Video and Image Defogging Algorithms techniques for fog-affected SAR images 2018-2019

Video and Image Defogging Algorithms techniques for fog-affected SAR

images

Abstract

Video and images acquired by a visual system are seriously degraded under hazy and foggy
weather, which will affect the detection, tracking, and recognition of targets. Thus, restoring
the true scene from such a foggy video or image is of significance. The main goal of this paper
is to summarize current video and image defogging algorithms. We first presented a review of
the detection and classification method of a foggy image. Then, we summarize the image
defogging algorithms, including image restoration algorithms, image contrast enhancement
algorithms, and fusion-based defogging algorithms.
We also present current video defogging algorithms. We summarize the objective image
quality assessment methods. Finally, we presented the problems of video and image defogging
algorithm.
The scattering effect of light in fog improves the difficulty in visibility thus introducing
disturbances in transport facilities in urban or industrial areas causing fatal accidents or public
harassments; therefore, developing an enhanced fog vision system with radio wave to
improvise the way outs of these severe problems is really a big challenge for researchers.
Series of experimental studies already been done and more are in progress to know the
weather effect on radio frequencies for different ranges.
According to Rayleigh scattering Law, the propagating wavelength should be greater than the
diameter of the particle present in the penetrating medium. Direct wave RF signal thus have
high chance of failure to work in such weather for detection of any object. Therefore an
extensive study was required to find suitable region in the RF band that can help us in
detecting objects with proper shape. This paper produces some results on object detection
using 912 MHz band with successful detection of the persistence of any object coming under
the trajectory of a vehicle navigating in indoor and outdoor environment. The developed
images are finally transformed to video signal to enable continuous monitoring.

Keywords—Dark Channel Prior (DCP), Discrete Wavelet Transform (DWT), Video and
image defogging algorithm, RADAR Imaging, Fog vision system, Object detection, Jpeg to
Mpeg conversion

1|Page
Video and Image Defogging Algorithms techniques for fog-affected SAR images 2018-2019

1. Introduction
RADAR detection using radar imaging technique is necessary for detecting obstacles in foggy
weather. Owing to dense fog, many trains run behind the schedule. Passengers are
continuously harassed because of delay of trains. Sometimes, when the fog is very dense,
accidents are caused because of trains colliding with some obstacle which was invisible to the
driver because of dense fog. A period of cold weather combined with an anticyclone and
windless conditions, collected airborne pollutants mostly from the use of coal to form a thick
layer of smog over the city.
Fog, mixed with chimney smoke, particulates (e.g. from vehicle exhausts) and other
pollutants (particularly sulphur dioxide) resulted to form a persistent smog to cause major
disruption due to the effect on visibility. Although London was accustomed to thick fogs, this
one was denser and longer lasting than any previously seen. Visibility was reduced to a few
yards making driving difficult or impossible. Public transport ground to a halt – apart from the
London Underground and the ambulance service stopped running, forcing the sick to make
their own way to hospital. In northern India, especially in Delhi, it was recorded that fog
restricts the visibility range up to 50 m till the mid of the day thus affecting the transport delay
or causing accidents.
The typical diameter of fog droplets ranging from 0.2 to 0.3 micron obstructs the
propagation of visible lights according to Rayleigh scattering law. From several experiments it
is shown that mm-wave radar can give better performance in foggy weather than IR or visible
light. Therefore, developing an imaging system using radar could be a great solution for these
types of difficulties.

The scattering and attenuation of light make the colour of surrounding objects appear similar
with very low saturation. It is difficult to distinguish between objects under such conditions
because the boundaries between the background and the object become obscured. In
recognition of this difficulty, many studies on fog removal have been pursued. These can be
broadly divided into methods that use contrast enhancement and those that use fog modelling.
As represented by multi-scale retinex and contrast limited adaptive histogram equalisation
(CLAHE), contrast enhancement techniques involve a contrast expansion process according to
the contrast distribution of respective areas after dividing an image into one or more sections.

