SlideShare ist ein Scribd-Unternehmen logo
1 von 36
ATMOSPHERIC TURBULENCE DEGRADED IMAGE RESTORATION USING BACK
PROPAGATION NEURAL NETWORK
A
DISSERTATION
Presented
In partial fulfillment of the requirement for the award of degree of
MASTER OF TECHNOLOGY
IN
INFORMATION TECHNOLOGY
Submitted by
Azad Singh
(0901IT13MT03)
Under the supervision of
Rajeev Kumar Singh
(Assistant Professor)
Department of Computer Science & Engineering and Information Technology
Madhav Institute of Technology & Science, Gwalior (MP) - 474005
Session 2013-2015
Contents
• Introduction
 Image restoration
Atmospheric blur
Artificial neural network
• Literature Survey
• Problem Statement
• Proposed methodology
• Result analysis
• Conclusions and future work
• References
• Publications
Introduction
• Image restoration is a technique used to recover image from degraded image. Image
may be distorted due to blur and noise; blur can occur due to atmospheric
turbulence, motion of objects and camera mis-focus[1].
• Image Restoration is an area that also deals with improving the appearance of an
image.
• The degradation model of imagine system can be define as,
𝑔 𝑘 = 𝐼 𝑘⦻ℎ + 𝑛 𝑘 (1)
Where, 𝑔 𝑘 is the degraded image, in which ⦻ denotes the two- dimensional linear
convolution operation,𝐼 𝑘 is the original image and h is point spread function 𝑛 𝑘 is
the additive noise.
Continue…
Image Restoration Techniques:
i. Deterministic method:-In deterministic method [2] the prior knowledge about
degradation of image so that we employ the method easily and get restored image.
In deterministic method we restore image using filters.
(1) Wiener filter[3]
(2) Median filter[4]
(3) Inverse filter[5]
ii. Stochastic method: - The stochastic technique [6] uses the probability theory to
form most likely result, given data. In stochastic method does not known prior
knowledge of degradation source. So in this method firstly estimate the point spread
function of blurred image. It is also called as blind de-convolution.
Continue…
• Atmospheric blur is the distortion of image due to long time exposure, fog, wind
speed and due to randomly change in refractive index of air through which light
travels.
• The point spread function of atmospheric turbulence image can be described
reasonably well by a Gaussian function[7].
ℎ 𝑥, 𝑦, 𝜎 𝐺 = 𝐶 𝑒𝑥𝑝
𝑥2+𝑦2
2𝜎2
𝐺
(2)
Here, 𝜎 𝐺 determines the amount of spread of the blur, and the constant 𝐶 is to be
chosen so that above equation is satisfied.
• The estimation of point spread function of atmospheric turbulence image is very
challenging without knowing prior knowledge of clear image[8]. So, it is very
difficult to restore image using point spread function (PSF).
Example of images
• Here see images which is affected due to atmospheric blur given below:
Figure 1: (a) image is blur (b) image is blur due to
due to fog temperature i.e., mirage
Continue…
• The effects to atmospheric turbulence can be measured by calculating the
scintillation index[9]. This index is related to the mean and standard deviation of
the intensity distribution of image.
• The atmospheric turbulence can be categorized in two categories:
 In rigid bodies restoration, we take single frame image which is degraded
through turbulence and restoration techniques will be applied in degraded image,
which is known as single image deconvolution [10].
 In non-rigid bodies, we take degraded videos and apply image registration
technique to restore the object frame of video sequences, so uses the different
registration algorithm for image registration.
Artificial Neural Network
• Artificial Neural Network [ANN] is an information paradigm which is inspired by
our biological nervous system, like brain and processing system. In which artificial
neurons are interconnected to each other through weights [11].
• It can be trained through the supervised learning process. An ANN is composed of a
large number of highly interconnected processing elements neurons working in
union to solve specific problem.
• The simple neural network is a two layers or multilayer feed-forward and feedback
network with 𝑀 input and 𝑁 output units. Each input unit is connected to each of
the output units, and each connection is associated with weight, which are denoted
as 𝑤1, 𝑤2 … 𝑤 𝑁 respectively or the strength of the connection
Continue…
.
X 1 Z 1 Y 1
X i-n Z i-n Y i-n
X n Z n Y n
Figure 2: Architecture of feed-forward network
A trained network can be thought as an “expert” in a particular category of
information, if it has given to analyze. This is architecture of simple feed
forward network
Input layer Hidden layer Output layer
Literature Survey
• Dalong Li and Steven Simske [12] proposed kurtosis minimization method for
restoration. Atmospheric turbulence is caused by the random fluctuations of the
refraction index of the medium. It can lead to blurring in images acquired from a
long distance away. Since the degradation is often not completely known, the
problem is viewed as blind image deconvolution or blur identification.
• Their work has observed that blurring increases kurtosis and introduced a new blur
identification method based on kurtosis minimization[13].
Continue…
• Luxin Yan, Mingzhi Jin, Houzhang Fang, Hai Liu, and Tianxu Zhang [14]
proposed Atmospheric turbulence affects imaging systems by virtue of wave
propagation through a medium with a non-uniform index of refraction. It can lead
to blurring in images acquired from a long distance away. In this letter, it is
observed that blurring increases the second-order central moment (SOCM)of
images, and we introduce a new parametric blur identification method by
minimizing SOCM.
• The method applies to finite-support images, in which the scene consists of a finite-
extent object against a uniformly black, gray, or white background. The SOCM
method has been validated by direct comparisons with other methods on simulated
and real degraded images.
• Dalong Li, Russell M. Mersereau, and Steven Simske [15] uses PCA revealed a
connection between blind image deconvolution and principal components analysis
(PCA).
Continue…
• Although PCA is derived from blur models that do not contain additive noise, it can
be justified on both theoretical and experimental grounds that the PCA-based
restoration algorithm is actually robust to the presence of white noise.
• The algorithm is applied to the restoration of atmospheric turbulence-degraded
imagery and compared to wiener filter [14] on both real and simulated atmospheric
turbulence blurred images. It is shown that the PCA-based blind image
deconvolution runs faster and is more robust to noise. PCA is applied on only shift
invariant blur image and do not consider spatially varying blur image.
• PCA based image deconvolution algorithm is fast, robust to noise and provide
better result in multichannel image.
Continue…
• Jin-Bao Wang, Ning He, Lu-Lu Zhang, and Ke Lu [15] provide fog removal
techniques using physical model and dark channel used for atmospheric scattering
and optical reflecting imaging for estimation of atmospheric light in the dehazed
image to get proper result.
• It is mostly for non-sky patches, as at least one color channel has low intensity at
some pixel, it can be happen due to shadows of objects [16]. As we know
atmospheric images are usually full of shadows, so due to fog image is more
contrast than its image without fog.
• After apply the normalize operation we get restored image and improve the
visibility through CLAHE (Contrast Limited Adaptive Histogram Equalisation).
• This technique is useful for removing haze but provide better result in presence of
different noise and other blur present in foggy image.
Continue…
• Anantrasirirchai, Achi,G.Kingsbury and David R. Bull revealed complex
wavelet [17] for image fusion for restore the image. A new approach that
overcomes the problem of ineffective occur due to conventional fusion methods for
large distortion and also time consuming.
• The conventional fusion methods require a large number of frames to select lucky
regions. Image registration using the phase of a DTCWT coefficient is robust to
noise and temporal intensity variations thereby providing an efficient tool for
removing distorting ripples. After fusion, the effect of haze is reduced using locally-
adaptive histogram equalization.
Continue…
• Zhu and Milanfar [18] explored the technique to restore atmospheric turbulence
image and reduces the space and time-varying problem to shift invariant one. Thus
method divides each frame into overlapping patches. These patches can be viewed
in small region containing space-invariant blur. So blur can be deblurred using
multi-frame blind deconvolution algorithm. In which calculate the PSF (Point
Spread Function) for each patches for deconvolution.
Problem Statement
The atmospheric turbulence image restoration becomes more beneficial for
numerous vision applications. The following gaps concluded using literature
survey:-
1. In the past years, many researchers have provided techniques for restoration of fog
or turbulence independently but not both and the restored image was not perfectly
retrieved by traditional filter.
2. The existing methods have neglected to reduce the noise issue, which is presented
in the output images.
3. The problem of the uneven illuminate in image is also neglected.
Proposed methodology
• In our proposed methodology, we create new feed-forward network which have 20
hidden layers and one output layer. The activation function is associated with each
layer of hidden layer and output layer of network. At the hidden layer tangent
