SlideShare a Scribd company logo
Exponential Contrast Restoration in Fog
Conditions for Driving Assistance
1
INDEX
• INTRODUCTION
• KOSCHMIEDER’S LAW
• METHOD
• RESULTS
• ADVANTAGES
• CONCLUSION
• REFERENCE
2
INTRODUCTION
• Different natural phenomena can reduce the quality of images and diminish
the visibility.
• Visibility distance is decreased because of the absorption and scattering of
light by the atmospheric particles.
• Images of outdoor scenes, captured during fog conditions, are drastically
degraded.
3
• Due to the presence of fog, the visibility distance decreases exponentially.
• Negative effects of fog on the quality of the image are the loss of contrast and
the alteration of the natural colours in the captured image.
• The scattering effect of the transmitted light causes additional lightness in parts
of the image. This effect is called air-light or atmospheric veil.
• Several algorithms for restoring the contrast of foggy images.
4
• These methods can be categorized in two groups: model and non-model based
enhancement techniques.
• Non-model based methods perform image enhancement relying only on the
information obtained from the image.
• Unfortunately, these methods do not maintain colour fidelity and are not
suitable for real time computer vision.
• Model based contrast restoration techniques can be further divided in two
categories : with given depth and unknown depth.
5
• When the depth is supposed to be known, this information can be used to
restore the original contrast of the image.
• The depth is inferred by using the altitude, tilt and position of the camera.
• It through the manual approximation of the sky area and vanishing point in the
captured image.
• Contrast restoration techniques can be used in a wide range of applications like
in photography fields.
6
KOSCHMIEDER’S LAW
• A relationship between the attenuation of an object’s luminance L at distance
d and the luminance L0 close to the object
• L∞ is the atmospheric luminance and β is the extinction coefficient
• This equation states that the luminance of an object seen through fog is
attenuated with an exponential factor e−βd
7
……………(1)
8
• The atmospheric veil obtained from daylight scattered by fog between the object
and the observer is expressed by L∞ (1 − e−βd )
• The response function of a camera can be applied to the Koschmieder’s
equation for to model the mapping from scene luminance to image intensity.
• Thus, the intensity perceived in the image is the result of a function f applied to
the sum of the air light A and the direct transmission T
……………(2)
9
The eqn represents a linear mapping and assuming that
So we get
where R represents the pixel intensity of the image without fog and A∞ is the
intensity of the sky in fog conditions.
……….(3)
……….(4)
……….(5)
10
METHOD
11
• Our approach is to estimate the atmospheric veil and then to use it in
order to compute the original fog free image.
A. Algorithm Overview
• By using a single image we are not able to compute the depth in the scene
hence we introduce the notion of atmospheric veil (V ) as
……….(6)
……….(7)
……….(8)
……….(9)
12
• The intensity of the sky (A∞) is considered to be equal to 255 or it can be
inferred as the maximum intensity in the image.
B. Obtaining the Atmospheric Veil
• Our visibility enhancement method must be able to work with both gray
scale and color images.
• For to compute the atmospheric veil for color images, use input as gray
level image W that consists in the minimum of each color channel (R,G,B).
13
• This image is called the dark channel prior of a color image.
• The obtained veil V provides the amount of white that must be subtracted
from each color channel
• In order to infer the atmospheric veil we must first examine some
properties that it must have.
• First, the photometric constraint must be introduced , V must be higher or
equal to zero and V must be lower than W
……….(10)
14
0 ≤ V ≤ W
• Another property of the atmospheric veil is that V must be smooth
function
• No black pixel constraint states that the local standard deviation of the
enhanced pixels around a given pixel position must be lower than its local
average.
• We can infer that the veil is smaller or equal to the difference between the
average and the standard deviation of the input image W
V ≤ Average(W) ≤ std(W)
……….(11)
……….(12)
15
• A median filter with a variable size (k) will be applied on the W image
instead of using the classical average
M = mediank (W)
• Use the classical standard deviation computed in each pixel of the image
on the obtained W image
• Fog is more present in the top part of the image and tends to disappear in
the bottom part of the image.
• So the median filter along columns is the one that yields the best results.
……….(13)
16
• Also consider that only a percentage p will be used to calculate the value for
the atmospheric veil in each pixel.
• This percentage is used to control the strength of the restoration process.
• The usual values for p are set from 85% to 99%.
• So the equation for computing the atmospheric veil becomes
……….(14)
17
C. Exponential Inference of the Atmospheric Veil
• The atmospheric veil computed with this method is over compensated in the
bottom part of the image, thus resulting in very dark restored images.
• So this method is not suited for traffic scenes, because it does not model the
link between the veil and the distance to the objects in the scene.
• There is a need for a smooth exponential function.
• For this reason we will model an exponential filter on the atmospheric veil
such that this exponential function decreases inversely with the distance.
18
• Our approach is to model the atmospheric veil with an exponential filter in
order to recover this link.
• We treat the whole atmospheric veil as an exponential function.
19
• The final formula for computing the atmospheric veil is
Vfinal = V . G
where G is an exponential function with values between 0 and 1.
• For modelling the function G, start from two exponential functions from the
partition of unity
• We can call them squared and modulus partition of unity functions.
• Let fso : [−a, a] → [0, 1] (squared) and fmo : [−a, a] → [0, 1] (modulus) having
the following form.
……….(15)
20
• For using the functions fso and fmo in image processing, we must modify
them such that they are defined in the image domain and take values in the
[0,1] interval.
• For this reason the variable x will denote the image lines.
• Hence we obtain fs : [0,H − 1] → [0, 1] and fm : [0,H − 1] → [0, 1], two new
exponential functions with the following form:
……….(17)
……….(16)
21
……….(19)
……….(18)
22
• A translation of our exponential function (along the x axis) has to be applied
by using a linear isomorphism A : [vh,Max] → [0,H − 1]
A(x) = ax + b
having the following properties
……….(20)
……….(21)
23
By solving the system of equations presented in equation (21) the final
exponential function becomes
where c = (H − 1)/(Max − vh)
……….(22)
24
RESULTS
25
26
27
28
• The rate of new visible edges in the enhanced images. This metric can be
used in order to assess the quality of the restoration process.
• The percent of new visible edges in the enhanced image e is represented by
the following formula: nr : total number of edge points in the
enhanced images
no : number of edge points in the original
foggy image.
• The percentage of pixels that become completely black or completely white
after restoration
……….(23)
……….(24)
ADVANTAGES
• Single-image is needed for this enhancement algorithm.
• Able to obtain superior reconstructions of the original fog-free image when
compared with traditional methods.
• It has the ability to adapt the model in accordance to the density of the fog.
• Suitable for contrast restoration in both homogeneous and heterogeneous
fog conditions.
• It can perform contrast restoration in real time for both color and grayscale
images.
29
CONCLUSION
• A new image enhancement method was proposed by taking into account the
exponential decay present in foggy images.
• Computes the restored image by estimating the atmospheric veil.
• The clarity of the reconstructed scene is higher than the one using other
median type filters especially in the regions of the image with many details
• Methods using the squared and modulus translated exponential functions
achieve the best results for image enhancement in fog conditions.
30
31
REFERENCE
• V. Cavallo, M. Colomb, and J. Doré, “Distance perception of vehicle rear lights in fog.” Human Factors,
vol. 43, no. 3, pp. 442–451, 2001.
• M. Negru and S. Nedevschi, “Image based fog detection and visibility estimation for driving assistance
systems,” in Proc. IEEE Int. Conf. ICCP,Sep. 2013, pp. 163–168.
• J. Oakley and H. Bu, “Correction of simple contrast loss in color images,” IEEE Trans. Image Process.,
vol. 16, no. 2, pp. 511–522, Feb. 2007.
• M. Pavlic, H. Belzner, G. Rigoll, and S. Ilic, “Image based fog detection in vehicles,” in Proc. IEEE IV,
Jun. 2012, pp. 1132–1137.
• K. Mori et al., “Recognition of foggy conditions by in-vehicle camera and millimeter wave radar,” in
Proc. IEEE Intell. Veh. Symp., Jun. 2007, pp. 87–92.
32

