PRILIMINARY PROJECT REPORT

ON
Fog Degreded Image Analysis

PREPARED BY

Patel Jesis J 60
Patel Bhavesh. 61
Surati Gaurav 63
Suthar Mrunal 45

Department of Electronics and Communication Engineering

2010-11

Guided by Co- Guided by

Prof.Neeta H. Chapatwala Prof.Chirag N.Paunwal
Prof.Bhargav Shah

Sarvajanik College of Engineering & Technology

Dr. R.K. Desai Road, Athwalines,
Surat-395009

1
 Sarvajanik Education Society

Sarvajanik College of Engineering &

Technology, Surat

Department of
Electronics and communication Engineering,

CERTIFICATE
This is to certify that the Midterm Project report
entitled Fog Degraded Image Analysis.
is prepared & presented by
Patel Jesis J 60
Patel Bhavesh. 61
Surati Gaurav 63
Suthar Mrunal 45

of B.E. IV Sem VII Electronics & communication Engineering department

during year 2010-11. Their work is satisfactory.

Signature of guide Head of the Department

2
 Acknowledgement

When a man walks a mile, he gets more Knowledge than what he can get from hundred
books. Every project is a synthesis of knowledge and experience of students and experts in
their related fields. However the guidance and the help extended by the experts to the students
with the sole benevolent purpose of intellectual development generate a feeling of profound
respect for them.

It gives us immense pleasure to express our sincere gratitude to all those Concerned with our
project entitled “ Fog Degraded Image Analysis ”.

We are heartily thankful to Mrs. Neeta H Chapatwala (Guide) for helping us tremendously in
analyzing and designing our project. Without her guidance, we could not have completed it
satisfactorily. The valuable knowledge, that she has imparted us proved to be a key success in
overcoming challenges that we faced during the course of the project

Our deep and sincere gratitude to our internal guide Mr. Chirag N. Paunwala.(Co-Guide) and
Mr. Bhargav Shah(Co-Guide) whose suggestion and constructive critics have been invaluable
assets throughout the project. We express our thankfulness to Mr. Nitin B. Patel (H. O. D. EC
Dept, SCET) & Mr. Maulin M. Joshi(D.I.C, EC Dept, SCET) for showing faith in our abilities and
giving us their valuable information, support and guidance which has given us immense help
and encouragement in completing this project in given duration.

Last but not the least; we are thankful to our FAMILY & the ALMIGHTY GOD. Without their
blessings, we would not have completed our project.

3
 ABSTRACT
Images of outdoor scenes captured in bad weather suffer from poor contrast. Under bad weather
conditions, the light reaching a camera is severely scattered by the atmosphere. The resulting decay in
contrast varies across the scene and is exponential in the depths of scene points. Therefore, traditional
space invariant image processing techniques are not sufficient to remove weather effects from images.
In this paper, we present a physics-based model that describes the appearances of scenes in uniform bad
weather conditions. Changes in intensities of scene points under different weather conditions provide
simple constraints to detect depth discontinuities in the scene and also to compute scene structure. Then,
a fast algorithm to restore scene contrast is presented. In contrast to previous techniques, our weather
removal algorithm does not require any a priori scene structure, distributions of scene reflectances, or
detailed knowledge about the particular weather condition. The method described in this report is
effective under a wide range of weather conditions including haze, mist, fog, and conditions arising due
to other aerosols

In this report we presents an image enhancement algorithm for fog degraded images. The proposed
method does not require any environmental condition, in which image is captured, to be known or
multiple images, it requires only single input image for the enhancement of its visibility. Also the
proposed method is computationaly faster as it applies the contrast stretching only of the affected areas
of the image, instead of the whole image.