function and output layer uniform sigmoidal function is used as shown in equation 3
and 4. The training of network is done by back propagation. The levenberg-
Marquardt algorithm uses in back propagation and MSE for performance measure,
perform number of iteration and validation check to get better result.
• Unipolar sigmoid: 𝑓 𝑥 =
1
1+𝑒−𝜆𝑥 (3)
• Tan hyperbolic: 𝑓 𝑥 =
𝑒 𝜆𝑥−𝑒−𝜆𝑥
𝑒 𝜆𝑥+𝑒−𝜆𝑥 (4)
Algorithm
The following steps are performed given below:
1. Take input of original image 𝐼(𝑥, 𝑦).
2. Normalize size of image 𝐼(𝑥, 𝑦) into 𝑛 𝘹 𝑛 blocks.
3. Normalize image 𝐼(𝑥, 𝑦) by dividing with 255.
4. Take input blurred image 𝐵(𝑥, 𝑦) and perform Step 2 and Step 3.
5. Reshape input and blurred image from 2-D to 1-D.
6. Create a new feed-forward neural network 20 hidden layer and one output layer.
The tangent sigmoidal and uniform activation function is associated with each
hidden and output layer respectively.
7. Randomly initializes weight of network to train network by back propagation
algorithm and set parameter for stop the training.
Continue…
8. Apply blurred image to trained network to get restored image.
9. Reshape image data from 1-D to 2-D.
10. Finally get restored image.
Start
Original image
Atmospheric
degraded images
Training neural
network
Simulate network
Restored Image
End Figure 3: The flow chart diagram and gui of proposed method
Yes
No
Start
Pre-processing
of image
Convert image
into 1D
Initialize weights
of network
Input blurred
image
Compute error
Error <=
acceptabl
e error
Weight update
Stop training and
save weights
Figure 4 : Flow chart of BPNN
Back
Continue…
Now the training phase of network using back propagation algorithm discuss
in figure 5, in which parameter is set for training, we have set 300 iteration
and 90 validation check. In this algorithm, the weights are initializing with some
small random values
Figure 5: Training process of proposed methodology
Result analysis
• The analysis of result obtain by proposed methodology is done by comparing the
result of other existing methods. On the basis of two parameters, MSE and PSNR.
• In the restoration, the imperceptibility or the quality of the image is measured by
using Peak Signal to Noise Ratio (PSNR). PSNR is used to calculate the similarity
in the original image and the blurred image. The PSNR is calculated by using two
images one is the original image and other is the restored image. The higher PSNR
means better the quality of the restored image. The basic formula of PSNR is given
below:
𝑃𝑆𝑁𝑅 = 10𝑙𝑜𝑔10
2552
𝑀𝑆𝐸
(5)
Where, 𝑀𝑆𝐸 the mean square error between two images.
Continue…
Mean Square Error
• The MSE is used to measures average squared disparities between the original
image and the restored image.
𝑀𝑆𝐸 =
1
𝑀×𝑁 𝑖=0
𝑀−1
𝑗=0
𝑁−1
𝑥𝑖,𝑗 − 𝑥𝑖,𝑗
2
(6)
Where, 𝑥𝑖,𝑗 and 𝑥𝑖,𝑗are the gray scale values of original and restored image
𝑀 × 𝑁 is the size of image
• Firstly the MSE (Mean Square Error) will be calculated then the PSNR value is
calculated. Therefore, higher value of PSNR denotes less distortion.
Restored Images
Example of Satellite and moon image:
Figure 6 (a) original image (b) blurred satellite
image
(c) Restored image using proposed
methodology
Figure 7 (a) original image (b) blurred moon (c) Restored image using proposed
image methodology
Continue…
Example of foggy image:
Figure 8 (a) original image (b) blurred foggy (c) Restored image using proposed
image methodology
Comparison of result
S.No Name of Image PSNR Value
of blurred
image
PSNR value
of restored
image using
kurtosis[10]
PSNR value
of restored
image using
PCA[13]
PSNR value
of restored
image using
proposed
methodology
1 Foggy image 58.81 59.54 63.54 69.47
2 Moon image 69.20 71.44 72.43 74.72
3 Satellite image 76.0207 78.89 79.69 82.2344
This table shows result compare with the PSNR value of restored image using
our proposed methodology and other existing techniques
Continue…
Figure 9: The bar graph of comparison PSNR value of restored image.
This figure shows graph comparison of PSNR value between Kurtosis, PCA and
Proposed method. These methods are used to restore image and calculate PSNR
value.
Continue…
S.No Name of Image PSNR
Value of
blurred
image
PSNR value
of restored
image using
kurtosis[10]
PSNR value
of restored
image using
PCA[13]
PSNR value
of restored
image using
proposed
methodology
1 Foggy image 56.91 58.89 61.24 67.57
2 Moon image 60.47 66.84 70.32 73.0934
3 Satellite image 59.578 65.81 71.79 77.154
This table shows result compare with the PSNR value of restored noisy
image using our proposed methodology and other existing techniques
Continue…
This graph comparison of restored noisy image PSNR value between
Kurtosis, PCA and Proposed method. These methods are used to restore
image which containing impulse noise and calculate PSNR value.
Figure 10: The bar graph of comparison PSNR value of restored noisy image.
Conclusion
• Atmospheric restoration technique is depth field in which a large scope to do
research work and have to increase efficiency of existing algorithm. The proposed
method is restoring atmospheric degraded image using neural network. This method
is designed according our analysis on image in low contrast, defogging techniques
improvement and also robust in presence of noise.
• The results obtained by our methodology in terms of PSNR values are acceptable
and better than other existing techniques. This method is robust towards noise
present in atmospheric degraded image.
Future Scope
Some new directions of research in the field of image restoration are not yet
fully explored.
1. Median filter can be used in pre-processing of image.
2. First order features extracted from image and it can be used to assign
weights of network for training purpose.
References
[1] M.R. Banham and A. K. Katsaggelos, “Digital Image Restoration”, IEEE Signal Processing Magazine, vol. 14,no.2, pp. 24-41,
1997.
[2] R.E. Hufnagel and N.R. Stanley, “ Modulation transfer function associated with image through turbulence media,” J. opt. Soc.
Amer: A, Opt. Image Sci., vol. 54, no. 1, pp. 52-61, 1964.
[3] M. K. Hu, “Visual pattern recognition by moment invariants,” IRE Trans. Inf. Theory, vol. 8, no. 2, pp. 179–187, Feb. 1962.
[4] Wu H.R. and Chen T. “Adaptive Impulse Detection Using Center Weighted Median Filters” IEEE signal processing letters,
Vol. 8, No. 1, January 2001.
[5] D. Li, S. Simske , and R.M. Mersereau, “ Blind Image deconvolution using constrained variance maximization,” in proc.
Asilomar Conf. Signals, Syst., Comput., 2004, pp. 1762-1765.
[6] R. Gregory, “A technique of minimizing the effects of atmospheric disturbance on photographic telescopes,” IEEE transaction
Nature, vol. 203,1964.
[7] J.L. Davidson and F. Hummer, “Morphological neural networks: An introduction with applications,” IEEE Systems Signal
Processing, Vol., 12, No.2, pp. 177-210, 1993.
[8] David L Donoho, “De-noising by soft thresholding”, IEEE Transactions on Information Theory, Vol. 41, No. 3, pp. 613–627,
May 2002.
[9] O. Rockinger, “Image sequence fusion using a shift-invariant wavelet transforms”, In Proceedings of the IEEE International
Conference on Image Processing, volume 3, pages 288–291, 1997.
[10] P. Baldi, J. Hornik, “Neural networks and principal component analysis: learning from examples without local minima”,
Neural Networks 2 (1), pp. 53-58, 1989.
[11] Y.T.Zhou, R. Chellappa and B.K. Jenkins, “Image restoration using a neural network,” IEEE Trans. Acoust., Speech, Signal
Processing, Vol., ASSP-36, pp. 1141-1151, July 1988.
[12] Dalong Li and Steven Simske, “Atmospheric Turbulence Degraded Image by Kurtosis Minimization,” IEEE Geoscience and
Remote Sensing Letters, Dec. 2008.
[13] Luxin Yan, Mingzhi Jin, Houzhang Fang, Hai Liu, and Tianxu Zhang, “Atmospheric-Turbulence-Degraded Astronomical
Image Restoration by Minimizing Second-Order Central moment,” IEEE Geoscience And Remote Sensing Letters, Vol. 9,
No. 4, pp. 672-676, July 2012.
[14] Dalong Li, Russell M. Mersereau, and Steven Simske, “Atmospheric Turbulence-Degraded Image Restoration Using
Principal Component Analysis,” IEEE Geoscience And Remote Sensing Letters, Vol. 4, No. 3, pp. 340-344, July 2012.
[15] Jin-Bao Wang, Ning He, Lu-Lu Zhang, and Ke Lu, “Single Image dehazing with a physical model and dark channel prior,”
Elsevier, neuro-computing Aug. 2014.
[16] K. Gibson and T. Nguyen, “Fast single image fog removal using the adaptive wiener filter,” in Proc. 20th IEEE ICIP, Sep.
2013, pp. 714–718.
[17] Nathaneera Anantrasirirchai, Alin Achim, Nick G.Kingsbury and David R. Bull, “Atmospheric Turbulence Mitigation Using
Complex Wavelet-Based Fusion,” IEEE Trans. Image processing, Vol. 22, no. 6, June 2013.
[18] X. Zhu and P. Milanfar, “Removing Atmospheric Turbulence via Space-Invariant Deconvolution,” in IEEE Transaction on
Pattern Analysis and Machine Intelligence, vol.35, no.1, January 2013.
Publications
• Azad Singh, Rajeev Kumar Singh, “A Survey: Image Restoration Techniques on
the basis of blurs”, AICTE Sponsored National Conference on Advances in
Information and Communication Technology (NCAICT), 2015.
• Azad Singh, Rajeev Kumar Singh, “Atmospheric Turbulence Degraded Image
Restoration Using Back Propagation Neural Network”, International Journal of
Signal Processing, Image Processing and Pattern Recognition, Volume 8, Issue 12,
pp. 181-194, ISSN: 2005-4254, 2015. (Published)
• Azad Singh, Rajeev Kumar Singh, “A Survey on Restoration Techniques of
Atmospheric Turbulence Blur Image”, in proceeding of IEEE, International
Conference on Innovations in Computer Science and Engineering, June 2015.
Thank you