More Related Content

What's hot

Gradient-Based Low-Light Image Enhancement
Gradient-Based Low-Light Image EnhancementGradient-Based Low-Light Image Enhancement
Gradient-Based Low-Light Image Enhancement
Masayuki Tanaka
 
06 spatial filtering DIP
06 spatial filtering DIP06 spatial filtering DIP
06 spatial filtering DIP
babak danyal
 
Sharpening spatial filters
Sharpening spatial filtersSharpening spatial filters
COM2304: Intensity Transformation and Spatial Filtering – I (Intensity Transf...
COM2304: Intensity Transformation and Spatial Filtering – I (Intensity Transf...COM2304: Intensity Transformation and Spatial Filtering – I (Intensity Transf...
COM2304: Intensity Transformation and Spatial Filtering – I (Intensity Transf...
Hemantha Kulathilake
 
Flash Photography and toonification
Flash Photography and toonificationFlash Photography and toonification
Flash Photography and toonification
Satya Sahoo
 
COM2304: Intensity Transformation and Spatial Filtering – II Spatial Filterin...
COM2304: Intensity Transformation and Spatial Filtering – II Spatial Filterin...COM2304: Intensity Transformation and Spatial Filtering – II Spatial Filterin...
COM2304: Intensity Transformation and Spatial Filtering – II Spatial Filterin...
Hemantha Kulathilake
 
Illumination models
Illumination modelsIllumination models
Illumination models
KABILESH RAMAR
 
Image Enhancement
Image Enhancement Image Enhancement
Image Enhancement
Deven Sahu
 
Filtering an image is to apply a convolution
Filtering an image is to apply a convolutionFiltering an image is to apply a convolution
Filtering an image is to apply a convolution
Abhishek Mukherjee
 
Adaptive unsharp masking
Adaptive unsharp maskingAdaptive unsharp masking
Adaptive unsharp masking
Ravi Teja
 
Ibica2014(p15)image fusion based on broveywavelet
Ibica2014(p15)image fusion based on broveywaveletIbica2014(p15)image fusion based on broveywavelet
Ibica2014(p15)image fusion based on broveywaveletAboul Ella Hassanien
 
Image enhancement ppt nal2
Image enhancement ppt nal2Image enhancement ppt nal2
Image enhancement ppt nal2
Surabhi Ks
 
