to clear the existing cutline and start again
3. Click the same button again to end the edit. Blend Seams Blending reduces the appearance of seams by mixing the pixels values on either side of the cutline to achieve a gradual transition between the images. In Automatic Mosaicking OrthoEngine blends the seams automatically. In Manual Mosaicking the Blend Width determines the number of pixels on either side of the cutline that are used to blend the seam. However, in areas containing bright or significantly different features, setting the Blend Width too high may cause "ghosting" or doubling of the features. To set the Blend width: 1. In the Blend width list, type 3. A Blend Width of three to five pixels is recommended. You can see the results of the blend in the Mosaic Preview window.
120
PCI Geomatics
To create match areas: 1. Under Mosaicking Steps, click Color balancing. The Match Areas section appears. 2. Under the Match Areas section, click New Area. 3. Using your mouse, drag a regulate in the overlap area. The match area is defined by your box. 4. To define another match area, click New Area. 5. Select Show mosaic preview to see how the histogram affects the image that you are adding to the mosaic. 6. To save the lookup table with the image, click Save Working. 7. To display the lookup table and the match areas that were saved with the image, click Show Saved. 8. To display the lookup table kept in memory (Working state), click Show Working. Define several match areas in the above manner. Any number of match areas can be defined, and are displayed in the viewer. The areas are only active during the current editing session As you collect Match Areas or when you import a lookup table, the Mosaic Preview window displays the changes in the color balancing. The changes are kept in memory referred to as the Working state. If you select new Match Areas and click Save Working, the previous lookup table in the project file is replaced.
PCI Geomatics
121
Brightness
To adjust the brightness values: In the Dark end and Light end boxes, type or select the values to determine the lower and upper limits of the range.
To return to the default values: Click Reset at the bottom of the Brightness section. This restores the default values of 0 and 255. After the color balancing step is completed and you are satisfied with the results in the Mosaic Preview window, proceed to the next mosaicking step, adding the image to the mosaic.
122
PCI Geomatics
Reapply Mosaicking
After the mosaic is complete, you may discover some areas that you want to change. You can use Manual Mosaicking to edit the cutlines or adjust the color balancing for the images as required, and then use Reapply Mosaicking to reassemble the mosaic file. You can use Reapply Mosaicking to: Preview the mosaic. Reproduce missing mosaic files. Regenerate the mosaic at different resolutions. Create a subset of the mosaic by changing the size of the Mosaic Area and then regenerating the mosaic.
To regenerate the mosaic: 1. On the Mosaic toolbar, click Reapply Mosaicking. The Reapply Manual Mosaicking dialog box opens.
Figure 6.6
Reapply Mosaicking dialog box
PCI Geomatics
123
2. Click in the Use column to select the images to be mosaicked. 3. Select from the following options: Current resolution - to create the mosaic using the resolution set in the Set Projection window Different resolution - to set a new resolution for the mosaic Regenerate stale orthos - to update any images labeled stale in the Status column Delete newly generated orthos after use - to save disk space Clear mosaic file before starting to mosaic - to delete images in the mosaic file and reset the background to the default before adding the selected images
4. Click Generate Mosaic. A Reapply Manual Mosaicking Progress monitor appears, which details the progress of the mosaic operation. 5. Click Close.
124
PCI Geomatics
Lesson 6.3
Automatic Mosaicking
Mosaicking is the second step in the Data Correction stage. This lesson describes how to set up the orthophotos, generate the mosaic automatically, and view the result. To create the mosaic from the orthophotos, you require: The project file from Lesson 6.1 An new area defined for mosaicking
PCI Geomatics
125
Now that you have defined a new mosaic area, you will create the mosaic using the automatic mosaicking tools. Afterwards, you will compare the manually generated mosaic with the automatically generated mosaic. To open the Automatic Mosaicking dialog box: On the Mosaic toolbar, click Automatic mosaicking. The Automatic Mosaicking dialog box opens.
Figure 6.7
Automatic Mosaicking dialog box
126
PCI Geomatics
Each row in the table represents an image in your project and lists the Image ID, Ortho ID, Status, whether the image is to be used what type of normalization is to be applied. By default, all images will be used in the mosaic. To set up the orthos for mosaicking: 1. Click in the Use column to ensure all images will be used. Only the images with check marks in the Use column will be mosaicked. 2. Under the Normalization column, select None. Normalization Normalization is used to even out the brightness in the images to achieve a more pleasing mosaic. You can set this feature differently for each image by clicking the corresponding arrow beside the Normalization column or you can set it for all the images by selecting the feature in the Normalization list and clicking Apply to All.