Weitere ähnliche Inhalte

Was ist angesagt?

Image pre processing - local processing
Image pre processing - local processingImage pre processing - local processing
Image pre processing - local processingAshish Kumar
 
Image restoration yogesh 201410048
Image restoration yogesh 201410048Image restoration yogesh 201410048
Image restoration yogesh 201410048yogesh kumar
 
Removal of Gaussian noise on the image edges using the Prewitt operator and t...
Removal of Gaussian noise on the image edges using the Prewitt operator and t...Removal of Gaussian noise on the image edges using the Prewitt operator and t...
Removal of Gaussian noise on the image edges using the Prewitt operator and t...IOSR Journals
 
The application of image enhancement in color and grayscale images
The application of image enhancement in color and grayscale imagesThe application of image enhancement in color and grayscale images
The application of image enhancement in color and grayscale imagesNisar Ahmed Rana
 
Gradient-Based Low-Light Image Enhancement
Gradient-Based Low-Light Image EnhancementGradient-Based Low-Light Image Enhancement
Gradient-Based Low-Light Image EnhancementMasayuki Tanaka
 
Video Forgery Detection: Literature review
Video Forgery Detection: Literature reviewVideo Forgery Detection: Literature review
Video Forgery Detection: Literature reviewTharindu Rusira
 
Non-Blind Deblurring Using Partial Differential Equation Method
Non-Blind Deblurring Using Partial Differential Equation MethodNon-Blind Deblurring Using Partial Differential Equation Method
Non-Blind Deblurring Using Partial Differential Equation MethodEditor IJCATR
 
Ghost free image using blur and noise estimation
Ghost free image using blur and noise estimationGhost free image using blur and noise estimation
Ghost free image using blur and noise estimationijcga
 
Image segmentation ppt
Image segmentation pptImage segmentation ppt
Image segmentation pptGichelle Amon
 
Speckle noise reduction from medical ultrasound images using wavelet thresh
Speckle noise reduction from medical ultrasound images using wavelet threshSpeckle noise reduction from medical ultrasound images using wavelet thresh
Speckle noise reduction from medical ultrasound images using wavelet threshIAEME Publication
 
IMPORTANCE OF IMAGE ENHANCEMENT TECHNIQUES IN COLOR IMAGE SEGMENTATION: A COM...
IMPORTANCE OF IMAGE ENHANCEMENT TECHNIQUES IN COLOR IMAGE SEGMENTATION: A COM...IMPORTANCE OF IMAGE ENHANCEMENT TECHNIQUES IN COLOR IMAGE SEGMENTATION: A COM...
IMPORTANCE OF IMAGE ENHANCEMENT TECHNIQUES IN COLOR IMAGE SEGMENTATION: A COM...Dibya Jyoti Bora
 
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
 
The super resolution technology 2016
The super resolution technology 2016The super resolution technology 2016
The super resolution technology 2016Testo Viet Nam
 
Adaptive Noise Reduction Scheme for Salt and Pepper
Adaptive Noise Reduction Scheme for Salt and PepperAdaptive Noise Reduction Scheme for Salt and Pepper
Adaptive Noise Reduction Scheme for Salt and Peppersipij
 
Speckle Noise Reduction in Ultrasound Images using Adaptive and Anisotropic D...
Speckle Noise Reduction in Ultrasound Images using Adaptive and Anisotropic D...Speckle Noise Reduction in Ultrasound Images using Adaptive and Anisotropic D...
Speckle Noise Reduction in Ultrasound Images using Adaptive and Anisotropic D...Md. Shohel Rana
 

Was ist angesagt? (20)

Image pre processing - local processing
Image pre processing - local processingImage pre processing - local processing
Image pre processing - local processing
 
Image restoration yogesh 201410048
Image restoration yogesh 201410048Image restoration yogesh 201410048
Image restoration yogesh 201410048
 
IMAGE SEGMENTATION.
IMAGE SEGMENTATION.IMAGE SEGMENTATION.
IMAGE SEGMENTATION.
 