Moving vehicle detection from Videos
Moving vehicle detection from VideosMoving vehicle detection from Videos
Moving vehicle detection from Videos
PRIYANK BHARDWAJ
 
Shading in OpenGL
Shading in OpenGLShading in OpenGL
Shading in OpenGL
Syed Zaid Irshad
 
Structure and Motion - 3D Reconstruction of Cameras and Structure
Structure and Motion - 3D Reconstruction of Cameras and StructureStructure and Motion - 3D Reconstruction of Cameras and Structure
Structure and Motion - 3D Reconstruction of Cameras and Structure
Giovanni Murru
 
Digital image processing techniques
Digital image processing techniquesDigital image processing techniques
Digital image processing techniques
Shab Bi
 
Chapter 6 Image Processing: Image Enhancement
Chapter 6 Image Processing: Image EnhancementChapter 6 Image Processing: Image Enhancement
Chapter 6 Image Processing: Image Enhancement
Varun Ojha
 
Pre processing of raw rs data
Pre processing of raw rs dataPre processing of raw rs data
Pre processing of raw rs data
guriaghosh
 
Image processing spatialfiltering
Image processing spatialfilteringImage processing spatialfiltering
Image processing spatialfilteringJohn Williams
 
Neural Scene Representation & Rendering: Introduction to Novel View Synthesis
Neural Scene Representation & Rendering: Introduction to Novel View SynthesisNeural Scene Representation & Rendering: Introduction to Novel View Synthesis
Neural Scene Representation & Rendering: Introduction to Novel View Synthesis
Vincent Sitzmann
 

What's hot (20)

Gradient-Based Low-Light Image Enhancement
Gradient-Based Low-Light Image EnhancementGradient-Based Low-Light Image Enhancement
Gradient-Based Low-Light Image Enhancement
 
06 spatial filtering DIP
06 spatial filtering DIP06 spatial filtering DIP
06 spatial filtering DIP
 
Sharpening spatial filters
Sharpening spatial filtersSharpening spatial filters
Sharpening spatial filters
 
COM2304: Intensity Transformation and Spatial Filtering – I (Intensity Transf...
COM2304: Intensity Transformation and Spatial Filtering – I (Intensity Transf...COM2304: Intensity Transformation and Spatial Filtering – I (Intensity Transf...
COM2304: Intensity Transformation and Spatial Filtering – I (Intensity Transf...
 
Flash Photography and toonification
Flash Photography and toonificationFlash Photography and toonification
Flash Photography and toonification
 
COM2304: Intensity Transformation and Spatial Filtering – II Spatial Filterin...
COM2304: Intensity Transformation and Spatial Filtering – II Spatial Filterin...COM2304: Intensity Transformation and Spatial Filtering – II Spatial Filterin...
COM2304: Intensity Transformation and Spatial Filtering – II Spatial Filterin...
 
Illumination models
Illumination modelsIllumination models
Illumination models
 
Image Enhancement
Image Enhancement Image Enhancement
Image Enhancement
 
Filtering an image is to apply a convolution
Filtering an image is to apply a convolutionFiltering an image is to apply a convolution
Filtering an image is to apply a convolution
 
Adaptive unsharp masking
Adaptive unsharp maskingAdaptive unsharp masking
Adaptive unsharp masking
 
Ibica2014(p15)image fusion based on broveywavelet
Ibica2014(p15)image fusion based on broveywaveletIbica2014(p15)image fusion based on broveywavelet
Ibica2014(p15)image fusion based on broveywavelet
 
Image enhancement ppt nal2
Image enhancement ppt nal2Image enhancement ppt nal2
Image enhancement ppt nal2
 
Moving vehicle detection from Videos
Moving vehicle detection from VideosMoving vehicle detection from Videos
Moving vehicle detection from Videos
 
Shading in OpenGL
Shading in OpenGLShading in OpenGL
Shading in OpenGL
 
Structure and Motion - 3D Reconstruction of Cameras and Structure
Structure and Motion - 3D Reconstruction of Cameras and StructureStructure and Motion - 3D Reconstruction of Cameras and Structure
Structure and Motion - 3D Reconstruction of Cameras and Structure
 
Digital image processing techniques
Digital image processing techniquesDigital image processing techniques
Digital image processing techniques
 
Chapter 6 Image Processing: Image Enhancement
Chapter 6 Image Processing: Image EnhancementChapter 6 Image Processing: Image Enhancement
Chapter 6 Image Processing: Image Enhancement
 
Pre processing of raw rs data
Pre processing of raw rs dataPre processing of raw rs data
Pre processing of raw rs data
 
Image processing spatialfiltering
Image processing spatialfilteringImage processing spatialfiltering
Image processing spatialfiltering
 
Neural Scene Representation & Rendering: Introduction to Novel View Synthesis
Neural Scene Representation & Rendering: Introduction to Novel View SynthesisNeural Scene Representation & Rendering: Introduction to Novel View Synthesis
Neural Scene Representation & Rendering: Introduction to Novel View Synthesis
 

Viewers also liked

Presentación power point
Presentación power pointPresentación power point
Presentación power point
haixa08
 
