SlideShare una empresa de Scribd logo
1 de 6
Descargar para leer sin conexión
International Journal of Computational Engineering Research||Vol, 04||Issue, 3||
||Issn 2250-3005 || || March || 2014 || Page 58
Image Performance Tuning Using Contrast Enhancement
Method with Quality Evaluation
Alphy George1,
S John Livingston2
1
PG Student, Department of Computer Science and Engineering, Karunya University, Tamilnadu, India
2
Assistant Professor, Department of Computer Science and Engineering, Karunya University, Tamilnadu, India
I. INTRODUCTION
Image quality has an important role in many image processing applications mainly in the case of
recognition. Quality of an image can be degraded because of the presence of various distortions like blur,
contrast change, noise, blocking artifacts etc. A low quality or degraded image may not match with the existing
image in the database. So image quality measurement is important. In the above case if some preprocessing is
done to remove some distortions, then it will increase the image quality. A quality measurement method is the
used to measure the image quality.
For measuring the quality, image quality assessment (IQA) algorithms are used. These IQA methods
are classified into three, 1) no reference 2) reduced reference and 3) full reference [1]. In no reference method,
only distorted image will be available. No information about the original image is given. Because of this, blind
calculation of the image quality is possible. In the case of reduced reference method, a portion of the original
reference image will be available whereas, in full reference, a full original image will be given. The quality
metric can be calculated by comparing distorted image with the original image. In full reference approach, many
quality metrics are available for calculating the quality. Structural similarity index measure (SSIM) [1], visual
information fidelity (VIF) [2], visual signal to noise ratio (VSNR) [3] etc. is some of the full reference methods.
This paper investigating a full reference image quality assessment metric based on kolmogorov
complexity and normalized information distance (NID) [7]. These two concepts are rarely been studied in the
field of image processing. As a new method of introducing kolmogorov complexity in image quality assessment,
it shows comparable performance with the other existing methods.
One particular type of distortion usually present in images is contrast change. This may occur due to
the bad lighting condition or because of the camera problem. In this paper we combined a preprocessing method
for image contrast enhancement with the quality metric calculation. There are many methods available for
contrast enhancement. Here, contrast enhancement based on gamma correction and a weighting distribution
function is used. A normalization function is also applied to the enhanced image to improve the quality of the
image.
ABSTRACT:
Quality of an image plays a very crucial role in various image processing applications such as,
recognition, identification, transmission etc. Therefore identifying the quality of an image is very
necessary in such areas. Restoration of the good image from a degraded image will improve the
quality of the image. The purpose of this paper is to find the quality value of an image using a new
metric after some preprocessing steps. Here one particular type of image distortion taken into account
that is contrast change and enhancing the contrast using an additive gamma correction method. After
this preprocessing the quality value is found using a new image quality assessment metric called
normalized perceptual information distance. For this metric, the main concept used is kolmogorov
complexity and normalized information distance.
Keywords: Contrast enhancement, Image quality assessment, gamma correction, normalized
information distance, kolmogorov complexity.
Image Performance Tuning Using Contrast Enhancement…
||Issn 2250-3005 || || March || 2014 || Page 59
This paper is organized as follows. In section 2, contrast enhancement method using additive gamma
correction (ACG) [4] is explained. This contrast enhancement method is taken as a preprocessing step and
embedded to the final image quality assessment framework. In section 3, a full reference image quality
measurement based on kolmogorov complexity and normalized information distance [6] is explained. In section
4, the results after implementing this method are given. Finally, in section 5, conclusions are described.
II. CONTRAST ENHANCEMENT USING ACG
The problem with most of the contrast enhancement methods is their high computational time and cost.
The proposed approach is resolving this problem. In ACG method [4], the densely associated pixels in the input
dimmed image are smoothened by using a weighting distribution. The flow chart for ACG is given in Fig. 1.