Removal of Gaussian noise on the image edges using the Prewitt operator and t...
Removal of Gaussian noise on the image edges using the Prewitt operator and t...Removal of Gaussian noise on the image edges using the Prewitt operator and t...
Removal of Gaussian noise on the image edges using the Prewitt operator and t...
 
Video Inpainting detection using inconsistencies in optical Flow
Video Inpainting detection using inconsistencies in optical FlowVideo Inpainting detection using inconsistencies in optical Flow
Video Inpainting detection using inconsistencies in optical Flow
 
The application of image enhancement in color and grayscale images
The application of image enhancement in color and grayscale imagesThe application of image enhancement in color and grayscale images
The application of image enhancement in color and grayscale images
 
Gradient-Based Low-Light Image Enhancement
Gradient-Based Low-Light Image EnhancementGradient-Based Low-Light Image Enhancement
Gradient-Based Low-Light Image Enhancement
 
Video Forgery Detection: Literature review
Video Forgery Detection: Literature reviewVideo Forgery Detection: Literature review
Video Forgery Detection: Literature review
 
Non-Blind Deblurring Using Partial Differential Equation Method
Non-Blind Deblurring Using Partial Differential Equation MethodNon-Blind Deblurring Using Partial Differential Equation Method
Non-Blind Deblurring Using Partial Differential Equation Method
 
Ghost free image using blur and noise estimation
Ghost free image using blur and noise estimationGhost free image using blur and noise estimation
Ghost free image using blur and noise estimation
 
Image segmentation ppt
Image segmentation pptImage segmentation ppt
Image segmentation ppt
 
Speckle noise reduction from medical ultrasound images using wavelet thresh
Speckle noise reduction from medical ultrasound images using wavelet threshSpeckle noise reduction from medical ultrasound images using wavelet thresh
Speckle noise reduction from medical ultrasound images using wavelet thresh
 
IMPORTANCE OF IMAGE ENHANCEMENT TECHNIQUES IN COLOR IMAGE SEGMENTATION: A COM...
IMPORTANCE OF IMAGE ENHANCEMENT TECHNIQUES IN COLOR IMAGE SEGMENTATION: A COM...IMPORTANCE OF IMAGE ENHANCEMENT TECHNIQUES IN COLOR IMAGE SEGMENTATION: A COM...
IMPORTANCE OF IMAGE ENHANCEMENT TECHNIQUES IN COLOR IMAGE SEGMENTATION: A COM...
 
Digital image processing - Image Enhancement (MATERIAL)
Digital image processing  - Image Enhancement (MATERIAL)Digital image processing  - Image Enhancement (MATERIAL)
Digital image processing - Image Enhancement (MATERIAL)
 
The super resolution technology 2016
The super resolution technology 2016The super resolution technology 2016
The super resolution technology 2016
 
Adaptive Noise Reduction Scheme for Salt and Pepper
Adaptive Noise Reduction Scheme for Salt and PepperAdaptive Noise Reduction Scheme for Salt and Pepper
Adaptive Noise Reduction Scheme for Salt and Pepper
 
Speckle Noise Reduction in Ultrasound Images using Adaptive and Anisotropic D...
Speckle Noise Reduction in Ultrasound Images using Adaptive and Anisotropic D...Speckle Noise Reduction in Ultrasound Images using Adaptive and Anisotropic D...
Speckle Noise Reduction in Ultrasound Images using Adaptive and Anisotropic D...
 
Seema dip
Seema dipSeema dip
Seema dip
 
G0443640
G0443640G0443640
G0443640
 
Image segmentation using wvlt trnsfrmtn and fuzzy logic. ppt
Image segmentation using wvlt trnsfrmtn and fuzzy logic. pptImage segmentation using wvlt trnsfrmtn and fuzzy logic. ppt
Image segmentation using wvlt trnsfrmtn and fuzzy logic. ppt
 

Ähnlich wie New microsoft power point presentation

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 ApproachesIRJET Journal
 
Progressive Image Denoising Through Hybrid Graph Laplacian Regularization: A ...
Progressive Image Denoising Through Hybrid Graph Laplacian Regularization: A ...Progressive Image Denoising Through Hybrid Graph Laplacian Regularization: A ...
Progressive Image Denoising Through Hybrid Graph Laplacian Regularization: A ...john236zaq
 
IRJET - Dehazing of Single Nighttime Haze Image using Superpixel Method
IRJET -  	  Dehazing of Single Nighttime Haze Image using Superpixel MethodIRJET -  	  Dehazing of Single Nighttime Haze Image using Superpixel Method
IRJET - Dehazing of Single Nighttime Haze Image using Superpixel MethodIRJET Journal
 
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 TechniqueBRNSSPublicationHubI
 
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
 
An efficient image segmentation approach through enhanced watershed algorithm
An efficient image segmentation approach through enhanced watershed algorithmAn efficient image segmentation approach through enhanced watershed algorithm
An efficient image segmentation approach through enhanced watershed algorithmAlexander Decker
 
NUMBER PLATE IMAGE DETECTION FOR FAST MOTION VEHICLES USING BLUR KERNEL ESTIM...
NUMBER PLATE IMAGE DETECTION FOR FAST MOTION VEHICLES USING BLUR KERNEL ESTIM...NUMBER PLATE IMAGE DETECTION FOR FAST MOTION VEHICLES USING BLUR KERNEL ESTIM...
NUMBER PLATE IMAGE DETECTION FOR FAST MOTION VEHICLES USING BLUR KERNEL ESTIM...paperpublications3
 
Image Denoising Based On Sparse Representation In A Probabilistic Framework
Image Denoising Based On Sparse Representation In A Probabilistic FrameworkImage Denoising Based On Sparse Representation In A Probabilistic Framework
Image Denoising Based On Sparse Representation In A Probabilistic FrameworkCSCJournals
 
A Pattern Classification Based approach for Blur Classification
A Pattern Classification Based approach for Blur ClassificationA Pattern Classification Based approach for Blur Classification
A Pattern Classification Based approach for Blur Classificationijeei-iaes
 
Review on Various Algorithm for Cloud Detection and Removal for Images
Review on Various Algorithm for Cloud Detection and Removal for ImagesReview on Various Algorithm for Cloud Detection and Removal for Images
Review on Various Algorithm for Cloud Detection and Removal for ImagesIJERA Editor
 
IRJET-A Review of Underwater Image Enhancement By Wavelet Decomposition using...
IRJET-A Review of Underwater Image Enhancement By Wavelet Decomposition using...IRJET-A Review of Underwater Image Enhancement By Wavelet Decomposition using...
IRJET-A Review of Underwater Image Enhancement By Wavelet Decomposition using...IRJET Journal
 
Atmospheric Correction of Remotely Sensed Images in Spatial and Transform Domain
Atmospheric Correction of Remotely Sensed Images in Spatial and Transform DomainAtmospheric Correction of Remotely Sensed Images in Spatial and Transform Domain
Atmospheric Correction of Remotely Sensed Images in Spatial and Transform DomainCSCJournals
 
Survey on Haze Removal Techniques
Survey on Haze Removal TechniquesSurvey on Haze Removal Techniques
Survey on Haze Removal TechniquesEditor IJMTER
 