Curso hechos
Curso hechosCurso hechos
Curso hechos
Avivamiento LA VID
 
Inteh
IntehInteh
Inteh
dnzcpto1
 
Hematopoyesis
HematopoyesisHematopoyesis
Hematopoyesis
anabellebarrera
 
My Documents
My DocumentsMy Documents
My Documents
Holly Meyer
 
Практична робота №3
Практична робота №3Практична робота №3
Практична робота №3
dnzcpto1
 
Valla maristas
Valla   maristasValla   maristas
Valla maristas
Santiago Robledo
 
Практична робота №4
Практична робота №4Практична робота №4
Практична робота №4
dnzcpto1
 
Anatomia piel1
Anatomia piel1Anatomia piel1
Anatomia piel1
anabellebarrera
 
COMMUNITY MANAGER
COMMUNITY MANAGERCOMMUNITY MANAGER
COMMUNITY MANAGER
sannriza
 
Video juegos
Video juegosVideo juegos
7
77
Effective image retrieval system using dot diffused
Effective image retrieval system using dot diffusedEffective image retrieval system using dot diffused
Effective image retrieval system using dot diffused
SREEKUTTY SREEKUMAR
 
Знайомство з комп’ютерними презентаціями
Знайомство з комп’ютерними презентаціямиЗнайомство з комп’ютерними презентаціями
Знайомство з комп’ютерними презентаціями
dnzcpto1
 
Bitacoratic
BitacoraticBitacoratic
Bitacoratic
samanta stolerman
 
Anatomía del corazón
Anatomía del corazónAnatomía del corazón
Anatomía del corazón
anabellebarrera
 

Viewers also liked (19)

Presentation1 (2)
Presentation1 (2)Presentation1 (2)
Presentation1 (2)
 
Presentación power point
Presentación power pointPresentación power point
Presentación power point
 
Curso hechos
Curso hechosCurso hechos
Curso hechos
 
Inteh
IntehInteh
Inteh
 
Hematopoyesis
HematopoyesisHematopoyesis
Hematopoyesis
 
My Documents
My DocumentsMy Documents
My Documents
 
Resume (Dominic Chandran)
Resume (Dominic Chandran)Resume (Dominic Chandran)
Resume (Dominic Chandran)
 
Практична робота №3
Практична робота №3Практична робота №3
Практична робота №3
 
Valla maristas
Valla   maristasValla   maristas
Valla maristas
 
Практична робота №4
Практична робота №4Практична робота №4
Практична робота №4
 
Anatomia piel1
Anatomia piel1Anatomia piel1
Anatomia piel1
 
COMMUNITY MANAGER
COMMUNITY MANAGERCOMMUNITY MANAGER
COMMUNITY MANAGER
 
leoni_insider_2_2014
leoni_insider_2_2014leoni_insider_2_2014
leoni_insider_2_2014
 
Video juegos
Video juegosVideo juegos
Video juegos
 
7
77
7
 
Effective image retrieval system using dot diffused
Effective image retrieval system using dot diffusedEffective image retrieval system using dot diffused
Effective image retrieval system using dot diffused
 
Знайомство з комп’ютерними презентаціями
Знайомство з комп’ютерними презентаціямиЗнайомство з комп’ютерними презентаціями
Знайомство з комп’ютерними презентаціями
 
Bitacoratic
BitacoraticBitacoratic
Bitacoratic
 
Anatomía del corazón
Anatomía del corazónAnatomía del corazón
Anatomía del corazón
 

Similar to Exponential contrast restoration in fog

An Efficient Visibility Enhancement Algorithm for Road Scenes Captured by Int...
An Efficient Visibility Enhancement Algorithm for Road Scenes Captured by Int...An Efficient Visibility Enhancement Algorithm for Road Scenes Captured by Int...
An Efficient Visibility Enhancement Algorithm for Road Scenes Captured by Int...
madhuricts
 
Lm342080283
Lm342080283Lm342080283
Lm342080283
IJERA Editor
 
New microsoft power point presentation
New microsoft power point presentationNew microsoft power point presentation
New microsoft power point presentation
Azad Singh
 
A Review on Haze Removal Techniques
A Review on Haze Removal TechniquesA Review on Haze Removal Techniques
A Review on Haze Removal Techniques
IRJET Journal
 
matdid950092.pdf
matdid950092.pdfmatdid950092.pdf
matdid950092.pdf
lencho3d
 
Image enhancement
Image enhancementImage enhancement
Image enhancementAyaelshiwi
 
A Survey on Single Image Dehazing Approaches
A Survey on Single Image Dehazing ApproachesA Survey on Single Image Dehazing Approaches
A Survey on Single Image Dehazing Approaches
IRJET Journal
 
High Efficiency Haze Removal Using Contextual Regularization Algorithm
High Efficiency Haze Removal Using Contextual Regularization AlgorithmHigh Efficiency Haze Removal Using Contextual Regularization Algorithm
High Efficiency Haze Removal Using Contextual Regularization Algorithm
IRJET Journal
 
Digital image processing
Digital image processingDigital image processing
Digital image processing
Chetan Hulsure
 
Review of Linear Image Degradation and Image Restoration Technique
Review of Linear Image Degradation and Image Restoration TechniqueReview of Linear Image Degradation and Image Restoration Technique
Review of Linear Image Degradation and Image Restoration Technique
BRNSSPublicationHubI
 