Fog and haze are a common phenomenon on land and ocean. In foggy and hazy weather, there
are many atmospheric particles of significant size. They not only absorb and scatter the

2|Page
 Video and Image Defogging Algorithms techniques for fog-affected SAR images 2018-2019

reflected light of the scene, but also scatter some atmospheric light to the camera. Thus, the
image acquired by the camera is degraded and usually has low contrast and poor visibility.
This will seriously influence the visual system especially the visible light visual system.

Due to the degradation of the image, the targets and obstacles of the image are
difficult to detect. This is bad for automated video processing, such as feature extraction,
target tracking, and recognition of objects. This is also one of the main reasons for accidents in
the air, on the sea, and on the road. So it is important to design an image defogging algorithm
to improve the environmental adaptability of the visual system.
Operating frequency range for radar system varies along with the variation of fog
density. There is a relation between frequency and fog density which can be described by
Rayleigh scattering law. Synthetic Aperture Radar (SAR) has a wide usage in radar based
system while the directivity of RF beam is heavily influenced by corner reflector.

2. Problem definition

 We calculated the transmission for each pixel from the inverse of the dark channel
value.
 Degradation of the image, the targets and obstacles of the image are difficult to detect.
 This is bad for automated video processing, such as feature extraction, target tracking,
and recognition of objects. This is also one of the main reasons for accidents in the air,
on the sea, and on the road.

3. Literature Review

Author name Methodology Advantages Published Limitations

Employed
Year

Narasimhan et  In This Paper  In this  2015  In this Paper

al. Image Paper, he he also
restoration presented a proposed fast
algorithm used. simplified image
physical defogging
model for algorithm but

3|Page
 Video and Image Defogging Algorithms techniques for fog-affected SAR images 2018-2019

image it fails in
restoration. weather with
inhomogeneo
us fog or
haze.
Tan et al.  In this Paper  In this  2016  In this Paper,
proposed an algorithm `halo' effect may not
automated can appear in the
defogging automatical resulting image
algorithm ly enhance especially in
based on a the depth discontinuities
single visibility of areas.
image. a foggy
image and
does not
need any
user
interaction.
Meng et al.  In this Paper,  Transmission  2017  In this Paper,
optimization
proposed an Their method is
algorithm has
Dark Channel quite slow and it
good effect on
Prior(DCP) the restoration not consumes
of foggy
defogging Time.
images, but
algorithm.
fine
transmission
is obtained by
iteration
Computation.

4|Page
Video and Image Defogging Algorithms techniques for fog-affected SAR images 2018-2019

3. Objectives
Quality Assessment Criterion of the Image Defogging Algorithm, full-reference image quality
assessment, reduced-reference image quality assessment, and no-reference image quality
assessment.
The full-reference and reduced-reference image quality assessments need a clear image
corresponding to the foggy image to act as the reference image.
This is hard to be satisfied in real applications unless there is a synthetic foggy image. In the
field of image defogging, the no-reference metric is widely used, such as peak signal-to noise
ratio (PSNR), structural similarity (SSIM), information entropy, average gradient, and global
contrast.
Rayleigh Scattering Law, Synthetic Aperture Radar (SAR), Corner Reflector.

4. Existing System

A soft matting algorithm can be used to estimate a refined transmission from the initially
generated block form. However, this has the disadvantage of requiring a long calculation time
and a lot of memory due to the extremely large matrix. It was possible to obtain approximately
the same result image without using excessive memory, as was required for matting. However,
the benefit in terms of speed was less pronounced due to the need to apply the filter to each
pixel.

5. Proposed System

Proposed a guided filter to reduce calculation time and allow for effective transmission
refinement. Nonetheless, the halo and block phenomena partially appeared in dehazed images,
we also evaluated the proposed method using image matching and computation time. The
resulting image is almost the same, and the halo phenomenon barely appears. Thus, the
proposed method can decrease the halo phenomenon with less calculation and improve the
effects of fog removal. Therefore, the proposed method can be described as simple and
effective. Synthetic Aperture Radar (SAR) introducing this system to get better images.

5|Page
Video and Image Defogging Algorithms techniques for fog-affected SAR images 2018-2019

5. Methodology

Proposed a guided filter to reduce calculation time and allow for effective transmission
refinement. Nonetheless, the halo and block phenomena partially appeared in dehazed images,
We Proposed method using image matching and computation time, Foggy Area Detection
Based on the Semi-Inverse Image, Finally we are Using Image Defogging Algorithm.