A Review on Deformation Measurement from Speckle Patterns using Digital Image...
A Review on Deformation Measurement from Speckle Patterns using Digital Image...A Review on Deformation Measurement from Speckle Patterns using Digital Image...
A Review on Deformation Measurement from Speckle Patterns using Digital Image...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 AlgorithmIRJET Journal
 
Visibility Enhancement of Hazy Images using Depth Estimation Concept
Visibility Enhancement of Hazy Images using Depth Estimation ConceptVisibility Enhancement of Hazy Images using Depth Estimation Concept
Visibility Enhancement of Hazy Images using Depth Estimation ConceptIRJET Journal
 
Exponential contrast restoration in fog
Exponential contrast restoration in fogExponential contrast restoration in fog
Exponential contrast restoration in fogSREEKUTTY SREEKUMAR
 
A Novel Dehazing Method for Color Accuracy and Contrast Enhancement Method fo...
A Novel Dehazing Method for Color Accuracy and Contrast Enhancement Method fo...A Novel Dehazing Method for Color Accuracy and Contrast Enhancement Method fo...
A Novel Dehazing Method for Color Accuracy and Contrast Enhancement Method fo...IRJET Journal
 
Enhanced Vision of Hazy Images Using Improved Depth Estimation and Color Anal...
Enhanced Vision of Hazy Images Using Improved Depth Estimation and Color Anal...Enhanced Vision of Hazy Images Using Improved Depth Estimation and Color Anal...
Enhanced Vision of Hazy Images Using Improved Depth Estimation and Color Anal...IRJET Journal
 

Ähnlich wie New microsoft power point presentation (20)

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
 
Progressive Image Denoising Through Hybrid Graph Laplacian Regularization: A ...
Progressive Image Denoising Through Hybrid Graph Laplacian Regularization: A ...Progressive Image Denoising Through Hybrid Graph Laplacian Regularization: A ...
Progressive Image Denoising Through Hybrid Graph Laplacian Regularization: A ...
 
IRJET - Dehazing of Single Nighttime Haze Image using Superpixel Method
IRJET -  	  Dehazing of Single Nighttime Haze Image using Superpixel MethodIRJET -  	  Dehazing of Single Nighttime Haze Image using Superpixel Method
IRJET - Dehazing of Single Nighttime Haze Image using Superpixel Method
 
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
 
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
 
An efficient image segmentation approach through enhanced watershed algorithm
An efficient image segmentation approach through enhanced watershed algorithmAn efficient image segmentation approach through enhanced watershed algorithm
An efficient image segmentation approach through enhanced watershed algorithm
 
NUMBER PLATE IMAGE DETECTION FOR FAST MOTION VEHICLES USING BLUR KERNEL ESTIM...
NUMBER PLATE IMAGE DETECTION FOR FAST MOTION VEHICLES USING BLUR KERNEL ESTIM...NUMBER PLATE IMAGE DETECTION FOR FAST MOTION VEHICLES USING BLUR KERNEL ESTIM...
NUMBER PLATE IMAGE DETECTION FOR FAST MOTION VEHICLES USING BLUR KERNEL ESTIM...
 
H1802054851
H1802054851H1802054851
H1802054851
 
Image Denoising Based On Sparse Representation In A Probabilistic Framework
Image Denoising Based On Sparse Representation In A Probabilistic FrameworkImage Denoising Based On Sparse Representation In A Probabilistic Framework
Image Denoising Based On Sparse Representation In A Probabilistic Framework
 
A Pattern Classification Based approach for Blur Classification
A Pattern Classification Based approach for Blur ClassificationA Pattern Classification Based approach for Blur Classification
A Pattern Classification Based approach for Blur Classification
 
Review on Various Algorithm for Cloud Detection and Removal for Images
Review on Various Algorithm for Cloud Detection and Removal for ImagesReview on Various Algorithm for Cloud Detection and Removal for Images
Review on Various Algorithm for Cloud Detection and Removal for Images
 
IRJET-A Review of Underwater Image Enhancement By Wavelet Decomposition using...
IRJET-A Review of Underwater Image Enhancement By Wavelet Decomposition using...IRJET-A Review of Underwater Image Enhancement By Wavelet Decomposition using...
IRJET-A Review of Underwater Image Enhancement By Wavelet Decomposition using...
 
Atmospheric Correction of Remotely Sensed Images in Spatial and Transform Domain
Atmospheric Correction of Remotely Sensed Images in Spatial and Transform DomainAtmospheric Correction of Remotely Sensed Images in Spatial and Transform Domain
Atmospheric Correction of Remotely Sensed Images in Spatial and Transform Domain
 
Survey on Haze Removal Techniques
Survey on Haze Removal TechniquesSurvey on Haze Removal Techniques
Survey on Haze Removal Techniques
 
A Review on Deformation Measurement from Speckle Patterns using Digital Image...
A Review on Deformation Measurement from Speckle Patterns using Digital Image...A Review on Deformation Measurement from Speckle Patterns using Digital Image...
A Review on Deformation Measurement from Speckle Patterns using Digital Image...
 
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
 
Visibility Enhancement of Hazy Images using Depth Estimation Concept
Visibility Enhancement of Hazy Images using Depth Estimation ConceptVisibility Enhancement of Hazy Images using Depth Estimation Concept
Visibility Enhancement of Hazy Images using Depth Estimation Concept
 
Exponential contrast restoration in fog
Exponential contrast restoration in fogExponential contrast restoration in fog
Exponential contrast restoration in fog
 
A Novel Dehazing Method for Color Accuracy and Contrast Enhancement Method fo...
A Novel Dehazing Method for Color Accuracy and Contrast Enhancement Method fo...A Novel Dehazing Method for Color Accuracy and Contrast Enhancement Method fo...
A Novel Dehazing Method for Color Accuracy and Contrast Enhancement Method fo...
 
Enhanced Vision of Hazy Images Using Improved Depth Estimation and Color Anal...
Enhanced Vision of Hazy Images Using Improved Depth Estimation and Color Anal...Enhanced Vision of Hazy Images Using Improved Depth Estimation and Color Anal...
Enhanced Vision of Hazy Images Using Improved Depth Estimation and Color Anal...
 

KĂźrzlich hochgeladen

Introduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptxIntroduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptxupamatechverse
 
(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Service
(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Service(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Service
(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Serviceranjana rawat
 
(SHREYA) Chakan Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Esc...
(SHREYA) Chakan Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Esc...(SHREYA) Chakan Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Esc...
(SHREYA) Chakan Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Esc...ranjana rawat
 
(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...ranjana rawat
 
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service NashikCall Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service NashikCall Girls in Nagpur High Profile
 
Microscopic Analysis of Ceramic Materials.pptx
Microscopic Analysis of Ceramic Materials.pptxMicroscopic Analysis of Ceramic Materials.pptx
Microscopic Analysis of Ceramic Materials.pptxpurnimasatapathy1234
 
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICSAPPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICSKurinjimalarL3
 
Call Girls in Nagpur Suman Call 7001035870 Meet With Nagpur Escorts
Call Girls in Nagpur Suman Call 7001035870 Meet With Nagpur EscortsCall Girls in Nagpur Suman Call 7001035870 Meet With Nagpur Escorts
Call Girls in Nagpur Suman Call 7001035870 Meet With Nagpur EscortsCall Girls in Nagpur High Profile
 
Model Call Girl in Narela Delhi reach out to us at 🔝8264348440🔝
Model Call Girl in Narela Delhi reach out to us at 🔝8264348440🔝Model Call Girl in Narela Delhi reach out to us at 🔝8264348440🔝
Model Call Girl in Narela Delhi reach out to us at 🔝8264348440🔝soniya singh
 