4
 Index

CHAPTER: 01 Introduction to image processing for fog degraded image ....................................... 7

1.1. Aim of the project- To improve the visibility of degraded image under bad
weather condition like fog, haze & mist. ................................................................................................ 8
1.2. Introduction. ........................................................................................................................... 8
1.2.1 Enhanacement using Atmospheric Model .......................................................................... 9
1.2.2. Contrast Enhancement ...................................................................................................... 10
CHAPTER: 02 Literature Survey ..................................................................................................... 12
Literature survey .................................................................................................................................. 13
2.1 An Improved fog degraded image enhancement algorithm [1] ................................................... 13
2.2 Image De-noising Based on Undecimated Discrete Wavelet Transform[2] ................................. 13
2.3 Enhancement of Image Degraded by Fog Using Cost Function Basedon Human
Visual Model[3] ................................................................................................................................... 14
2.4 Improving Image Quality in Poor Visibility Conditions Using a Physical Model for Contrast
Degradation[4] ..................................................................................................................................... 14
2.5 Contrast Restoration of Weather Degraded Images[5] ................................................................ 15
2.2 Existing Methodology with references ....................................................................................... 16
2.2.1 Contrast restoration using atmospheric model ......................................................................... 16
2.2.2 Contrast restoration of iso-depth regions ................................................................................. 16
2.2.3 Contrast enhancement using histogram equilization ................................................................ 17
2.2.4 Contrast restoration using scene structure................................................................................ 17
2.2.5 Ddepth edges from two weather conditions ............................................................................. 17
2.6 Enhancement of Image Using Cost Function Based on Human Visual Model ............................ 18
CHAPTER: 03 Design Specificcation\Observation .......................................................................... 19
3.1. Design Specification. ................................................................................................................ 20
3.2 Flow Chart. ................................................................................................................................ 21
3.4. Required Tools and Alternatives ............................................................................................... 23
3.5. Standard Stimulation Result. ..................................................................................................... 24
3.6 Work carried out in odd sem....................................................................................................... 26
3.7 Work targeted for even sem. ...................................................................................................... 26
CHAPTER: 04 ................................................................................................................................... 27
Conclusion ........................................................................................................................................... 28
Reference ....................................................................................................................................... 29
History Sheet ...................................................................................................................................... 30

5
 List of figures
Figure 1 Attenuation of light by scattering [1] ........................................................................................ 9
Figure 3 Fog- Degraded Image[1] ........................................................................................................ 10
Figure 2 Histogram Equalized Image of Figure 1[1] ............................................................................. 10
Figure 4 Enhancement of Figure 1 by the proposed[1] ......................................................................... 11
Figure 5 Scattering of light in atmosphere ............................................................................................ 16
Figure 6 Original Fog degraded Image ................................................................................................. 24
Figure 7 gray scale image of figure 3.1 ................................................................................................ 24
Figure 8 Histogram equalization on non edge image obtained. ............................................................. 25
Figure 9 Improved image ..................................................................................................................... 25
Figure 10 Comparison between gray scale image and the degraded image obtained. ............................ 26

6
CHAPTER: 01
INTRODUCTION TO IMAGE
PROCESSING FOR FOG DEGRADED
IMAGES

7
1.1. Aim of the project- To improve the visibility of degraded image under bad
weather condition like fog, haze & mist.

1.2. Introduction.
The Degradation of images by fog and mist is a familiar problem. In the literature on atmospheric
propagation, distributions of particles such as fog, mist, cloud, and haze are collectively known as
atmospheric aerosols. The effect of such aerosols is to progressively reduce image contrast with
increasing distance. This is due to the following two scattering processes: 1) light reflected from the
object surface is attenuated due to scattering by aerosol particles; and 2) some direct light flux is
scattered toward the camera. These effects result in a loss of contrast that is characteristic of poor
visibility conditions. The level of contrast reduction increases with the distance from the camera to the
object.

Images taken under bad weather conditions suffer from degradation and severe contrast loss. The degree
of degradation increases exponentially with the distance of scene points from the sensor. The standard
filtering methods cannot restore images degraded by bad weather conditions like fog, mist, haze etc.,
hence contrast enhancement methods are used. There can be two approaches for improving the visibility
of fog degraded images. One is based on the atmospheric model and the other is based on the contrast
enhancement.