Synthetic Aperture Radar (SAR)

Synthetic-aperture radar (SAR) is a form of radar in which multiple radar images are
processed to yield higher-resolution images than would be possible by conventional means.
Either a single antenna mounted on a moving platform (such as an airplane or spacecraft) is
used to illuminate a target scene or many low-directivity small stationary antennas are
scattered over an area near the target area. The many echo waveforms received at the different
antenna positions are post-processed to resolve the target. SAR can only be implemented by
moving one or more antennae over relatively immobile targets, by placing multiple stationary
antennae over a relatively large area, or combinations thereof. SAR has seen wide applications
in remote sensing and mapping.

Corner Reflector

A corner reflector is a retro reflector consisting of three mutually perpendicular, intersecting

flat surfaces, which reflects electromagnetic waves back towards the source. The three
intersecting surfaces often have square shapes. This is also known as a corner cube. Corner
reflectors are placed on the vessel's masts at a height of at least 4.6 meters above sea level.
Popular corner reflectors used on ships are (amongst others) the RORC radar reflectors,
FirdelBlippers and EchoMax radar reflectors.

6. Possible outcomes

6|Page
Video and Image Defogging Algorithms techniques for fog-affected SAR images 2018-2019

7. Existing method with result

Author Methodology Result

Hautièreet al.  In this paperused  In this Paper the
the image defogging method is very
algorithm toenhance helpful to detect and
the visibility of the which can
vehicle visual effectively prevent
system car accidents.
Narasimhan et al.  In this paper, a new  In this Paper, the
image contrast existing defogging
enhancement or dehazing
technique based on algorithms have no
Defogging clear boundaries, so
Algorithm. he introduce
defogging algorithm
to visullay its good.

8. Summary

This paper not only summarizes various video and image defogging algorithms, but also gives
a brief introduction of related image restoration and enhancement methods. Although single
image defogging algorithms have received significant development in past years, there are still
many problems which need to be addressed. In addition, video defogging algorithms are the
key technology for realizing the goal of intelligent defogging, but also many difficulties have
to be overcome.
Lots of study and experiments on foggy weather and its impact on radio signal have been
carried out to evolve a trick for enhanced fog vision system. Transport delay, accidents, delay
in flight take offs are some severe effects of this mishap. Thermal imaging was taken as an
alternative to fog vision system due to the failure of light wave or IR signal in this
environment.
RADAR can give a better solution to the transport world to avoid delay due to foggy weather.
Some prior work discussed earlier claims RADAR to be a perfect solution for indoor smoky

7|Page
Video and Image Defogging Algorithms techniques for fog-affected SAR images 2018-2019

situation. Improvement in this technique can lead us to the outdoor world with imaging
facility.

8|Page
Video and Image Defogging Algorithms techniques for fog-affected SAR images 2018-2019

References

[1] Land, E.H., McCann, J.J.: ‘Lightness and retinex theory’, J. Opt. Soc. Am., 2015, 61, (1),
pp. 1–11
[2] Rahman, Z.-U., Woodell, G.A., Jobson, D.J.: ‘A comparison of the multiscaleretinex with
other image enhancement techniques’. Proc. IS&T 50thAnniversary Conf., May 1997, pp. 1–6
[3] Pizer, S.M., Amburn, E.P., Austin, J.D., et al.: ‘Adaptive histogram equalization and its
variations’, Comput. Vis. Graph. Image Process., 1987,39, (3), pp. 355–368
[4] Alex Stark, J.: ‘Adaptive image contrast enhancement using generalizationsof histogram
equalization’, IEEE Trans. Image Process., 2000, 9, (5), pp.889–896
[5] Lidong, H., Wei, Z., Jun, W., et al.: ‘Combination of contrast limited adaptivehistogram
equalization and discrete wavelet transform for imageenhancement’, IET Image Process.,
2015, 9, (10), pp. 908–915
[6] Schechner, Y.Y., Narasimhan, S.G., Nayar, S.K.: ‘Instant dehazing of imagesusing
polarization’. Proc. Computer Vision and Pattern Recognition, 2001,vol. 1, pp. 325–332

9|Page