Introduction to IEEE STANDARDS and its different types.pptx
Introduction to IEEE STANDARDS and its different types.pptxIntroduction to IEEE STANDARDS and its different types.pptx
Introduction to IEEE STANDARDS and its different types.pptxupamatechverse
 
IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...
IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...
IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...RajaP95
 
Coefficient of Thermal Expansion and their Importance.pptx
Coefficient of Thermal Expansion and their Importance.pptxCoefficient of Thermal Expansion and their Importance.pptx
Coefficient of Thermal Expansion and their Importance.pptxAsutosh Ranjan
 
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...Dr.Costas Sachpazis
 
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINEMANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINESIVASHANKAR N
 
(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...ranjana rawat
 
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur EscortsHigh Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escortsranjana rawat
 
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur EscortsCall Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur EscortsCall Girls in Nagpur High Profile
 
OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...
OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...
OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...Soham Mondal
 

KĂźrzlich hochgeladen (20)

Introduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptxIntroduction and different types of Ethernet.pptx
Introduction and different types of Ethernet.pptx
 
(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Service
(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Service(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Service
(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Service
 
(SHREYA) Chakan Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Esc...
(SHREYA) Chakan Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Esc...(SHREYA) Chakan Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Esc...
(SHREYA) Chakan Call Girls Just Call 7001035870 [ Cash on Delivery ] Pune Esc...
 
(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(ANVI) Koregaon Park Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
 
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service NashikCall Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
 
Microscopic Analysis of Ceramic Materials.pptx
Microscopic Analysis of Ceramic Materials.pptxMicroscopic Analysis of Ceramic Materials.pptx
Microscopic Analysis of Ceramic Materials.pptx
 
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICSAPPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
 
Call Girls in Nagpur Suman Call 7001035870 Meet With Nagpur Escorts
Call Girls in Nagpur Suman Call 7001035870 Meet With Nagpur EscortsCall Girls in Nagpur Suman Call 7001035870 Meet With Nagpur Escorts
Call Girls in Nagpur Suman Call 7001035870 Meet With Nagpur Escorts
 
Roadmap to Membership of RICS - Pathways and Routes
Roadmap to Membership of RICS - Pathways and RoutesRoadmap to Membership of RICS - Pathways and Routes
Roadmap to Membership of RICS - Pathways and Routes
 
Model Call Girl in Narela Delhi reach out to us at 🔝8264348440🔝
Model Call Girl in Narela Delhi reach out to us at 🔝8264348440🔝Model Call Girl in Narela Delhi reach out to us at 🔝8264348440🔝
Model Call Girl in Narela Delhi reach out to us at 🔝8264348440🔝
 
Introduction to IEEE STANDARDS and its different types.pptx
Introduction to IEEE STANDARDS and its different types.pptxIntroduction to IEEE STANDARDS and its different types.pptx
Introduction to IEEE STANDARDS and its different types.pptx
 
IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...
IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...
IMPLICATIONS OF THE ABOVE HOLISTIC UNDERSTANDING OF HARMONY ON PROFESSIONAL E...
 
Coefficient of Thermal Expansion and their Importance.pptx
Coefficient of Thermal Expansion and their Importance.pptxCoefficient of Thermal Expansion and their Importance.pptx
Coefficient of Thermal Expansion and their Importance.pptx
 
9953056974 Call Girls In South Ex, Escorts (Delhi) NCR.pdf
9953056974 Call Girls In South Ex, Escorts (Delhi) NCR.pdf9953056974 Call Girls In South Ex, Escorts (Delhi) NCR.pdf
9953056974 Call Girls In South Ex, Escorts (Delhi) NCR.pdf
 
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
Sheet Pile Wall Design and Construction: A Practical Guide for Civil Engineer...
 
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINEMANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
 
(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
(PRIYA) Rajgurunagar Call Girls Just Call 7001035870 [ Cash on Delivery ] Pun...
 
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur EscortsHigh Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
High Profile Call Girls Nagpur Isha Call 7001035870 Meet With Nagpur Escorts
 
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur EscortsCall Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
 
OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...
OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...
OSVC_Meta-Data based Simulation Automation to overcome Verification Challenge...
 