Fig. 1 Flow chart of the AGC method
Firstly, a dimmed image is given as input to this method. Histogram analysis can be done to see the
densely oriented pixels. The gamma correction function used here is,
where gives the maximum intensity of the dimmed input image. represents the transformed
intensities of the dimmed image after gamma correction. This gamma correction method includes cdf
calculation. For cdf calculation, probability density function (pdf) has to be calculated. A weighting distribution
function is using here for pdf calculation. This will modify the histogram and decrease any adverse effects. The
weighting distribution function is,
where is the maximum pdf and is the minimum pdf. is the adjusting parameter. This can be
used to find cdf as follows,
sum of is calculated as follows,
(1)
(2)
(3)
(4)
Degraded input image
Histogram analysis
Weight distribution function
Gamma correction using calculation of cdf
Normalization of intensities
Enhanced image
Image Performance Tuning Using Contrast Enhancement…
||Issn 2250-3005 || || March || 2014 || Page 60
This can be applied in Eq. 1 to get the transformed intensities. While implementing this algorithm, first
the image is converted into HSV. Here H and S contain the color information. We are preserving the color
information and changing only value V. Sometimes the output enhanced image may be appeared as whitish after
gamma correction. In order to solve this problem, a normalization of the intensity levels of the output image is
used. Equation for that is given in Eq. (2.6).
where are original intensities of the input image. are the intensities produced after gamma correction. w
is the non negative weight. w is selected based on the improvement in the quality.
III. IMAGE QUALITY ASSESSMENT USING NPID
The problem with most of the contrast enhancement methods is their high computational time and cost.
The overall architecture of the proposed approach is given in Fig. 2.
Figure 2. Flow chart for the proposed system
Contrast enhancement method is described in section 2. For quality evaluation using NPID [6],
important mathematical concepts used are kolmogorov complexity and normalized information distance (NID).
These two concepts are rarely studied in the context of image processing. Kolmogorov complexity [8] is a non
computable concept. General equation for NID is,
where is the kolmogorov complexity of two objects x and y. For the implementation of NPID
metric, first transform the original and the reference image into the wavelet domain. The two images are
decomposed into subbands in that domain. Particular type of wavelet transform used here is 5 scale laplacian
pyramid transformation [11]. Then kolmogorov complexity can be formulated as,
where and are the subbands of reference image and the distorted image. Using this, NID can be calculated
as,
Because of the non computable nature of kolmogorov complexity, NID is also non computable. To
solve this problem, kolmogorov complexity can be replaced with Shannon entropy. Shannon entropy gives how
much information contained in an image. And also for finding information content, subband can be modeled
using a gaussian scale mixture model [9]. So the final formula for finding NPID image quality metric is,
Let are the N elements of subband , and , are defined in the same way.
Information content between the subbands can be calculated as [10],
(5)
(6)
(7)
(8)
(9)
(10)
Input dimmed
image
Contrast enhancement
using additive gamma
correction
Quality
evaluation
using NPID
Quality value
Enhanced
image
Image Performance Tuning Using Contrast Enhancement…
||Issn 2250-3005 || || March || 2014 || Page 61
To calculate the information content estimation of the parameters g, s, , are important. These parameters are
calculated based on the information from [6].
IV. RESULT ANALYSIS
The proposed algorithm is applied on the images in CSIQ (Categorial Image Quality) database [5]. This
database is publicly available for download. The database contains 866 distorted images for 30 reference
images. Different types of distortions include contrast change, jpeg compression, noise, blur etc. Each distortion
is at four to five distortion levels.
Figure. 3. Some of the reference images present in CSIQ database
Figure 4. Five levels of contrast change for child_swimming image in CSIQ database and its corresponding
contrast enhanced images.
For the proposed approach, contrast change distortion is taken into account. For each image, five levels
of contrast change is there in the database. For each level, contrast enhancement is done, and to the enhanced
image, the quality metric is applied. The result obtained after performing the contrast enhancement for some of
the images in the CSIQ database is given in Fig. 4.
(11)