Many algorithms are available for enhancing image contrast. Of these algorithms, perhaps the best
known is histogram equalization. However, these algorithms are designed for images that are stationary
in the sense that the image properties are roughly constant across the image. One of the main
characteristics of atmospheric aerosol degradation is that the local image contrast depends strongly on
distance. Histogram equalization and its variants may be applied locally but low spatial frequencies are
then lost. It seems reasonable to suppose that better results could be obtained by exploiting knowledge
of the characteristics of aerosol degradation.

A considerable amount of research has been directed toward greater understanding of image propagation
through the atmosphere, and this has resulted in the design of sophisticated forward imaging models.
However, most of the effort has been in predicting the performance of imaging systems, establishing
optimal imaging wavelengths for particular meteorological and geographical conditions , and in
estimating levels of aerosol pollution, and not in image processing. One exception to this is the work of
Tajbakhsh and Boyce who propose a model-based method of compensating for the attenuation of
terrain-reflected light due to scattering when imaging from an airborne camera. They suggest that a
range-dependent scaling should be applied to reverse the Mie scattering effect, and so maintain a
uniform terrain signal across the image. However, they offer no method for determining the appropriate
value of the extinction coefficient; this is assumed to be known in advance. More seriously, their image
processing procedure takes no account of the direct flux that is scattered toward the camera. This means
that their algorithm will inevitably fail in daylight viewing conditions when the flux due to scattered
light may be several orders of magnitude greater than the flux reflected from the terrain. Some of
enhancement method are mentioned below

8
1.2.1 Enhanacement using Atmospheric Model

These methods use physical models to predict the pattern of image degradation and then restore image
contrast with appropriate compensations. They provide better image rendition but usually require extra
information about the imaging system or the imaging environment. Oakley used a physics-based method
to restore scene contrast without any predicted weather information by approximating the distribution of
radiances in the scene by a single gaussian with known variance, however, in most of atmospheric based
models , scene depth need to be estimated beforehand, but it requires more information about the scene
environment, like multiple degraded images taken from the same point or both the clear day and foggy
day images. Narasimhan and Nayar use two or more different bad weather images taken from the same
point of view to restore scene structure and contrast based on atmospheric scattering model by assuming
the atmospheric scattering properties invariably as shown in figure 1. Narasimhan, presented an
interactive scene depth estimate method, in which the image contrast can be restored using a single
image when the biggest and smallest scene depth is assigned beforehand.
All methods based on physical model either need scene depth information to be known beforehand or
multiple degraded images taken from the same point, these requirements make this approach impractical
in some cases.[1].

Figure 1 Attenuation of light by scattering [1]

9
1.2.2. Contrast Enhancement

The most commonly used contrast enhancement method is histogram equalization and its variations.
Other image enhancement techniques, that uses histogram equalization are also presented, e.g. in order
to restore contrast of fog-degraded images, which do not need any information regarding the scene depth
and avoids complicated atmospheric scattering model.

Histogram equalization is one of the popular contrast enhancement algorithm, due to its simplicity and
effectiveness. However, since it uses histogram information over the whole image, as its transformation
function, in order to stretch contrast, so it may not reflect scene depth change in different parts of the
image.

Thus, the enhancement effect may not be satisfying when depth changes in the scene. As an illustration,
Figure 1.2 below shows the result of histogram equlization when applied on Figure 1.1 which shows that
visibility after histrogram equilization may not improve or there can be added noise which degrades the
visibilty. Thus, defining which areas of the image have to be enhanced becomes important.