M.sc. m hassan
M.sc. m hassanM.sc. m hassan
M.sc. m hassan
Ashraf Aboshosha
 
C0363016021
C0363016021C0363016021
C0363016021
theijes
 
image enhancement technique for aerial imagery
image enhancement technique for aerial imageryimage enhancement technique for aerial imagery
image enhancement technique for aerial imagery
Gottipati Mohana Krishna
 
Digital image processing - Image Enhancement (MATERIAL)
Digital image processing  - Image Enhancement (MATERIAL)Digital image processing  - Image Enhancement (MATERIAL)
Digital image processing - Image Enhancement (MATERIAL)
Mathankumar S
 
Digital Image Processing - Image Enhancement
Digital Image Processing  - Image EnhancementDigital Image Processing  - Image Enhancement
Digital Image Processing - Image Enhancement
Mathankumar S
 
Single Image Fog Removal Based on Fusion Strategy
Single Image Fog Removal Based on Fusion Strategy Single Image Fog Removal Based on Fusion Strategy
Single Image Fog Removal Based on Fusion Strategy
csandit
 
graphics notes
graphics notesgraphics notes
graphics notes
Sonia Pahuja
 
Multiexposure Image Fusion
Multiexposure Image FusionMultiexposure Image Fusion
Multiexposure Image Fusion
IJMER
 
Fisheye Omnidirectional View in Autonomous Driving
Fisheye Omnidirectional View in Autonomous DrivingFisheye Omnidirectional View in Autonomous Driving
Fisheye Omnidirectional View in Autonomous Driving
Yu Huang
 
Shadow removal using Image Processing (Case study and code Implementation)
Shadow removal using Image Processing (Case study and code Implementation)Shadow removal using Image Processing (Case study and code Implementation)
Shadow removal using Image Processing (Case study and code Implementation)
Ardra
 

Similar to Exponential contrast restoration in fog (20)

An Efficient Visibility Enhancement Algorithm for Road Scenes Captured by Int...
An Efficient Visibility Enhancement Algorithm for Road Scenes Captured by Int...An Efficient Visibility Enhancement Algorithm for Road Scenes Captured by Int...
An Efficient Visibility Enhancement Algorithm for Road Scenes Captured by Int...
 
Lm342080283
Lm342080283Lm342080283
Lm342080283
 
New microsoft power point presentation
New microsoft power point presentationNew microsoft power point presentation
New microsoft power point presentation
 
A Review on Haze Removal Techniques
A Review on Haze Removal TechniquesA Review on Haze Removal Techniques
A Review on Haze Removal Techniques
 
matdid950092.pdf
matdid950092.pdfmatdid950092.pdf
matdid950092.pdf
 
Image enhancement
Image enhancementImage enhancement
Image enhancement
 
A Survey on Single Image Dehazing Approaches
A Survey on Single Image Dehazing ApproachesA Survey on Single Image Dehazing Approaches
A Survey on Single Image Dehazing Approaches
 
High Efficiency Haze Removal Using Contextual Regularization Algorithm
High Efficiency Haze Removal Using Contextual Regularization AlgorithmHigh Efficiency Haze Removal Using Contextual Regularization Algorithm
High Efficiency Haze Removal Using Contextual Regularization Algorithm
 
Digital image processing
Digital image processingDigital image processing
Digital image processing
 
Review of Linear Image Degradation and Image Restoration Technique
Review of Linear Image Degradation and Image Restoration TechniqueReview of Linear Image Degradation and Image Restoration Technique
Review of Linear Image Degradation and Image Restoration Technique
 
M.sc. m hassan
M.sc. m hassanM.sc. m hassan
M.sc. m hassan
 
C0363016021
C0363016021C0363016021
C0363016021
 
image enhancement technique for aerial imagery
image enhancement technique for aerial imageryimage enhancement technique for aerial imagery
image enhancement technique for aerial imagery
 
Digital image processing - Image Enhancement (MATERIAL)
Digital image processing  - Image Enhancement (MATERIAL)Digital image processing  - Image Enhancement (MATERIAL)
Digital image processing - Image Enhancement (MATERIAL)
 
Digital Image Processing - Image Enhancement
Digital Image Processing  - Image EnhancementDigital Image Processing  - Image Enhancement
Digital Image Processing - Image Enhancement
 
Single Image Fog Removal Based on Fusion Strategy
Single Image Fog Removal Based on Fusion Strategy Single Image Fog Removal Based on Fusion Strategy
Single Image Fog Removal Based on Fusion Strategy
 
graphics notes
graphics notesgraphics notes
graphics notes
 
Multiexposure Image Fusion
Multiexposure Image FusionMultiexposure Image Fusion
Multiexposure Image Fusion
 
Fisheye Omnidirectional View in Autonomous Driving
Fisheye Omnidirectional View in Autonomous DrivingFisheye Omnidirectional View in Autonomous Driving
Fisheye Omnidirectional View in Autonomous Driving
 
Shadow removal using Image Processing (Case study and code Implementation)
Shadow removal using Image Processing (Case study and code Implementation)Shadow removal using Image Processing (Case study and code Implementation)
Shadow removal using Image Processing (Case study and code Implementation)
 

Recently uploaded

Railway Signalling Principles Edition 3.pdf
Railway Signalling Principles Edition 3.pdfRailway Signalling Principles Edition 3.pdf
Railway Signalling Principles Edition 3.pdf
TeeVichai
 
AKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdf
AKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdfAKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdf
AKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdf
SamSarthak3
 
Tutorial for 16S rRNA Gene Analysis with QIIME2.pdf
Tutorial for 16S rRNA Gene Analysis with QIIME2.pdfTutorial for 16S rRNA Gene Analysis with QIIME2.pdf
Tutorial for 16S rRNA Gene Analysis with QIIME2.pdf
aqil azizi
 
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
zwunae
 
AP LAB PPT.pdf ap lab ppt no title specific
AP LAB PPT.pdf ap lab ppt no title specificAP LAB PPT.pdf ap lab ppt no title specific
AP LAB PPT.pdf ap lab ppt no title specific
BrazilAccount1
 
Unbalanced Three Phase Systems and circuits.pptx
Unbalanced Three Phase Systems and circuits.pptxUnbalanced Three Phase Systems and circuits.pptx
Unbalanced Three Phase Systems and circuits.pptx
ChristineTorrepenida1
 
Water Industry Process Automation and Control Monthly - May 2024.pdf
Water Industry Process Automation and Control Monthly - May 2024.pdfWater Industry Process Automation and Control Monthly - May 2024.pdf
Water Industry Process Automation and Control Monthly - May 2024.pdf
Water Industry Process Automation & Control
 
Top 10 Oil and Gas Projects in Saudi Arabia 2024.pdf
Top 10 Oil and Gas Projects in Saudi Arabia 2024.pdfTop 10 Oil and Gas Projects in Saudi Arabia 2024.pdf
Top 10 Oil and Gas Projects in Saudi Arabia 2024.pdf
Teleport Manpower Consultant
 
Design and Analysis of Algorithms-DP,Backtracking,Graphs,B&B
Design and Analysis of Algorithms-DP,Backtracking,Graphs,B&BDesign and Analysis of Algorithms-DP,Backtracking,Graphs,B&B
Design and Analysis of Algorithms-DP,Backtracking,Graphs,B&B
Sreedhar Chowdam
 
Immunizing Image Classifiers Against Localized Adversary Attacks
Immunizing Image Classifiers Against Localized Adversary AttacksImmunizing Image Classifiers Against Localized Adversary Attacks
Immunizing Image Classifiers Against Localized Adversary Attacks
gerogepatton
 
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdf
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdfHybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdf
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdf
fxintegritypublishin
 
Fundamentals of Electric Drives and its applications.pptx
Fundamentals of Electric Drives and its applications.pptxFundamentals of Electric Drives and its applications.pptx
Fundamentals of Electric Drives and its applications.pptx
manasideore6
 
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
bakpo1
 
HYDROPOWER - Hydroelectric power generation
HYDROPOWER - Hydroelectric power generationHYDROPOWER - Hydroelectric power generation
HYDROPOWER - Hydroelectric power generation
Robbie Edward Sayers
 
Forklift Classes Overview by Intella Parts
Forklift Classes Overview by Intella PartsForklift Classes Overview by Intella Parts
Forklift Classes Overview by Intella Parts
Intella Parts
 
Nuclear Power Economics and Structuring 2024
Nuclear Power Economics and Structuring 2024Nuclear Power Economics and Structuring 2024
Nuclear Power Economics and Structuring 2024
Massimo Talia
 
space technology lecture notes on satellite
space technology lecture notes on satellitespace technology lecture notes on satellite
space technology lecture notes on satellite
ongomchris
 
Building Electrical System Design & Installation
Building Electrical System Design & InstallationBuilding Electrical System Design & Installation
Building Electrical System Design & Installation
symbo111
 
Gen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdfGen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdf
gdsczhcet
 
Standard Reomte Control Interface - Neometrix
Standard Reomte Control Interface - NeometrixStandard Reomte Control Interface - Neometrix
Standard Reomte Control Interface - Neometrix
Neometrix_Engineering_Pvt_Ltd
 

Recently uploaded (20)

Railway Signalling Principles Edition 3.pdf
Railway Signalling Principles Edition 3.pdfRailway Signalling Principles Edition 3.pdf
Railway Signalling Principles Edition 3.pdf
 
AKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdf
AKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdfAKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdf
AKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdf
 
Tutorial for 16S rRNA Gene Analysis with QIIME2.pdf
Tutorial for 16S rRNA Gene Analysis with QIIME2.pdfTutorial for 16S rRNA Gene Analysis with QIIME2.pdf
Tutorial for 16S rRNA Gene Analysis with QIIME2.pdf
 
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
 
AP LAB PPT.pdf ap lab ppt no title specific
AP LAB PPT.pdf ap lab ppt no title specificAP LAB PPT.pdf ap lab ppt no title specific
AP LAB PPT.pdf ap lab ppt no title specific
 
Unbalanced Three Phase Systems and circuits.pptx
Unbalanced Three Phase Systems and circuits.pptxUnbalanced Three Phase Systems and circuits.pptx
Unbalanced Three Phase Systems and circuits.pptx
 
Water Industry Process Automation and Control Monthly - May 2024.pdf
Water Industry Process Automation and Control Monthly - May 2024.pdfWater Industry Process Automation and Control Monthly - May 2024.pdf
Water Industry Process Automation and Control Monthly - May 2024.pdf
 