(12)
Image Performance Tuning Using Contrast Enhancement…
||Issn 2250-3005 || || March || 2014 || Page 62
The quality metric values obtained for the enhanced images after performing the image quality
assessment algorithm based on normalized perceptual information distance (NPID) is given in Table 1.
Table 1. Results obtained before and after performing the algorithm to the images in CSIQ database
Image name 1st
level 2nd
level 3rd
level 4th
level 5th
level
before after before after before after before after before after
1600 0.9057 0.7675 0.8963 0.8037 0.8419 0.8454 0.8194 0.8477 0.8194 0.8473
Aerial_city 0.8886 0.7273 0.8488 0.7712 0.7495 0.8261 0.7083 0.8222 0.7083 0.8222
boston 0.8921 0.7043 0.8433 0.755 0.7084 0.8059 0.6277 0.7868 0.6277 0.7868
bridge 0.9115 0.6655 0.8948 0.6812 0.8582 0.7071 0.8398 0.7207 0.8398 0.7207
butter_flower 0.9014 0.7324 0.8653 0.7815 0.7245 0.7909 0.6913 0.7241 0.6913 0.7241
child_swimmin
g
0.8934 0.6295 0.8461 0.6854 0.7107 0.7567 0.646 0.7489 0.646 0.7489
couple 0.9201 0.6138 0.9013 0.6301 0.8568 0.6538 0.8377 0.6607 0.8377 0.6607
family 0.8824 0.7255 0.8476 0.7668 0.7739 0.8239 0.724 0.8275 0.724 0.8275
lake 0.9145 0.7257 0.8861 0.7673 0.8182 0.8253 0.7893 0.8363 0.7893 0.8363
log_seaside 0.9161 0.5775 0.8969 0.5946 0.8523 0.6202 0.835 0.6269 0.835 0.6269
monument 0.9154 0.651 0.8958 0.6729 0.8518 0.7116 0.8378 0.7216 0.8378 0.7216
native_america
n
0.9069 0.6664 0.8839 0.6863 0.8325 0.7156 0.8098 0.7229 0.8098 0.7229
redwood 0.9032 0.7592 0.8761 0.7904 0.8099 0.8351 0.7869 0.8413 0.7869 0.8413
roping 0.9031 0.7203 0.8765 0.7557 0.8172 0.8049 0.782 0.8147 0.782 0.8147
rushmore 0.9054 0.4891 0.8814 0.507 0.8282 0.5347 0.8064 0.543 0.8064 0.543
Results from the above figure shows that the contrast enhancement function works well if the image
quality lies in the range less than 0.8. If the quality value is greater than 0.8 then enhancing the image may
decrease the image quality.
V. CONCLUSIONS
In this paper, we combined the performance tuning of images with quality evaluation. Major two steps
performed in the proposed method is contrast enhancement using additive gamma correction and quality metric
calculation using normalized perceptual information distance. The algorithm is implemented on publicly
available CSIQ database and the results are obtained. If the quality is very poor, the contrast enhancement works
well and quality is improved in a good manner. If the image is of good quality, then no change is needed. This
work can be extended for finding the quality of video after performance tuning of video frames.
REFERENCES
[1]. Wang. Z., Bovik. A.C., Sheikh, H.R. and Simoncelli. E.P., “Image quality assessment: from error visibility to structural
similarity”, IEEE Trans. Image Process. 13(4), 600–612, 2011.
[2]. Sheikh. H.R. and Bovik. A.C., “ Image information and visual quality”, IEEE Trans. Image Process. 15(2), 430–444, (2006)
[3]. Chandler. D. M. and Hemami, S. S., “Vsnr: a wavelet-based visual signal-to-noise-ratio for natural images”, IEEE Trans. Image
Process. 16, 2284–2298, 2007.
[4]. Huang S., H.,Cheng F., Chiu Y., “Efficient contrast enhancement using adaptive gamma correction with weighting distribution”
IEEE Trans. Image Process. Vol.22, 2013.
[5]. Larson. E. C., Chandler. D.M.: Categorial image quality (CSIQ) database. (Online) Available http://vision.okstate.edu/csiq.
[6]. Nikwand M. and Wang Z., “Image distortion analysis based on normalized perceptual information distance”, Springer, SlViP
7:403-410, 2013.
[7]. Li. M., Chen. X., Li. X., Ma. B., and Vitanyi. P.M.B., “The similarity metric”, IEEE Trans. Inf. Theory 50,3250-3264, 2004.
[8]. Li. M., and Vitanyi. P., “An introduction to kolmogorov complexity and its applications”, 2nd
edn. Springer, Berlin, 1997.
[9]. Wainwright, M.J., Simoncelli, E.P.: Scale mixtures of gaussians and the statistics of natural images. In: Solla, S.A., Leen, T.K.,
Müller, K.-R. (eds.) Advances in Neural Information Processing Systems, vol. 12, pp. 855–861. MIT Press, Cambridge, 2000.
[10]. Wang, Z., Li, Q.: Information content weighting for perceptual image quality assessment. IEEE Trans Image Process. 20(5),
1185–1198 (2011)
[11]. Burt, P.J., Adelson, E.H.: The laplacian pyramid as a compact image code. IEEE Trans. Commun. 31, 532–540 (1983)
Image Performance Tuning Using Contrast Enhancement…
||Issn 2250-3005 || || March || 2014 || Page 63
BIOGRAPHIES
Alphy George received her B.Tech (Computer Science and Engineering) degree from SCMS school
of Engineering and Technology affiliated by Mahatma Gandhi University in the year 2012. She is
pursuing her M. Tech degree in Computer Science and Engineering at Karunya University.
S John Livingston received his M.E. degree from karunya university. He is currently working as an
Assistant Professor at Karunya University.