Figure 3 Fog- Degraded Image[1] Figure 2 Histogram Equalized Image of Figure

1[1]

10
Figure 1.3 shows the results of the proposed method, that has significant improvement in the visiblity as
compared to Figure 2.In this paper, we present an image enhancement algorithm which finds out which
areas of the image are affected by poor contrast, and then enhances the contrast of only those regions.[1]

Figure 4 Enhancement of Figure 1 by the proposed[1]

11
CHAPTER: 02
LITERATURE SURVEY

12
Literature survey
2.1 An Improved fog degraded image enhancement algorithm [1]
In fog weather, Images captured by outdoor surveillance system degrade significantly and suffer
from poor contrast. This paper presents an improved algorithm for enhancing fog-degraded image
contrast in which a moving mask is used. By assuming the pixels in a mask having same scene
depth, the algorithm applies the modified partially overlapped sub-block histogram equalization to
implement contrast enhancement in every mask.

It involves two steps: first, sky region is segmented in order to restrain over-enhancement in the flat
region and reduce noise; then, remove the sky pixels in masks and modify the histogram
information in mask, thus the modified partially overlapped histogram equalization transformation
function for enhancement can be gotten.

This paper also proposes anovel fuzzy edge detection algorithm to evaluate above contrast
enhancement effect. Experiments on many fog-degraded images demonstrate that the proposed
improved algorithm is simple and effective in contrast enhancement of fog-degraded images.

2.2 Image De-noising Based on Undecimated Discrete Wavelet Transform[2]

A new image denoising approach based on undecimated discrete wavelet transform (UDWT) is
proposed. The proposed method combines a technique of cone of influence (COI) analyzing and
UDWT. Therefore it can effectively remove the impulse noise and preserve theimage edges.
Furthermore, combining with the traditional wavelet thresholding denoising method, it can be used to
welldecrease more widely type of the noise such as Gaussiannoise, poisson noise and even mixed noise.
Simulation resultsshow that the filtering performance of the proposed approach is very satisfactory.

This method is different from the tradition one based on the Fourier transformation signal analysis,
which separates the signal and the noise in the frequency domain, and then filters the noise with the
linearity invariable-time filter. The basic thought of the wavelet de noising method by thresholding is:
To make the signal scope and the noise scope is as far as possible different,although the signal and the
noise overlap is in the same place in the wavelet domain, so long as their scope is different, we can use
the threshold filter method to remove the influence of noise.

13
2.3 Enhancement of Image Degraded by Fog Using Cost Function Basedon
Human Visual Model[3]

In foggy weather conditions, images become degraded due to the presence of airlight that is generated
by scattering light by fog particles. In this paper, we propose an effective method to correct the degraded
image by subtracting the estimated airlight map from the degraded image. The airlight map is generated
using multiple linear regression, which models the relationship between regional airlight and the
coordinates of the image pixels. Airlight can then be estimated using a cost function that is based on the
human visual model, wherein a human is more insensitive to variations of the luminance in bright
regions than in dark regions. For this objective, the luminance image is employed for airlight estimation.
The luminance image is generated by an appropriate fusion of the R, G, and B components.
Representative experiments on real foggy images confirm significant enhancement in image quality over
the degraded image.

In this paper, we improve the Oakley method to make it applicable even when the airlight distribution
is not uniform over the image. In order to estimate the airlight, a cost function that is based on the
human visual model is used in the luminance image. The luminance image can be estimated by an
appropriate fusion of the R, G, and B components. Also, the airlight map is estimated using least squares
fitting, which models the relationship between regional airlight and the coordinates of the image pixels.

2.4 Improving Image Quality in Poor Visibility Conditions Using a Physical Model for
Contrast Degradation[4]

In daylight viewing conditions, image contrast is ften significantly degraded by atmospheric aerosols
such as haze and fog. This paper introduces a method for reducing this degradation in situations in
which the scene geometry is known. Contrast is lost because light is scattered toward the sensor by the
aerosol particles and because the light reflectedby the terrain is attenuated by the aerosol. This
degradation is approximately characterized by a simple, physically based model with three parameters.

The method involves two steps: first, aninverse problem is solved in order to recover the three model
parameters; then, for each pixel, the relative contributions of scattered and reflected flux are estimated.
The estimated scattercontribution is simply subtracted from the pixel value and the remainder is scaled
to compensate for aerosol attenuation. This paper describes the image processing algorithm and
presentsan analysis of the signal-to-noise ratio (SNR) in the resulting enhanced image.