Top 10 Oil and Gas Projects in Saudi Arabia 2024.pdf
Top 10 Oil and Gas Projects in Saudi Arabia 2024.pdfTop 10 Oil and Gas Projects in Saudi Arabia 2024.pdf
Top 10 Oil and Gas Projects in Saudi Arabia 2024.pdf
 
Design and Analysis of Algorithms-DP,Backtracking,Graphs,B&B
Design and Analysis of Algorithms-DP,Backtracking,Graphs,B&BDesign and Analysis of Algorithms-DP,Backtracking,Graphs,B&B
Design and Analysis of Algorithms-DP,Backtracking,Graphs,B&B
 
Immunizing Image Classifiers Against Localized Adversary Attacks
Immunizing Image Classifiers Against Localized Adversary AttacksImmunizing Image Classifiers Against Localized Adversary Attacks
Immunizing Image Classifiers Against Localized Adversary Attacks
 
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdf
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdfHybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdf
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdf
 
Fundamentals of Electric Drives and its applications.pptx
Fundamentals of Electric Drives and its applications.pptxFundamentals of Electric Drives and its applications.pptx
Fundamentals of Electric Drives and its applications.pptx
 
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
 
HYDROPOWER - Hydroelectric power generation
HYDROPOWER - Hydroelectric power generationHYDROPOWER - Hydroelectric power generation
HYDROPOWER - Hydroelectric power generation
 
Forklift Classes Overview by Intella Parts
Forklift Classes Overview by Intella PartsForklift Classes Overview by Intella Parts
Forklift Classes Overview by Intella Parts
 
Nuclear Power Economics and Structuring 2024
Nuclear Power Economics and Structuring 2024Nuclear Power Economics and Structuring 2024
Nuclear Power Economics and Structuring 2024
 
space technology lecture notes on satellite
space technology lecture notes on satellitespace technology lecture notes on satellite
space technology lecture notes on satellite
 
Building Electrical System Design & Installation
Building Electrical System Design & InstallationBuilding Electrical System Design & Installation
Building Electrical System Design & Installation
 
Gen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdfGen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdf
 
Standard Reomte Control Interface - Neometrix
Standard Reomte Control Interface - NeometrixStandard Reomte Control Interface - Neometrix
Standard Reomte Control Interface - Neometrix
 