Más contenido relacionado

La actualidad más candente

Interpolation Technique using Non Linear Partial Differential Equation with E...
Interpolation Technique using Non Linear Partial Differential Equation with E...Interpolation Technique using Non Linear Partial Differential Equation with E...
Interpolation Technique using Non Linear Partial Differential Equation with E...CSCJournals
 
Intensity Preserving Cast Removal in Color Images Using Particle Swarm Optimi...
Intensity Preserving Cast Removal in Color Images Using Particle Swarm Optimi...Intensity Preserving Cast Removal in Color Images Using Particle Swarm Optimi...
Intensity Preserving Cast Removal in Color Images Using Particle Swarm Optimi...IJECEIAES
 
Efficient contrast enhancement using gamma correction with multilevel thresho...
Efficient contrast enhancement using gamma correction with multilevel thresho...Efficient contrast enhancement using gamma correction with multilevel thresho...
Efficient contrast enhancement using gamma correction with multilevel thresho...eSAT Publishing House
 
Property based fusion for multifocus images
Property based fusion for multifocus imagesProperty based fusion for multifocus images
Property based fusion for multifocus imagesIAEME Publication
 
Performance Evaluation of Filters for Enhancement of Images in Different Appl...
Performance Evaluation of Filters for Enhancement of Images in Different Appl...Performance Evaluation of Filters for Enhancement of Images in Different Appl...
Performance Evaluation of Filters for Enhancement of Images in Different Appl...IOSR Journals
 
Histogram Equalization with Range Offset for Brightness Preserved Image Enhan...
Histogram Equalization with Range Offset for Brightness Preserved Image Enhan...Histogram Equalization with Range Offset for Brightness Preserved Image Enhan...
Histogram Equalization with Range Offset for Brightness Preserved Image Enhan...CSCJournals
 
Review on Image Enhancement in Spatial Domain
Review on Image Enhancement in Spatial DomainReview on Image Enhancement in Spatial Domain
Review on Image Enhancement in Spatial Domainidescitation
 
CONTRAST ENHANCEMENT AND BRIGHTNESS PRESERVATION USING MULTIDECOMPOSITION HIS...
CONTRAST ENHANCEMENT AND BRIGHTNESS PRESERVATION USING MULTIDECOMPOSITION HIS...CONTRAST ENHANCEMENT AND BRIGHTNESS PRESERVATION USING MULTIDECOMPOSITION HIS...
CONTRAST ENHANCEMENT AND BRIGHTNESS PRESERVATION USING MULTIDECOMPOSITION HIS...sipij
 
Dataset Pre-Processing and Artificial Augmentation, Network Architecture and ...
Dataset Pre-Processing and Artificial Augmentation, Network Architecture and ...Dataset Pre-Processing and Artificial Augmentation, Network Architecture and ...
Dataset Pre-Processing and Artificial Augmentation, Network Architecture and ...INFOGAIN PUBLICATION
 
SEMANTIC IMAGE RETRIEVAL USING MULTIPLE FEATURES
SEMANTIC IMAGE RETRIEVAL USING MULTIPLE FEATURESSEMANTIC IMAGE RETRIEVAL USING MULTIPLE FEATURES
SEMANTIC IMAGE RETRIEVAL USING MULTIPLE FEATUREScscpconf
 