This analysis shows that the SNR decreases exponentially with range. A temporal filter structure is
proposed to solve this problem. Results are presented for two image sequences taken from an airborne
camera in hazy conditions and one sequence in clear conditions. A significant improvement in image
quality is demonstrated when using the contrast enhancement algorithm in conjuction with a temporal
filter.

14
2.5 Contrast Restoration of Weather Degraded Images[5]

Images of outdoor scenes captured in bad weather suffer from poor contrast. Under bad weather
conditions, the light reaching a camera is severely scattered by the atmosphere. The resulting decay in
contrast varies across the scene and is exponential in the depths of scene points. Therefore, traditional
space invariant image processing techniques are not sufficient to remove weather effects from images.
In this paper, we present a physics-based model that describes the appearances of scenes in uniform bad
weather conditions. Changes in intensities of scene points under different weather conditions provide
simple constraints to detect depth discontinuities in the scene and also to compute scene structure. Then,
a fast algorithm to restore scene contrast is presented. In contrast to previous techniques, our weather
removal algorithm does not require any a priori scene structure, distributions of scene reflectances, or
detailed knowledge about the particular weather condition. All the methods described in this paper are
effective under a wide range of weather conditions including haze, mist, fog, and conditions arising due
to other aerosols. Further, our methods can be applied to gray scale, RGB color, multispectral and even
IR images. We also extend our techniques to restore contrast of scenes with moving objects, captured
using a video camera

15
2.2 Existing Methodology with references
2.2.1 Contrast restoration using atmospheric model

These methods use physical models to predict the pattern of image degradation and then restore image
contrast with appropriate compensations. They provide better image rendition but usually require extra
information about the imaging system or the imaging environment. Oakley used a physics-based method
to restore scene contrast without any predicted weather information by approximating the distribution of
radiances in the scene by a single gaussian with known variance, however, in most of atmospheric based
models , scene depth need to be estimated beforehand, but it requires more information about the scene
environment, like multiple degraded images taken from the same point or both the clear day and foggy
day images. [1]

Figure 5 Scattering of light in atmosphere

2.2.2 Contrast restoration of iso-depth regions

We now describe a simple method to restore scene contrast from one bad weather image using depth
segmentation of the scene. We define depth segmentation as the extraction of iso-depth regions in the
scene. This does not mean that actual scene depths have to be known. In several situations, it may be
easy to interactively provide the necessary segmentation. For instance, in urban scenes with frontal
views of buildings, a user can easily mark out regions that roughly have the same depths. Later, we will
present two automatic depth segmentation techniques using images taken under different weather
conditions. [5]

16
2.2.3 Contrast enhancement using histogram equilization

The most commonly used contrast enhancement method is histogram equalization and its variations.
Other image enhancement techniques, that uses histogram equalization are also presented, in order to
restore contrast of fog-degraded images, which do not need any information regarding the scene depth
and avoids complicated atmospheric scattering model. Histogram equalization is one of the popular
contrast enhancement algorithm, due to its simplicityand effectiveness. However, since it uses histogram
information over the whole image, as its transformation function, in order to stretch contrast, so it may
not reflect scene depth change in different parts of the image. Thus, the enhancement effect may not be
satisfying when depth changes in the scene. Thus, defining which areas of the image have to be
enhanced becomes important. This method gives, that has significant improvement in the visiblity of an
image enhancement algorithm which finds out which areas of the image are affected by poor contrast,
and then enhances the contrast of only those regions.[1]

2.2.4 Contrast restoration using scene structure

We described a method to restore scene contrast given a depth segmentation of the scene. This method
is simple and effective for scenes where depth changes are abrupt (for example, an urban scene with
frontal views of buildings). However, it is hard to define good depth segmentation when scene depths
change gradually (for instance, a natural scene with mountains or an urban scene with a oblique view of
a road). In this section, we present a method to restore contrast of an arbitary scene using scaled
depths of scene points.We assume that there exists a patch Ezero in the scene whose direct transmission
is zero. This can happen in two instances. First, Ezero can be a black patch with its scene radiance equal to
zero. Note that the black scene patch will not appear black in the image due to the addition of airlight.
Second, Ezero could be a distant scene patch that is completely invisible due to strong airlight. In other
words, this distant scene patch has zero direct transmission and its contrast cannot be restored from a
bad weather image.[3]