None: To leave the images as is. Hot Spot: To remove hot spots from the image. A hot spot is a common distortion
that results from solar reflections. Hot Spot normalizes the brightness over the image, but it does not remove spot reflections from lakes, cars, and buildings.
Across Image 1st Order: To correct the gradual change in brightness from one side of the image to the other. Recommended for ScanSAR and other imagery. Across Image 2nd Order: To correct the gradual change from dark to bright to
dark or vice versa across the image, also known as an antenna pattern. Recommended for ScanSAR and other imagery.
Across Image 3rd Order: To correct the gradual bright and dark patterns from
one side of the image to the other. Recommended for ScanSAR and other imagery.
Mosaicking Options
Regenerate offline orthos You can select this option to regenerate orthorectified images with a Stale or Offline status. OrthoEngine can orthorectify the images and mosaic them in one step. If you have already generated your orthos, as outlined in the previous module, you do not need to select this option. To regenerate offline orthos: Clear mosaic file before mosaicking Click Regenerate offline orthos.
This options deletes existing data from the mosaic file. If you are already starting with a blank mosaic file or are adding images to a large existing mosaic, you can disable this option to leave the mosaic file as is, which can save time. Although regenerating a large mosaic just to add a few images may be time consuming, it may produce better color balancing results.
127
PCI Geomatics
To clear this option: Click to clear the Clear mosaic file before mosaicking. Since your mosaic file has not yet been created, there is no need to delete any existing data from the mosaic file. Starting image The Starting image list lets you select the corrected image to be the basis for the mosaic, the color balancing, and the cutline selection. To select the Starting image: Color balance In the Starting image list, select an image.
Automatic color balancing applies tonal and contrast adjustments over the mosaic. There are three methods:
None: No color balancing is applied. Entire image: The histogram of each entire image is used to compute the color
balancing histogram. This method is recommended for images with low overlap or for images with systematic effects such as when images are bright at the top and dark at the bottom.
Overlap area: This method computes the color balancing histogram using only the pixels in the overlapping area of the images being added to the mosaic file. This method is recommended for most images. Match to ref image: This method matches the color balancing for the mosaic to
the image identified in the Mosaic reference image box. To apply color balancing during mosaicking: Ignore pixels under bitmap mask Cutlines In the Color balance list, select a method.
Select Ignore pixels under bitmap mask to disregard the pixel values under the mask when calculating the color balancing histogram. OrthoEngine uses the last bitmap segment in each image file as the mask. Cutlines are drawn in areas where the seams are the least visible based on the radiometric values of the overlapping images. There are four methods:
Min difference: To place the cutline in areas where there is the least amount of difference in gray values between the images. Min relative difference: To place the cutline in areas where there is the least
amount of difference in gradient values between the images.
128
PCI Geomatics
Use Entire image: To mosaic images that do not overlap. OrthoEngine uses the four corner coordinates of the images as the cutlines to avoid gaps between the images.
If you want to use existing or imported cutlines, select Use existing cutlines. To select a cutline method: Blend width From the Cutline Selection Method list, select Min Difference.
This is the number of pixels on each side of the cutline to delimit the area for smoothing the radiometric differences between the images. A Blend width of 5 will create a total blending width of 10 pixels - 5 pixels on either side of the cutline.
File Options
Preview file The Preview file box specifies the path and file name for the file that will contain a low resolution version of the full mosaic. You can click Browse to change the default file name and select a location. This box specifies the path for the temporary working files. You can click Browse to change the default location. The temporary files are deleted when the mosaic is complete. This option lets you select a mask area for your entire mosaic to exclude certain pixel values when calculating the color balancing histograms. This is the image you want to use as a reference for color balancing.
To set the Generation Start Time: Under Generation Start Time, click Start now or Start at (hh:mm) and set the time when you want the operation to begin (within the next 24 hours).
PCI Geomatics
129
To generate the mosaic: Click Generate Mosaic. This creates the full resolution version of the mosaic. It is saved in the auto_mosaic.pix file you created.
Tip
This is a good time to save your project file.