Extended Performance Appraise of Image Retrieval Using the Feature Vector as ...
Extended Performance Appraise of Image Retrieval Using the Feature Vector as ...Extended Performance Appraise of Image Retrieval Using the Feature Vector as ...
Extended Performance Appraise of Image Retrieval Using the Feature Vector as ...Waqas Tariq
 
International Journal of Engineering Research and Development (IJERD)
International Journal of Engineering Research and Development (IJERD)International Journal of Engineering Research and Development (IJERD)
International Journal of Engineering Research and Development (IJERD)IJERD Editor
 
Automatic Determination Number of Cluster for NMKFC-Means Algorithms on Image...
Automatic Determination Number of Cluster for NMKFC-Means Algorithms on Image...Automatic Determination Number of Cluster for NMKFC-Means Algorithms on Image...
Automatic Determination Number of Cluster for NMKFC-Means Algorithms on Image...IOSR Journals
 
An Efficient Approach for Image Enhancement Based on Image Fusion with Retine...
An Efficient Approach for Image Enhancement Based on Image Fusion with Retine...An Efficient Approach for Image Enhancement Based on Image Fusion with Retine...
An Efficient Approach for Image Enhancement Based on Image Fusion with Retine...ijsrd.com
 
CONTRAST ENHANCEMENT TECHNIQUES USING HISTOGRAM EQUALIZATION METHODS ON COLOR...
CONTRAST ENHANCEMENT TECHNIQUES USING HISTOGRAM EQUALIZATION METHODS ON COLOR...CONTRAST ENHANCEMENT TECHNIQUES USING HISTOGRAM EQUALIZATION METHODS ON COLOR...
CONTRAST ENHANCEMENT TECHNIQUES USING HISTOGRAM EQUALIZATION METHODS ON COLOR...IJCSEA Journal
 
Efficient Image Compression Technique using JPEG2000 with Adaptive Threshold
Efficient Image Compression Technique using JPEG2000 with Adaptive ThresholdEfficient Image Compression Technique using JPEG2000 with Adaptive Threshold
Efficient Image Compression Technique using JPEG2000 with Adaptive ThresholdCSCJournals
 
A comprehensive method for image contrast enhancement based on global –local ...
A comprehensive method for image contrast enhancement based on global –local ...A comprehensive method for image contrast enhancement based on global –local ...
A comprehensive method for image contrast enhancement based on global –local ...eSAT Publishing House
 

La actualidad más candente (19)

Interpolation Technique using Non Linear Partial Differential Equation with E...
Interpolation Technique using Non Linear Partial Differential Equation with E...Interpolation Technique using Non Linear Partial Differential Equation with E...
Interpolation Technique using Non Linear Partial Differential Equation with E...
 
Intensity Preserving Cast Removal in Color Images Using Particle Swarm Optimi...
Intensity Preserving Cast Removal in Color Images Using Particle Swarm Optimi...Intensity Preserving Cast Removal in Color Images Using Particle Swarm Optimi...
Intensity Preserving Cast Removal in Color Images Using Particle Swarm Optimi...
 
Efficient contrast enhancement using gamma correction with multilevel thresho...
Efficient contrast enhancement using gamma correction with multilevel thresho...Efficient contrast enhancement using gamma correction with multilevel thresho...
Efficient contrast enhancement using gamma correction with multilevel thresho...
 
Property based fusion for multifocus images
Property based fusion for multifocus imagesProperty based fusion for multifocus images
Property based fusion for multifocus images
 
Performance Evaluation of Filters for Enhancement of Images in Different Appl...
Performance Evaluation of Filters for Enhancement of Images in Different Appl...Performance Evaluation of Filters for Enhancement of Images in Different Appl...
Performance Evaluation of Filters for Enhancement of Images in Different Appl...
 
Histogram Equalization with Range Offset for Brightness Preserved Image Enhan...
Histogram Equalization with Range Offset for Brightness Preserved Image Enhan...Histogram Equalization with Range Offset for Brightness Preserved Image Enhan...
Histogram Equalization with Range Offset for Brightness Preserved Image Enhan...
 