2.2.5 Depth edges from two weather conditions

We present a simple cue to automatically locate the depth edges (discontinuities) present in a scene
from two monochrome images taken under different but unknown weather conditions. In other words,
we present a method to label image edges as reflectance edges and depth edges. Several researchers
have pursued the problem of classifying different types of edges (diffuse, specular, and occluding) based
on image intensity/color cues and polarization cues . As we shall show, changes in weather conditions
can be exploited as a cue to differentiate depth edges from reflectance edges. this method to classify
edges can be sensitive to noise, especially under poor weather conditions. Under poor weather
conditions, due to the limited dynamic range of the sensor (typically 8 bits)[5], the direct transmission
(signal) to airlight (noise) ratio can be so low that the direct transmission magnitude can be compared to
the sensor noise level. In this case, the results produced by the method may not be trusted.[5]

17
2.6 Enhancement of Image Using Cost Function Based on Human Visual Model

Improving the Oakley method to make it applicable even when the airlight distribution is not uniform
over the image. In order to estimate the airlight, a cost function that is based on the human visual model
is used in the luminance image. Airlight can then be estimated using a cost function that is based on the
human visual model, wherein a human is more insensitive to variations of the luminance in bright
regions than in dark regions. For this objective, the luminance image is employed for airlight
estimation.The luminance image can be estimated by an appropriate fusion of the R, G, and B
components. Also,the airlight map is estimated using least squares fitting, which models the relationship
between regional airlight and the coordinates of the image pixels. estimating the airlight for each region
and modeling the airlight for each region and the coordinates within the image to generate the airlight
map. In the case of an image with various depth, the contribution of airlight can be varied according to
the region. Estimating the airlight for each region can reflect the variation of depth within the image.
Regions are segmented uniformly to estimate the regional contribution of airlght. to estimate the airlight
using cost function, which is based on human visual model, and Blurred image due to fog is restored by
subtracting airlight map from degraded image.[3]

18
CHAPTER: 03
SOFTWARE DESIGN/
OBSERVATIONS

19
3.1. Design Specification.

 The designed algorithm here deals with every pixel of degraded image
enhancing part degraded by image.
 Here we consider an image of any size to be taken (256 * 256). Sub
block of size desirable is made to find the brightest region and highest
gray level in image to know the highest pixel value to get data of pixel
been degraded.
 Knowing the pixel degraded a new matrix is made with the original
values of image, inserting a threshold finding
 the gradient of every pixel vertically and horizontally for degrading
every pixel of into edge and non edge pixels
 Histogram equalization is performed on edge pixel image.
 The sum of edge pixel and histogram of non-edge pixel is done to get
the improved degraded image.

20
3.2 Flow Chart.

21
Perform histogram equalization on
Non edge pixels

22
3.4. Required Tools and Alternatives

• Matlab 7.0

What is Matlab?

Matlab is a high-level technical computing language and interactive environment for algorithm
development, data visualization, data analysis, and numeric computation. Using the Matlab product, you
can solve technical computing problems faster than with traditional programming languages, such as C,
C++, and Fortran.

You can use Matlab in a wide range of applications, including signal and image processing,
communications, control design, test and measurement, financial modeling and analysis,
andcomputational biology. Add-on toolboxes (collections of special-purpose Matlab functions, available
separately) extend the Matlab environment to solve particular classes of problems in these application
areas.

Matlab provides a number of features for documenting and sharing your work. You can integrate your
Matlab code with other languages and applications, and distribute your Matlab algorithms and
applications.

23
3.5. Standard Stimulation Result.