130
PCI Geomatics
PCI Geomatics
131
132
PCI Geomatics
OrthoEngine Componentization
Module 7 has Three Lessons:
Lesson 7.1 Lesson 7.2 Lesson 7.3
Module
Data Import and Automatic GCP Collection Project Creation and Tie Point Collection Automatic GCP Collection and Mosaicking
OrthoEngine Components
Many functions which were previously available only in the Focus and OrthoEngine GUI environments have been componentized. Componentization means that these functions can now be linked together into automated workflows and run in batch processes. Automation and batch processing of workflows is available through both visual modeling and command-line scripting. Component tasks include: Project creation Data import Ground control and tie point collection Radiometric adjustments Orthorectification
Modeler
Geomatica Modeler provides an interactive methodology for the development of both simple and complex data processing flows. Modeler provides access to a number of standard operations such as data import and export, as well as most EASI/PACE processing algorithms. You build processing models by placing modules on the Modeler canvas and then connecting the modules with pipes to create a process flow. You first configure the modules and then execute the model in either single execution mode or batch mode. During the execution of the model, graphical cues indicate the data flow through the process. The Module Librarian enables quick access to all modules.
PCI Geomatics
133
Lesson 7.1
A common operation in desktop photogrammetry is the registration of new images to existing geocoded images. Often this entails registration from a previous years data, or the updating of a database with new overlapping imagery. Traditionally, this registration was done through the manual or semi-automated collection of ground control points. Geomatica 10 introduces automated ground control collection through automatic image-to-image registration. Combined with OrthoEngines accurate satellite and airphoto models this technology enables fast, automated, rigorous orthorectification.
The model for this lesson uses a total of seven modules from the Module Librarian. First, you will add a CDLAND7 module to the canvas to import a Landsat-7 scene in HDF to pix format. The IMPORT modules will import a geocoded image and a digital elevation module. AUTOGCP will be used to collect GCPs from the geocoded image. These GCPs will be refined with GCPREFN. SATMODEL applies the satellite model after which the image can be orthorectified and exported with ORTHO and EXPORT, respectively. To start Modeler:
Figure 7.1
Geomatica toolbar Modeler command
The Module Librarian and the Modeler workspace open on your desktop.
134
PCI Geomatics
Canvas
Status Area
Figure 7.3
Module Librarian
The Module Librarian provides access to the modules you can use to process your data. Modules are the basic building blocks for your model. You access modules from the Algorithm Library tree view in the Module Librarian. Modules are sorted into categories and subcategories according to their functionality and can also be listed alphabetically.
Note
Modules for which you are not licensed are identified with a lock icon.
PCI Geomatics
135
To place a module on the canvas: 1. In the Module Librarian, expand the All Algorithms folder. 2. Select the CDLAND7 module. The CDLAND7 graphical element displays in the Selected Module window. 3. Click the CDLAND7 graphical element in the Selected Module window. 4. Click anywhere on the canvas. The CDLAND7 graphical element displays on the canvas.
Tip
You can also click Add to Canvas to place the selected module in the canvas.
5. Add the following modules to the Modeler canvas: 2 IMPORT modules 1 SPLIT module 1 AUTOGCP module 1 GCPREFN module 1 SATMODEL module 1 ORTHO module 1 EXPORT module
Now that all the modules have been added to the model, you will configure the IMPORT modules.
136
PCI Geomatics
To configure the first IMPORT module: 1. Double-click the IMPORT module located below the CDLAND7 module. The IMPORT Module Control Panel opens. 2. On the Input Params 1 tab, click Browse and locate the LANDSAT folder. 3. Select geo_landsat.pix and click Open. The Available Layers from this file are listed. 4. From the Available Layers list, select 3 [8U] red band.
Note
If using multispectral data for automatic image to image registration, you may choose to avoid using the blue band as it tends to be noisier than the other bands.
Figure 7.5
IMPORT Module Control Panel
5. Click Accept. The output raster port glyph displays on the IMPORT module and the status indicator bar turns green. To configure the second IMPORT module: 1. Double-click the second IMPORT module. The IMPORT Module Control Panel opens. 2. On the Input Params 1 tab, click Browse and locate the LANDSAT folder.
PCI Geomatics 137
3. Select dem.pix and click Open. The Available Layers from this file are listed. 4. From the Available Layers list, select the DEM from raster files layer. 5. Click Accept. To configure the CDLAND7 module: 1. Double-click the CDLAND7 module. The CDLAND7 Module Control Panel opens. 2. On the Input Params 1 tab, click Browse. The File Selector dialog box open. 3. From the LANDSAT\raw folder, select L71048026_02620000923_HDF.L1G file and click Open. 4. In the CD Input Layer(s) List box, type 1, 2, 3, 4, 5. 5. In the File Description box, type Southern Vancouver Island.