New microsoft power point presentation

  • 1. ATMOSPHERIC TURBULENCE DEGRADED IMAGE RESTORATION USING BACK PROPAGATION NEURAL NETWORK A DISSERTATION Presented In partial fulfillment of the requirement for the award of degree of MASTER OF TECHNOLOGY IN INFORMATION TECHNOLOGY Submitted by Azad Singh (0901IT13MT03) Under the supervision of Rajeev Kumar Singh (Assistant Professor) Department of Computer Science & Engineering and Information Technology Madhav Institute of Technology & Science, Gwalior (MP) - 474005 Session 2013-2015
  • 2. Contents • Introduction  Image restoration Atmospheric blur Artificial neural network • Literature Survey • Problem Statement • Proposed methodology • Result analysis • Conclusions and future work • References • Publications
  • 3. Introduction • Image restoration is a technique used to recover image from degraded image. Image may be distorted due to blur and noise; blur can occur due to atmospheric turbulence, motion of objects and camera mis-focus[1]. • Image Restoration is an area that also deals with improving the appearance of an image. • The degradation model of imagine system can be define as, 𝑔 𝑘 = 𝐼 𝑘⦻ℎ + 𝑛 𝑘 (1) Where, 𝑔 𝑘 is the degraded image, in which ⌝ denotes the two- dimensional linear convolution operation,𝐼 𝑘 is the original image and h is point spread function 𝑛 𝑘 is the additive noise.
  • 4. Continue… Image Restoration Techniques: i. Deterministic method:-In deterministic method [2] the prior knowledge about degradation of image so that we employ the method easily and get restored image. In deterministic method we restore image using filters. (1) Wiener filter[3] (2) Median filter[4] (3) Inverse filter[5] ii. Stochastic method: - The stochastic technique [6] uses the probability theory to form most likely result, given data. In stochastic method does not known prior knowledge of degradation source. So in this method firstly estimate the point spread function of blurred image. It is also called as blind de-convolution.
  • 5. Continue… • Atmospheric blur is the distortion of image due to long time exposure, fog, wind speed and due to randomly change in refractive index of air through which light travels. • The point spread function of atmospheric turbulence image can be described reasonably well by a Gaussian function[7]. ℎ 𝑥, 𝑦, 𝜎 𝐺 = 𝐶 𝑒𝑥𝑝 𝑥2+𝑦2 2𝜎2 𝐺 (2) Here, 𝜎 𝐺 determines the amount of spread of the blur, and the constant 𝐶 is to be chosen so that above equation is satisfied. • The estimation of point spread function of atmospheric turbulence image is very challenging without knowing prior knowledge of clear image[8]. So, it is very difficult to restore image using point spread function (PSF).
  • 6. Example of images • Here see images which is affected due to atmospheric blur given below: Figure 1: (a) image is blur (b) image is blur due to due to fog temperature i.e., mirage
  • 7. Continue… • The effects to atmospheric turbulence can be measured by calculating the scintillation index[9]. This index is related to the mean and standard deviation of the intensity distribution of image. • The atmospheric turbulence can be categorized in two categories:  In rigid bodies restoration, we take single frame image which is degraded through turbulence and restoration techniques will be applied in degraded image, which is known as single image deconvolution [10].  In non-rigid bodies, we take degraded videos and apply image registration technique to restore the object frame of video sequences, so uses the different registration algorithm for image registration.
  • 8. Artificial Neural Network • Artificial Neural Network [ANN] is an information paradigm which is inspired by our biological nervous system, like brain and processing system. In which artificial neurons are interconnected to each other through weights [11]. • It can be trained through the supervised learning process. An ANN is composed of a large number of highly interconnected processing elements neurons working in union to solve specific problem. • The simple neural network is a two layers or multilayer feed-forward and feedback network with 𝑀 input and 𝑁 output units. Each input unit is connected to each of the output units, and each connection is associated with weight, which are denoted as 𝑤1, 𝑤2 … 𝑤 𝑁 respectively or the strength of the connection
  • 9. Continue… . X 1 Z 1 Y 1 X i-n Z i-n Y i-n X n Z n Y n Figure 2: Architecture of feed-forward network A trained network can be thought as an “expert” in a particular category of information, if it has given to analyze. This is architecture of simple feed forward network Input layer Hidden layer Output layer
  • 10. Literature Survey • Dalong Li and Steven Simske [12] proposed kurtosis minimization method for restoration. Atmospheric turbulence is caused by the random fluctuations of the refraction index of the medium. It can lead to blurring in images acquired from a long distance away. Since the degradation is often not completely known, the problem is viewed as blind image deconvolution or blur identification. • Their work has observed that blurring increases kurtosis and introduced a new blur identification method based on kurtosis minimization[13].
  • 11. Continue… • Luxin Yan, Mingzhi Jin, Houzhang Fang, Hai Liu, and Tianxu Zhang [14] proposed Atmospheric turbulence affects imaging systems by virtue of wave propagation through a medium with a non-uniform index of refraction. It can lead to blurring in images acquired from a long distance away. In this letter, it is observed that blurring increases the second-order central moment (SOCM)of images, and we introduce a new parametric blur identification method by minimizing SOCM. • The method applies to finite-support images, in which the scene consists of a finite- extent object against a uniformly black, gray, or white background. The SOCM method has been validated by direct comparisons with other methods on simulated and real degraded images. • Dalong Li, Russell M. Mersereau, and Steven Simske [15] uses PCA revealed a connection between blind image deconvolution and principal components analysis (PCA).
  • 12. Continue… • Although PCA is derived from blur models that do not contain additive noise, it can be justified on both theoretical and experimental grounds that the PCA-based restoration algorithm is actually robust to the presence of white noise. • The algorithm is applied to the restoration of atmospheric turbulence-degraded imagery and compared to wiener filter [14] on both real and simulated atmospheric turbulence blurred images. It is shown that the PCA-based blind image deconvolution runs faster and is more robust to noise. PCA is applied on only shift invariant blur image and do not consider spatially varying blur image. • PCA based image deconvolution algorithm is fast, robust to noise and provide better result in multichannel image.
  • 13. Continue… • Jin-Bao Wang, Ning He, Lu-Lu Zhang, and Ke Lu [15] provide fog removal techniques using physical model and dark channel used for atmospheric scattering and optical reflecting imaging for estimation of atmospheric light in the dehazed image to get proper result. • It is mostly for non-sky patches, as at least one color channel has low intensity at some pixel, it can be happen due to shadows of objects [16]. As we know atmospheric images are usually full of shadows, so due to fog image is more contrast than its image without fog. • After apply the normalize operation we get restored image and improve the visibility through CLAHE (Contrast Limited Adaptive Histogram Equalisation). • This technique is useful for removing haze but provide better result in presence of different noise and other blur present in foggy image.
  • 14. Continue… • Anantrasirirchai, Achi,G.Kingsbury and David R. Bull revealed complex wavelet [17] for image fusion for restore the image. A new approach that overcomes the problem of ineffective occur due to conventional fusion methods for large distortion and also time consuming. • The conventional fusion methods require a large number of frames to select lucky regions. Image registration using the phase of a DTCWT coefficient is robust to noise and temporal intensity variations thereby providing an efficient tool for removing distorting ripples. After fusion, the effect of haze is reduced using locally- adaptive histogram equalization.
  • 15. Continue… • Zhu and Milanfar [18] explored the technique to restore atmospheric turbulence image and reduces the space and time-varying problem to shift invariant one. Thus method divides each frame into overlapping patches. These patches can be viewed in small region containing space-invariant blur. So blur can be deblurred using multi-frame blind deconvolution algorithm. In which calculate the PSF (Point Spread Function) for each patches for deconvolution.
  • 16. Problem Statement The atmospheric turbulence image restoration becomes more beneficial for numerous vision applications. The following gaps concluded using literature survey:- 1. In the past years, many researchers have provided techniques for restoration of fog or turbulence independently but not both and the restored image was not perfectly retrieved by traditional filter. 2. The existing methods have neglected to reduce the noise issue, which is presented in the output images. 3. The problem of the uneven illuminate in image is also neglected.