Exponential contrast restoration in fog

  • 1. Exponential Contrast Restoration in Fog Conditions for Driving Assistance 1
  • 2. INDEX • INTRODUCTION • KOSCHMIEDER’S LAW • METHOD • RESULTS • ADVANTAGES • CONCLUSION • REFERENCE 2
  • 3. INTRODUCTION • Different natural phenomena can reduce the quality of images and diminish the visibility. • Visibility distance is decreased because of the absorption and scattering of light by the atmospheric particles. • Images of outdoor scenes, captured during fog conditions, are drastically degraded. 3
  • 4. • Due to the presence of fog, the visibility distance decreases exponentially. • Negative effects of fog on the quality of the image are the loss of contrast and the alteration of the natural colours in the captured image. • The scattering effect of the transmitted light causes additional lightness in parts of the image. This effect is called air-light or atmospheric veil. • Several algorithms for restoring the contrast of foggy images. 4
  • 5. • These methods can be categorized in two groups: model and non-model based enhancement techniques. • Non-model based methods perform image enhancement relying only on the information obtained from the image. • Unfortunately, these methods do not maintain colour fidelity and are not suitable for real time computer vision. • Model based contrast restoration techniques can be further divided in two categories : with given depth and unknown depth. 5
  • 6. • When the depth is supposed to be known, this information can be used to restore the original contrast of the image. • The depth is inferred by using the altitude, tilt and position of the camera. • It through the manual approximation of the sky area and vanishing point in the captured image. • Contrast restoration techniques can be used in a wide range of applications like in photography fields. 6
  • 7. KOSCHMIEDER’S LAW • A relationship between the attenuation of an object’s luminance L at distance d and the luminance L0 close to the object • L∞ is the atmospheric luminance and β is the extinction coefficient • This equation states that the luminance of an object seen through fog is attenuated with an exponential factor e−βd 7 ……………(1)
  • 8. 8 • The atmospheric veil obtained from daylight scattered by fog between the object and the observer is expressed by L∞ (1 − e−βd ) • The response function of a camera can be applied to the Koschmieder’s equation for to model the mapping from scene luminance to image intensity. • Thus, the intensity perceived in the image is the result of a function f applied to the sum of the air light A and the direct transmission T ……………(2)
  • 9. 9 The eqn represents a linear mapping and assuming that So we get where R represents the pixel intensity of the image without fog and A∞ is the intensity of the sky in fog conditions. ……….(3) ……….(4) ……….(5)
  • 11. 11 • Our approach is to estimate the atmospheric veil and then to use it in order to compute the original fog free image. A. Algorithm Overview • By using a single image we are not able to compute the depth in the scene hence we introduce the notion of atmospheric veil (V ) as ……….(6) ……….(7) ……….(8) ……….(9)
  • 12. 12 • The intensity of the sky (A∞) is considered to be equal to 255 or it can be inferred as the maximum intensity in the image. B. Obtaining the Atmospheric Veil • Our visibility enhancement method must be able to work with both gray scale and color images. • For to compute the atmospheric veil for color images, use input as gray level image W that consists in the minimum of each color channel (R,G,B).
  • 13. 13 • This image is called the dark channel prior of a color image. • The obtained veil V provides the amount of white that must be subtracted from each color channel • In order to infer the atmospheric veil we must first examine some properties that it must have. • First, the photometric constraint must be introduced , V must be higher or equal to zero and V must be lower than W ……….(10)
  • 14. 14 0 ≤ V ≤ W • Another property of the atmospheric veil is that V must be smooth function • No black pixel constraint states that the local standard deviation of the enhanced pixels around a given pixel position must be lower than its local average. • We can infer that the veil is smaller or equal to the difference between the average and the standard deviation of the input image W V ≤ Average(W) ≤ std(W) ……….(11) ……….(12)
  • 15. 15 • A median filter with a variable size (k) will be applied on the W image instead of using the classical average M = mediank (W) • Use the classical standard deviation computed in each pixel of the image on the obtained W image • Fog is more present in the top part of the image and tends to disappear in the bottom part of the image. • So the median filter along columns is the one that yields the best results. ……….(13)
  • 16. 16 • Also consider that only a percentage p will be used to calculate the value for the atmospheric veil in each pixel. • This percentage is used to control the strength of the restoration process. • The usual values for p are set from 85% to 99%. • So the equation for computing the atmospheric veil becomes ……….(14)
  • 17. 17 C. Exponential Inference of the Atmospheric Veil • The atmospheric veil computed with this method is over compensated in the bottom part of the image, thus resulting in very dark restored images. • So this method is not suited for traffic scenes, because it does not model the link between the veil and the distance to the objects in the scene. • There is a need for a smooth exponential function. • For this reason we will model an exponential filter on the atmospheric veil such that this exponential function decreases inversely with the distance.
  • 18. 18 • Our approach is to model the atmospheric veil with an exponential filter in order to recover this link. • We treat the whole atmospheric veil as an exponential function.
  • 19. 19 • The final formula for computing the atmospheric veil is Vfinal = V . G where G is an exponential function with values between 0 and 1. • For modelling the function G, start from two exponential functions from the partition of unity • We can call them squared and modulus partition of unity functions. • Let fso : [−a, a] → [0, 1] (squared) and fmo : [−a, a] → [0, 1] (modulus) having the following form. ……….(15)
  • 20. 20 • For using the functions fso and fmo in image processing, we must modify them such that they are defined in the image domain and take values in the [0,1] interval. • For this reason the variable x will denote the image lines. • Hence we obtain fs : [0,H − 1] → [0, 1] and fm : [0,H − 1] → [0, 1], two new exponential functions with the following form: ……….(17) ……….(16)
  • 22. 22 • A translation of our exponential function (along the x axis) has to be applied by using a linear isomorphism A : [vh,Max] → [0,H − 1] A(x) = ax + b having the following properties ……….(20) ……….(21)
  • 23. 23 By solving the system of equations presented in equation (21) the final exponential function becomes where c = (H − 1)/(Max − vh) ……….(22)
  • 25. 25
  • 26. 26
  • 27. 27
  • 28. 28 • The rate of new visible edges in the enhanced images. This metric can be used in order to assess the quality of the restoration process. • The percent of new visible edges in the enhanced image e is represented by the following formula: nr : total number of edge points in the enhanced images no : number of edge points in the original foggy image. • The percentage of pixels that become completely black or completely white after restoration ……….(23) ……….(24)
  • 29. ADVANTAGES • Single-image is needed for this enhancement algorithm. • Able to obtain superior reconstructions of the original fog-free image when compared with traditional methods. • It has the ability to adapt the model in accordance to the density of the fog. • Suitable for contrast restoration in both homogeneous and heterogeneous fog conditions. • It can perform contrast restoration in real time for both color and grayscale images. 29
  • 30. CONCLUSION • A new image enhancement method was proposed by taking into account the exponential decay present in foggy images. • Computes the restored image by estimating the atmospheric veil. • The clarity of the reconstructed scene is higher than the one using other median type filters especially in the regions of the image with many details • Methods using the squared and modulus translated exponential functions achieve the best results for image enhancement in fog conditions. 30
  • 31. 31 REFERENCE • V. Cavallo, M. Colomb, and J. Doré, “Distance perception of vehicle rear lights in fog.” Human Factors, vol. 43, no. 3, pp. 442–451, 2001. • M. Negru and S. Nedevschi, “Image based fog detection and visibility estimation for driving assistance systems,” in Proc. IEEE Int. Conf. ICCP,Sep. 2013, pp. 163–168. • J. Oakley and H. Bu, “Correction of simple contrast loss in color images,” IEEE Trans. Image Process., vol. 16, no. 2, pp. 511–522, Feb. 2007. • M. Pavlic, H. Belzner, G. Rigoll, and S. Ilic, “Image based fog detection in vehicles,” in Proc. IEEE IV, Jun. 2012, pp. 1132–1137. • K. Mori et al., “Recognition of foggy conditions by in-vehicle camera and millimeter wave radar,” in Proc. IEEE Intell. Veh. Symp., Jun. 2007, pp. 87–92.
  • 32. 32