Figure 7.6
CDLAND7 Module Control Panel
6. Click Accept. The status indicator bar turns green. You will now connect the modules with pipes. Pipes Pipes are graphical elements that represent data transmission paths between modules. A pipe can be thin or fat. A thin pipe contains only one layer of information. A fat pipe, which is wider than a thin pipe, contains multiple layers of information. Modules are connected by clicking the output port of the module to connect from and then clicking the input port of the module to connect to. You can also connect between a pipe and a module by first clicking the pipe, and then clicking the input port on the connecting module.
138 PCI Geomatics
A pipe is default color-coded according to the type of data that it transmits. Some examples are listed below.
Table 1: Pipe Colors and Data Types Green White Red Blue Yellow Cyan Black Rasters Vectors Bitmaps Pseudocolor tables Lookup tables Binary Dead pipe
Tip
The most frequent causes for dead pipes are moving, renaming, and deleting an input file or layer.
Tip
Always connect modules before you configure them, because for some modules the default settings of an input layer can override a modules configuration. An exception is the IMPORT module, which you must configure for it to display a port.
To connect CDLAND7 and AUTOGCP: 1. Click the raster output port on CDLAND7 and then click the input port on the SPLIT module. A green pipe connects these two modules. 2. Click the second output port on the SPLIT module and then click the input raster port on SATMODEL. This is a temporary connection that allows you to access the third channel from the raw Landsat image. 3. Click the third output port on the SPLIT module and connect to the Input Image Layer port on AUTOGCP. 4. Click the pipe containing channel 2 of the raw Landsat image. The pipe displays with a blue border. 5. Press the DELETE key. 6. Click the Satellite Orbital Ephemeris Layer port on the CDLAND7 module and connect this to the Orbit Layer port on the AUTOGCP module. A blue pipe connects these two modules.
PCI Geomatics
139
Note
A SPLIT module will split a 'Fat' pipe into a 'Thin' pipe. Since a 'fat' pipe contains many layers, while a 'thin' pipe contains only 1 layer, this module will be transferring 1 layer from a 'fat' pipe and placing it into its own 'thin' pipe. This is useful if only a particular layer needs to pass on to another module.
To connect the IMPORT modules to AUTOGCP: 1. Click the raster port on the IMPORT module below the CDLAND7 module and connect this to the Reference Image Layer raster port on the AUTOGCP module. 2. Click the raster port of the lower IMPORT module and connect this to the Elevation Layer port on AUTOGCP.
Figure 7.7
CDLAND7 and IMPORT pipe connections to AUTOGCP
To connect AUTOGCP and GCPREFN: 1. Connect the GCP Layer port on AUTOGCP to the GCP Layer to be Refined port on GCPREFN. A cyan pipe connects these modules. 2. Connect the orbit port on CDLAND7 to the orbit port on GCPREFN. To set up the connections for SATMODEL: 1. Connect the GCP port on GCPREFN to the GCP port on SATMODEL. 2. Connect the output raster port on CDLAND7 to the raster input port on SATMODEL. 3. Connect the output orbit layer port on CDLAND7 to the orbit layer port on SATMODEL.
140
PCI Geomatics
To set up the connections for SATMODEL and ORTHO: 1. Connect the output raster port on CDLAND7 to the Image Layers to be Processed raster port on ORTHO. 2. Connect the output raster port from the second IMPORT to the Elevation Layer raster port on ORTHO. 3. Connect the Output Math Model Layer port on SATMODEL to the Math Model Layer port on ORTHO. 4. Connect the raster port on ORTHO to the Any port on EXPORT.
Figure 7.9
Model to orthorectify a Landsat-7 image
PCI Geomatics
141
To configure the AUTOGCP module: 1. Double-click the AUTOGCP module. 2. For the Sampling Count, enter a value of 50. This is the number of GCPs to be matched. 3. For the No Data Image Value, enter 0. The defaults for the remaining parameters will be used. 4. Click Accept. To configure the GCPREFN: 1. Double-click the GCPREFN module. 2. For the Rejection Method, enter 5, 1, 1. Mode 5 rejection (RMS error rejection) will be used. GCPs with an X or Y RMS higher than 1 will be rejected. 3. Click Accept. For this model, there are no parameters to configure for the SATMODEL module. The Output Projection for the ortho images will be set in the ORTHO MCP. To configure the ORTHO module: 1. Double-click the ORTHO module. 2. On the Input Parms 1 tab, enter UTM 10 U E012 for the Output Projection. 3. For the Output Pixel Ground Size: X, Y, enter 30, 30. 4. Set the Resample Mode to Cubic. 5. Click Accept. To configure the EXPORT module: 1. Double-click the EXPORT module. 2. On the Input Params 1 tab, click Browse and locate the LANDSAT folder. 3. For the File name, enter ortho.pix and click Save. 4. Select the Overwrite Existing File option. If the model is run more than once, ortho.pix will be overwritten. 5. Click Accept.