Figure 6 Original Fog degraded Image

This is the input image given to algorithm which is to be altered and treated at parts of image effected by
fog.

Figure 7 gray scale image of figure 3.1

Image convert to gray scale due to reducing process time. As if original image is taken the process get
longer dealing with R,G,& B as it gets 24 bit image. So converting results in working with 8 bit to get
process farter for now.

24
 Figure 8 Histogram equalization on non edge image obtained.

Histogram equalization done on the nonedge image obtained for alteration of contrast in image to
obtaine a part of visibility from the degradation.

Figure 9 Improved image

Image obtained after sum of edge pixel and histogram equalization of non edge pixel showing us the
difference in level and pixel of degraded image.

25
 Figure 10 Comparison between gray scale image and the degraded image obtained.

Here it can be clearly seen that that after the alteration of degraded pixel in image. The difference
between visibility can be clearly seen.

3.6 Work carried out in odd sem.

• Detailed study of Matlab Image processing toolbox.

• Literature survey from different IEEE technical papers.
• studying various methods to enhance an image.
• Implementation of an existing algorithm with an output of an improved visible image.

3.7 Work targeted for even sem.

• Implementation of other algorithm’s using different methods so to compare and make out the
best way of image enhancing with more probability of degrading an image visibility to greater
extent.

26
CHAPTER: 04
CONCLUSION

27
 Conclusion

we propose an improved fog-degraded image enhancement algorithm applying modified partially

Overlapped sub-block histogram equalization. Experimental results demonstrate that the algorithm is
simple and effective. On observing the visibility of the enhanced image as compared to the degraded
images, we can conclude that the proposed enhancement method is able to enhance the image, better
than the histogram equalization applied globally over the whole image which stretches contrast of
complete image giving unsatisfactory o/p wherelse this algorithm works on every part of image
degraded by fog showing a better enhancement of image.

28
 REFERENCES

[1] Manoj Alwani,Hitendra Gupta,K.K Sharma,” CONTRAST ENHANCEMENT

ALGORITHM FOR FOG-DEGRADED IMAGES” NCVPRIPG 2010,pp 77-78.

[2] YU-FENG LI,”Image Denoising Based on Undecimated Discrete Wavelet Transform”

Proceedings of the 2007 International Conference on Wavelet Analysis and Pattern
Recognition,Beijing, China, 2-4 Nov. 2007.

[3] Y. S. Zhai and X. M. Liu, "An improved fog-degraded image enhancement algorithm,"
Wavelet Analysis and Pattern Recognition,007. ICWAPR'07. International Conference on,
vol. 2, 2007.

[4] Dongjun Kim, Changwon Jeon, Bonghyup Kang and Hanseok Ko,”Enhancement of
Image Degraded by Fog Using Cost Function Based on Human Visual Model”,
Proceedings of IEEE International Conference on Multisensor Fusion and Integration for
Intelligent SystemsSeoul, Korea, August 20 - 22, 2008.

[5] Srinivasa G. Narasimhan and Shree K. Nayar,” Contrast Restoration of Weather Degraded
Images”IEEE Transaction on pattern analysis and machine intelligence, vol 25, no 6, june
2003

29
 HISTORY SHEET

Date and Discussion and Resolutions of the Signature of guide and

Time Meeting students

1. 4\8\10 PPT for Introduction of Project

Small program related to image toolbox
in matlab
2. 11\8\10 Introductory Presentation of project

3. 18\8\10 Algorithm overview and understanding

4. 25\8\10 Implementation of algorithm till

threshold implementation
5. 8\9\10 Understanding of padding command
Literature study
6. 15\9\10 Implementation of Padding in program
& making of sub blocks in image
7. 22\9\10 Complete literature survey and
completion of 40% of program
8. 29\9\10 Mid term presentation of odd sem and
presenting the work done.
9. 13\10\10 Further implementation of program

10. 20\10\10 Showed the program completion till

step 5.
Start report making and ppt.
11. 27\10\10 Correction in reports.
Completion of ppt.

30