  • 17. Proposed methodology • In our proposed methodology, we create new feed-forward network which have 20 hidden layers and one output layer. The activation function is associated with each layer of hidden layer and output layer of network. At the hidden layer tangent function and output layer uniform sigmoidal function is used as shown in equation 3 and 4. The training of network is done by back propagation. The levenberg- Marquardt algorithm uses in back propagation and MSE for performance measure, perform number of iteration and validation check to get better result. • Unipolar sigmoid: 𝑓 𝑥 = 1 1+𝑒−𝜆𝑥 (3) • Tan hyperbolic: 𝑓 𝑥 = 𝑒 𝜆𝑥−𝑒−𝜆𝑥 𝑒 𝜆𝑥+𝑒−𝜆𝑥 (4)
  • 18. Algorithm The following steps are performed given below: 1. Take input of original image 𝐼(𝑥, 𝑦). 2. Normalize size of image 𝐼(𝑥, 𝑦) into 𝑛 𝘹 𝑛 blocks. 3. Normalize image 𝐼(𝑥, 𝑦) by dividing with 255. 4. Take input blurred image 𝐵(𝑥, 𝑦) and perform Step 2 and Step 3. 5. Reshape input and blurred image from 2-D to 1-D. 6. Create a new feed-forward neural network 20 hidden layer and one output layer. The tangent sigmoidal and uniform activation function is associated with each hidden and output layer respectively. 7. Randomly initializes weight of network to train network by back propagation algorithm and set parameter for stop the training.
  • 19. Continue… 8. Apply blurred image to trained network to get restored image. 9. Reshape image data from 1-D to 2-D. 10. Finally get restored image.
  • 20. Start Original image Atmospheric degraded images Training neural network Simulate network Restored Image End Figure 3: The flow chart diagram and gui of proposed method
  • 21. Yes No Start Pre-processing of image Convert image into 1D Initialize weights of network Input blurred image Compute error Error <= acceptabl e error Weight update Stop training and save weights Figure 4 : Flow chart of BPNN Back
  • 22. Continue… Now the training phase of network using back propagation algorithm discuss in figure 5, in which parameter is set for training, we have set 300 iteration and 90 validation check. In this algorithm, the weights are initializing with some small random values Figure 5: Training process of proposed methodology
  • 23. Result analysis • The analysis of result obtain by proposed methodology is done by comparing the result of other existing methods. On the basis of two parameters, MSE and PSNR. • In the restoration, the imperceptibility or the quality of the image is measured by using Peak Signal to Noise Ratio (PSNR). PSNR is used to calculate the similarity in the original image and the blurred image. The PSNR is calculated by using two images one is the original image and other is the restored image. The higher PSNR means better the quality of the restored image. The basic formula of PSNR is given below: 𝑃𝑆𝑁𝑅 = 10𝑙𝑜𝑔10 2552 𝑀𝑆𝐸 (5) Where, 𝑀𝑆𝐸 the mean square error between two images.
  • 24. Continue… Mean Square Error • The MSE is used to measures average squared disparities between the original image and the restored image. 𝑀𝑆𝐸 = 1 𝑀×𝑁 𝑖=0 𝑀−1 𝑗=0 𝑁−1 𝑥𝑖,𝑗 − 𝑥𝑖,𝑗 2 (6) Where, 𝑥𝑖,𝑗 and 𝑥𝑖,𝑗are the gray scale values of original and restored image 𝑀 × 𝑁 is the size of image • Firstly the MSE (Mean Square Error) will be calculated then the PSNR value is calculated. Therefore, higher value of PSNR denotes less distortion.
  • 25. Restored Images Example of Satellite and moon image: Figure 6 (a) original image (b) blurred satellite image (c) Restored image using proposed methodology Figure 7 (a) original image (b) blurred moon (c) Restored image using proposed image methodology
  • 26. Continue… Example of foggy image: Figure 8 (a) original image (b) blurred foggy (c) Restored image using proposed image methodology
  • 27. Comparison of result S.No Name of Image PSNR Value of blurred image PSNR value of restored image using kurtosis[10] PSNR value of restored image using PCA[13] PSNR value of restored image using proposed methodology 1 Foggy image 58.81 59.54 63.54 69.47 2 Moon image 69.20 71.44 72.43 74.72 3 Satellite image 76.0207 78.89 79.69 82.2344 This table shows result compare with the PSNR value of restored image using our proposed methodology and other existing techniques
  • 28. Continue… Figure 9: The bar graph of comparison PSNR value of restored image. This figure shows graph comparison of PSNR value between Kurtosis, PCA and Proposed method. These methods are used to restore image and calculate PSNR value.
  • 29. Continue… S.No Name of Image PSNR Value of blurred image PSNR value of restored image using kurtosis[10] PSNR value of restored image using PCA[13] PSNR value of restored image using proposed methodology 1 Foggy image 56.91 58.89 61.24 67.57 2 Moon image 60.47 66.84 70.32 73.0934 3 Satellite image 59.578 65.81 71.79 77.154 This table shows result compare with the PSNR value of restored noisy image using our proposed methodology and other existing techniques
  • 30. Continue… This graph comparison of restored noisy image PSNR value between Kurtosis, PCA and Proposed method. These methods are used to restore image which containing impulse noise and calculate PSNR value. Figure 10: The bar graph of comparison PSNR value of restored noisy image.
  • 31. Conclusion • Atmospheric restoration technique is depth field in which a large scope to do research work and have to increase efficiency of existing algorithm. The proposed method is restoring atmospheric degraded image using neural network. This method is designed according our analysis on image in low contrast, defogging techniques improvement and also robust in presence of noise. • The results obtained by our methodology in terms of PSNR values are acceptable and better than other existing techniques. This method is robust towards noise present in atmospheric degraded image.
  • 32. Future Scope Some new directions of research in the field of image restoration are not yet fully explored. 1. Median filter can be used in pre-processing of image. 2. First order features extracted from image and it can be used to assign weights of network for training purpose.
  • 33. References [1] M.R. Banham and A. K. Katsaggelos, “Digital Image Restoration”, IEEE Signal Processing Magazine, vol. 14,no.2, pp. 24-41, 1997. [2] R.E. Hufnagel and N.R. Stanley, “ Modulation transfer function associated with image through turbulence media,” J. opt. Soc. Amer: A, Opt. Image Sci., vol. 54, no. 1, pp. 52-61, 1964. [3] M. K. Hu, “Visual pattern recognition by moment invariants,” IRE Trans. Inf. Theory, vol. 8, no. 2, pp. 179–187, Feb. 1962. [4] Wu H.R. and Chen T. “Adaptive Impulse Detection Using Center Weighted Median Filters” IEEE signal processing letters, Vol. 8, No. 1, January 2001. [5] D. Li, S. Simske , and R.M. Mersereau, “ Blind Image deconvolution using constrained variance maximization,” in proc. Asilomar Conf. Signals, Syst., Comput., 2004, pp. 1762-1765. [6] R. Gregory, “A technique of minimizing the effects of atmospheric disturbance on photographic telescopes,” IEEE transaction Nature, vol. 203,1964. [7] J.L. Davidson and F. Hummer, “Morphological neural networks: An introduction with applications,” IEEE Systems Signal Processing, Vol., 12, No.2, pp. 177-210, 1993. [8] David L Donoho, “De-noising by soft thresholding”, IEEE Transactions on Information Theory, Vol. 41, No. 3, pp. 613–627, May 2002. [9] O. Rockinger, “Image sequence fusion using a shift-invariant wavelet transforms”, In Proceedings of the IEEE International Conference on Image Processing, volume 3, pages 288–291, 1997. [10] P. Baldi, J. Hornik, “Neural networks and principal component analysis: learning from examples without local minima”, Neural Networks 2 (1), pp. 53-58, 1989. [11] Y.T.Zhou, R. Chellappa and B.K. Jenkins, “Image restoration using a neural network,” IEEE Trans. Acoust., Speech, Signal Processing, Vol., ASSP-36, pp. 1141-1151, July 1988.
  • 34. [12] Dalong Li and Steven Simske, “Atmospheric Turbulence Degraded Image by Kurtosis Minimization,” IEEE Geoscience and Remote Sensing Letters, Dec. 2008. [13] Luxin Yan, Mingzhi Jin, Houzhang Fang, Hai Liu, and Tianxu Zhang, “Atmospheric-Turbulence-Degraded Astronomical Image Restoration by Minimizing Second-Order Central moment,” IEEE Geoscience And Remote Sensing Letters, Vol. 9, No. 4, pp. 672-676, July 2012. [14] Dalong Li, Russell M. Mersereau, and Steven Simske, “Atmospheric Turbulence-Degraded Image Restoration Using Principal Component Analysis,” IEEE Geoscience And Remote Sensing Letters, Vol. 4, No. 3, pp. 340-344, July 2012. [15] Jin-Bao Wang, Ning He, Lu-Lu Zhang, and Ke Lu, “Single Image dehazing with a physical model and dark channel prior,” Elsevier, neuro-computing Aug. 2014. [16] K. Gibson and T. Nguyen, “Fast single image fog removal using the adaptive wiener filter,” in Proc. 20th IEEE ICIP, Sep. 2013, pp. 714–718. [17] Nathaneera Anantrasirirchai, Alin Achim, Nick G.Kingsbury and David R. Bull, “Atmospheric Turbulence Mitigation Using Complex Wavelet-Based Fusion,” IEEE Trans. Image processing, Vol. 22, no. 6, June 2013. [18] X. Zhu and P. Milanfar, “Removing Atmospheric Turbulence via Space-Invariant Deconvolution,” in IEEE Transaction on Pattern Analysis and Machine Intelligence, vol.35, no.1, January 2013.
  • 35. Publications • Azad Singh, Rajeev Kumar Singh, “A Survey: Image Restoration Techniques on the basis of blurs”, AICTE Sponsored National Conference on Advances in Information and Communication Technology (NCAICT), 2015. • Azad Singh, Rajeev Kumar Singh, “Atmospheric Turbulence Degraded Image Restoration Using Back Propagation Neural Network”, International Journal of Signal Processing, Image Processing and Pattern Recognition, Volume 8, Issue 12, pp. 181-194, ISSN: 2005-4254, 2015. (Published) • Azad Singh, Rajeev Kumar Singh, “A Survey on Restoration Techniques of Atmospheric Turbulence Blur Image”, in proceeding of IEEE, International Conference on Innovations in Computer Science and Engineering, June 2015.