Tip
The COMMENT module allows you to enter comments relevant to the particular model.
142
PCI Geomatics
To add comments: 1. Add a COMMENT module to the canvas and place it above the CDLAND7 module. 2. Double-click the COMMENT module. 3. On the Input Params 1 tab, enter Read Landsat Imagery from HDF file. 4. Click Accept. 5. Resize the comment box so all the text is visible.
Figure 7.10
Model with comment
6. Add any other comments you would like to the model. You are now ready to execute your model. To execute a model: From the Execute menu, select Run or click the Run button on the toolbar. The status indicator bars on the modules show the progress of each operation as it is executed. Another progress indicator in the display area of the Modeler window monitors the progress of the entire model. It will take a few minutes for the model to run to completion. When the execution of the model has completed, double-click the AUTOGCP and GCPREFN modules and examine the reports in the LOG tab. To see how well the ortho image is registered to geo_landsat.pix, open the files in Focus.
PCI Geomatics
143
Saving a Model
Now that you have created your first model with Modeler, you will save the model as a MOD file. To save your model: 1. From the File menu, select Save Model. The File Selector dialog box opens. 2. Navigate to the LANDSAT folder. 3. In the File Name box, type ortho_landsat.mod. 4. Click Save. The File Selector dialog box closes and your model is saved as a MOD file. 5. From the File menu, click Close Model.
Note
When you save a model after it has executed, the intermediate and output files are not saved with the model.
144
PCI Geomatics
Lesson 7.2
OrthoEngine projects can now be created and saved in Modeler. This allows you to set up a work flow in Modeler to perform some of the image processing of your data. You can then continue working in OrthoEngine with the exported project file.
The model for this lesson uses a total of seven modules from the Module Librarian. First, you will batch import two ASTER images. An OrthoEngine project file will be created for the data and tie points will be automatically collected and refined. The project file will be saved once the tie points have been collected and could later be used in OrthoEngine for further data extraction. CPMMSEG will apply the sensor model and the images will be orthorectified with ORTHO. To start setting up the model: 1. Add the following modules to the Modeler canvas: 2 IMPORT modules 1 ACCUMULATE module 1 CRPROJ module 1 AUTOTIE module 1 TPREFN module 1 CPMMSEG 2 RELEASE modules 1 ORTHO module 2 EXPORT modules
PCI Geomatics
145
To configure the first IMPORT module: 1. Double-click the first IMPORT module. The IMPORT Module Control Panel opens. 2. Click Browse and navigate to the ASTER_MODELER folder. 3. Select 3b.pix and click Open. 4. From the Available Layers list, select the raster layer. 5. Click Batch. The Module Control Panel expands. 6. Click the + button. 7. Select the second batch parameter set in the list. 8. Browse and select the file 3n.pix.
Figure 7.12
IMPORT Module Control Panel with Batch parameter sets
9. Click Accept.
146
PCI Geomatics
To configure the second IMPORT module: 1. Double-click the IMPORT module above the ORTHO module. 2. Click Browse and navigate to the ASTER_MODELER folder. 3. Select aster_dem.pix and click Open. 4. From the Available Layers list, select the 32R SRTM DEM layer.
Figure 7.13
IMPORT Module Control Panel
5. Click Accept.
Note
The pseudo GCPs delivered with ASTER data cannot be imported into the model because the 3N and 3B images have different GCPs with the same GCP ID.
PCI Geomatics
147
To connect the IMPORT and CRPROJ modules: 1. Click the raster port on the IMPORT module and connect with the input port of the upper ACCUMULATE module. 2. Connect the output port on the ACCUMULATE module to the raster port on the CRPROJ module.
Note
The ACCUMULATE module accumulates all incoming layers during batch execution. These layers are released when all batch runs have been exhausted and all modules executed.
Figure 7.14
IMPORT module connected to CRPROJ
To connect CRPROJ to CPMMSEG and EXPORT: 1. Connect the Output OrthoEngine Project port on CRPROJ to the Input OrthoEngine Project port on AUTOTIE. 2. Connect the Output OrthoEngine Project port on AUTOTIE to the Input OrthoEngine Project port on TPREFN. 3. Connect the Output OrthoEngine Project port on TPREFN to the Input OrthoEngine Project port on the first EXPORT module. 4. Connect the Output OrthoEngine Project port on TPREFN to the Input OrthoEngine Project port on CPMMSEG. 5. Connect the Output raster port from the ACCUMULATE module to the Input raster port on the first RELEASE module. 6. Connect the Output raster port from the first RELEASE module to the Input raster port on the CPMMSEG module.
Note
The RELEASE module releases all incoming layers in sequence by layer or by group during batch execution. Release takes place during each model execution until all groups or layers have been released.
148
PCI Geomatics
To connect ORTHO and EXPORT: 1. Click the output raster port from the ACCUMULATE module and click the input port on the second RELEASE module. 2. On the second RELEASE module click the output port and connect it to the input Image Layers to be Processed raster port on the ORTHO module. 3. Connect the output Math Model Layer port on CPMMSEG to the input Math Model port on the ORTHO module. 4. Connect the output Raster port of the second IMPORT module to the input Elevation Layer port of the ORTHO module. 5. Connect the output Raster port on ORTHO to the input port on the EXPORT module.
Figure 7.16
Model showing all pipe connections
Now that all of the modules are connected, you will now configure them. To configure the CRPROJ module: 1. Double-click the CRPROJ module. This module creates an OrthoEngine Project file. 2. For the Output Projection, enter UTM 34 T E012. 3. From the Model Type list, select SAT (Toutins Model). 4. Click Accept.
PCI Geomatics 149
To configure the AUTOTIE module: 1. Double-click the AUTOTIE module. This module performs automatic tie point collection from a pair of overlapping images. 2. On the Input Params 1 tab, set Number of Points per Area to 60. 3. For Distribution, select Overlap. 4. For the Approximate Elevation, enter 1300. 5. For the Search Radius, enter 200 and click Accept. To configure the TPREFN module: 1. Double-click the TPREFN module. This module automatically refines tie points by eliminating those with large residual errors. 2. For the Rejection Criteria, enter 5, 10, 10. Mode 5 (RMS Error rejection) criteria will be used. Tie points with an X or Y RMS greater than 10 pixels will be rejected. 3. Click Accept. To configure the first EXPORT module: 1. Double-click the EXPORT module below the CPMMSEG module. 2. Click Browse and navigate to the ASTER_MODELER folder. 3. For the File name, enter aster_tp.prj and click Save. This will export the OrthoEngine project file.
Note
CPMMSEG computes and copies the math model contained in a project file to a Math Model layer in the input file. There are no parameters to configure for this module.
To configure the ORTHO module: 1. Double-click the ORTHO module: ORTHO orthorectifies a raw image based on the math model computed by CPMMSEG. 2. For the Output Projection, enter UTM 34 T E012. 3. For Output Pixel Ground Size: X, Y, enter 15, 15. 4. Click Accept. To set the batch parameters for the EXPORT module: 1. Double-click the EXPORT module to the right of the ORTHO module. 2. Click Batch.
150 PCI Geomatics
3. In the Batch parameter sets table, right-click the File column heading, and choose From Input Module. The From Input Module dialog box opens. If your model contains more than one IMPORT module, you would select from which module to select the input file names. 4. Click OK. The Batch parameter sets table updates with the names of the input files. 5. Press the SHIFT key, and select both parameter sets. 6. In the Batch parameter sets table, right-click the File column heading, and choose Add Prefix/Suffix. The Add Prefix/Suffix dialog box opens. 7. In the Add Prefix/Suffix dialog box, click the Prefix Text check box and enter ortho_. This option applies to both selected cells.
Figure 7.17
Add Prefix/Suffix dialog box
8. Click OK. The File names update in the Batch parameter sets table. 9. Click Accept. To execute the model in batch mode: From the Execute menu, select Run Batch or click the Run Batch button on the toolbar. It will take a few minutes for the model to run to completion. When the model has finished executing, double-click the Log tab of the TPREFN MCP. You will see a report of the deleted tie points, the original number of tie points as well as RMS errors. The output ortho images can be viewed in Focus. To save the model: 1. From the File menu, select Save Model. The File Selector dialog box opens. 2. Navigate to the ASTER_MODELER folder. 3. In the File Name box, type autotie_aster.mod and click Save. The File Selector dialog box closes and your model is saved as a MOD file.
PCI Geomatics 151
152
PCI Geomatics
Lesson 7.3
Entire work flows can be set up within OrthoEngine to perform automatic image-toimage registration, calculate the sensor model, orthorectify imagery and perform automatic mosaicking.
The model for this lesson uses a total of eight modules from the Module Librarian. First, you will batch import two SPOT scenes. Two other IMPORT modules will be used for a geocoded image and a DEM. GCPs will be automatically collected from the geocoded image with AUTOGCP and then refined with GCPREFN. The satellite model will be applied and the images will be orthorectified with ORTHO. The two orthorectified images will be automatically mosaicked with AUTOMOS. The individual ortho images and the mosaicked image will be exported. To start setting up the model: 1. Add the following modules to the Modeler canvas: 3 IMPORT modules 1 AUTOGCP module 1 GCPREFN module 1 SATMODEL 1 ORTHO module 1 ACCUMULATE module 1 AUTOMOS module 2 EXPORT modules
PCI Geomatics
153
To setup the Batch Parameter Sets for the first IMPORT module: 1. Double-click the upper IMPORT module. 2. Click Browse and navigate to the SPOT folder. 3. Select SPOTLEFT.PIX and click Open. 4. From the list of Available Layers, select the 8U raster layer and the orbital segment. 5. Click Batch. 6. Click the + sign. 7. Select the second Batch parameter set. 8. Browse to the SPOT folder and select SPOTRIGHT.PIX.
154
PCI Geomatics
9. Click Accept. To configure the second IMPORT module: 1. Double-click the middle IMPORT module. 2. Click Browse and navigate to the SPOT folder. 3. Select SPOT_MOSAIC.PIX and click Open. 4. Select the 8U raster layer. 5. Click Accept. To configure the third IMPORT module: 1. Double-click the lower IMPORT module. 2. Click Browse and navigate to the SPOT folder. 3. Select SPOTDEM.PIX and click Open. 4. Select the 16S raster layer. 5. Click Accept. Now that the IMPORT modules are configured, you will set up the pipes to connect the modules in the model.
PCI Geomatics
155
To connect the IMPORT modules to the AUTOGCP module: 1. Click the OutputRaster1 port on the upper IMPORT module, then click the Input Image Layer port on the AUTOGCP module. 2. Click the OutputORB1 port on the upper IMPORT module, then click the Orbit Layer input port on the AUTOGCP module. 3. Click the OutputRaster1 port on the middle IMPORT module, then click the Reference Image Layer port on the AUTOGCP module. 4. Click the OutputRaster1 port on the lower IMPORT module, then click the Elevation Layer input port on the AUTOGCP module.
Figure 7.20
Connections to AUTOGCP
To connect GCPREFN and SATMODEL: 1. Connect the GCP ports on the AUTOGCP and GCPREFN modules. 2. Connect the Orbital ports on the upper IMPORT module to the GCPREFN module. 3. Connect the GCP ports on the GCPREFN and SATMODEL modules. 4. Click the OutputRaster1 port on the upper IMPORT module, then click the Image Layers to be Processed port on the SATMODEL module. 5. Click the OutputORB1 port on the upper IMPORT module, then click the Orbit Layer port on the SATMODEL module.
156
PCI Geomatics
To connect ORTHO and ACCUMULATE: 1. Connect the raster layer port on the upper IMPORT module to the Image Layers to be Processed port on ORTHO. 2. Connect the raster layer port on the lower IMPORT module to the Elevation Layer port on ORTHO. 3. Connect the Math Model ports on SATMODEL and ORTHO. 4. Connect the Output Raster Layer port on ORTHO to the input port on the ACCUMULATE module. To connect EXPORT and AUTOMOS: 1. Connect the output raster port from ORTHO to the first EXPORT module. 2. Connect the output port on the ACCUMULATE module to the Image Layers to be Processed port on AUTOMOS. 3. Connect the Output Mosaic Layers port on the AUTOMOS module to the second EXPORT module. The AUTOGCP module automatically registers a raw image to a geocoded image and creates a GCP segment stored in the raw image file. You can open the AUTOGCP Module Control Panel to see the parameters, however in this lesson, you will use the default parameters.
Tip
You can rotate the ports on a module by clicking the Rotate button on the Modeler toolbar. This is useful to help organize your pipes.
PCI Geomatics
157
To configure the GCPREFN module: 1. Double-click the GCPREFN module. This module refines GCPs in a GCP segment and removes points with large errors. 2. Ensure the Model Type SAT (Toutins Model) is selected. 3. For the Rejection Method, enter 4,1,1. Mode 4 (absolute distance rejection) will be used. GCPs with residual values greater than 1 pixel X and 1 pixel Y will be rejected. 4. Click Accept. To configure the ORTHO module: 1. Double-click the ORTHO module. ORTHO orthorectifies a raw image based on the math model computed by SATMODEL. 2. For the Output Projection, enter UTM 11 E000. 3. For the Output Pixel Ground Size: X, Y, enter 10, 10. The remaining default parameters will used. 4. Click Accept. To configure the AUTOMOS module: 1. Double-click the AUTOMOS module. This module performs automatic mosaicking of a set of geocoded images. 2. For the Color Balancing Method, select OVERLAP. 3. For the Cutline Generation Method, select EDGE. 4. Click Accept. To set the batch parameters for the first EXPORT module: 1. Double-click the EXPORT module to the right of the ORTHO module. 2. Click Batch. 3. In the Batch parameter sets table, right-click the File column heading, and choose From Input Module. The From Input Module dialog box opens. If your model contains more than one IMPORT module, you would select from which module to select the input file names. 4. Click OK. The Batch parameter sets table updates with the names of the input files. 5. Press the SHIFT key and select both parameter sets. 6. In the Batch parameter sets table, right-click the File column heading, and choose Add Prefix/Suffix. The Add Prefix/Suffix dialog box opens.
158 PCI Geomatics
7. In the Add Prefix/Suffix dialog box, click the Prefix Text check box and enter ortho_. This option applies to both selected cells. 8. Click OK. The File names update in the Batch parameter sets table. 9. Click Accept. To configure the second EXPORT module. 1. Double-click the EXPORT module below the AUTOMOS module. 2. Click Browse and navigate to the SPOT folder. 3. Enter the Filename auto_mosaic.pix and click Save. 4. Click Accept. The AUTOGCP and SATMODEL modules do not need to be configured as default parameters will be used.
Figure 7.22
SPOT model
You are now ready to execute your model. To execute the model: From the Execute menu, select Run Batch or click the Run Batch button on the toolbar. It will take a few minutes for the model to run to completion. To save the model: 1. From the File menu, select Save Model. The File Selector dialog box opens.
PCI Geomatics
159
2. Navigate to the SPOT folder. 3. In the File Name box, type spot_autogcp.mod and click Save. The File Selector dialog box closes and your model is saved as a MOD file.
160
PCI Geomatics
Appendix
The following table lists the minimum number of ground control points (GCPs) to collect, but it is recommend that you collect more than the minimum to ensure accuracy. However, collecting over 20 GCPs per image does not significantly improve the accuracy for most math models. To improve the accuracy, collect GCPs evenly throughout the image at a variety of elevations and in areas where images overlap. Also, the quality of the GCPs impacts the number needed to ensure accuracy.
Table 1: Minimum Number of GCPs Math Model / Data Aerial Photography Aerial Photography with GPS/INS Satellite Orbital: SPOT 1 to 4 ASTER, AVNIR-2, CARTOSAT, CBERS, DMC, EOC, IRS, LANDSAT, ORBVIEW, PRISM, QUICKBIRD Basic, SPOT 5 ASAR, EROS, ERS, FORMOSAT-2, IKONOS, JERS, PALSAR, QUICKBIRD Ortho Ready, RADARSAT ASAR/PALSAR/RADARSAT Specific Model Rational Functions: If Computed from GCPs 5 per image * 19 per image * For IKONOS Ortho Kit, improve accuracy with 1 or more GCPs (using zero-order RPC adjustment); for CARTOSAT and QUICKBIRD Ortho Ready, minimum of 3 GCPs (using first-order RPC adjustment); for ORBVIEW, minimum of 1 GCP (using zero-order RPC adjustment) Collecting more than the minimum will average out errors introduced by inaccurate GCPs or terrain variations 4 per image 6 to 10 per image Minimum GCPs 3 to 4 per project GCPs optional Recommended 3 per image for highest accuracy
6 per image
10 to 15 per image
8 per image
10 to 15 per image Improve accuracy with a minimum of 8 GCPs for RADARSAT only
GCPs optional
GCPs optional
3 per image
PCI Geomatics
161
Geomatica OrthoEngine - Appendix A Table 1: Minimum Number of GCPs Math Model / Data Polynomial: First-order Second-order Third-order Fourth-order Fifth-order 4 per image 7 per image 11 per image 16 per image 22 per image * Depending on the number of coefficients that you want to use. Collecting more than the minimum will average out errors introduced by inaccurate GCPs Minimum GCPs Recommended
162
PCI Geomatics