Review on Image Enhancement in Spatial Domain
Review on Image Enhancement in Spatial DomainReview on Image Enhancement in Spatial Domain
Review on Image Enhancement in Spatial Domain
 
CONTRAST ENHANCEMENT AND BRIGHTNESS PRESERVATION USING MULTIDECOMPOSITION HIS...
CONTRAST ENHANCEMENT AND BRIGHTNESS PRESERVATION USING MULTIDECOMPOSITION HIS...CONTRAST ENHANCEMENT AND BRIGHTNESS PRESERVATION USING MULTIDECOMPOSITION HIS...
CONTRAST ENHANCEMENT AND BRIGHTNESS PRESERVATION USING MULTIDECOMPOSITION HIS...
 
Dataset Pre-Processing and Artificial Augmentation, Network Architecture and ...
Dataset Pre-Processing and Artificial Augmentation, Network Architecture and ...Dataset Pre-Processing and Artificial Augmentation, Network Architecture and ...
Dataset Pre-Processing and Artificial Augmentation, Network Architecture and ...
 
B42020710
B42020710B42020710
B42020710
 
SEMANTIC IMAGE RETRIEVAL USING MULTIPLE FEATURES
SEMANTIC IMAGE RETRIEVAL USING MULTIPLE FEATURESSEMANTIC IMAGE RETRIEVAL USING MULTIPLE FEATURES
SEMANTIC IMAGE RETRIEVAL USING MULTIPLE FEATURES
 
F0342032038
F0342032038F0342032038
F0342032038
 
Extended Performance Appraise of Image Retrieval Using the Feature Vector as ...
Extended Performance Appraise of Image Retrieval Using the Feature Vector as ...Extended Performance Appraise of Image Retrieval Using the Feature Vector as ...
Extended Performance Appraise of Image Retrieval Using the Feature Vector as ...
 
International Journal of Engineering Research and Development (IJERD)
International Journal of Engineering Research and Development (IJERD)International Journal of Engineering Research and Development (IJERD)
International Journal of Engineering Research and Development (IJERD)
 
Automatic Determination Number of Cluster for NMKFC-Means Algorithms on Image...
Automatic Determination Number of Cluster for NMKFC-Means Algorithms on Image...Automatic Determination Number of Cluster for NMKFC-Means Algorithms on Image...
Automatic Determination Number of Cluster for NMKFC-Means Algorithms on Image...
 
An Efficient Approach for Image Enhancement Based on Image Fusion with Retine...
An Efficient Approach for Image Enhancement Based on Image Fusion with Retine...An Efficient Approach for Image Enhancement Based on Image Fusion with Retine...
An Efficient Approach for Image Enhancement Based on Image Fusion with Retine...
 
CONTRAST ENHANCEMENT TECHNIQUES USING HISTOGRAM EQUALIZATION METHODS ON COLOR...
CONTRAST ENHANCEMENT TECHNIQUES USING HISTOGRAM EQUALIZATION METHODS ON COLOR...CONTRAST ENHANCEMENT TECHNIQUES USING HISTOGRAM EQUALIZATION METHODS ON COLOR...
CONTRAST ENHANCEMENT TECHNIQUES USING HISTOGRAM EQUALIZATION METHODS ON COLOR...
 
Efficient Image Compression Technique using JPEG2000 with Adaptive Threshold
Efficient Image Compression Technique using JPEG2000 with Adaptive ThresholdEfficient Image Compression Technique using JPEG2000 with Adaptive Threshold
Efficient Image Compression Technique using JPEG2000 with Adaptive Threshold
 
A comprehensive method for image contrast enhancement based on global –local ...
A comprehensive method for image contrast enhancement based on global –local ...A comprehensive method for image contrast enhancement based on global –local ...
A comprehensive method for image contrast enhancement based on global –local ...
 

Similar a G04302058063

A Comparative Study on Image Contrast Enhancement Techniques
A Comparative Study on Image Contrast Enhancement TechniquesA Comparative Study on Image Contrast Enhancement Techniques
A Comparative Study on Image Contrast Enhancement TechniquesIRJET Journal
 
An image enhancement method based on gabor filtering in wavelet domain and ad...
An image enhancement method based on gabor filtering in wavelet domain and ad...An image enhancement method based on gabor filtering in wavelet domain and ad...
An image enhancement method based on gabor filtering in wavelet domain and ad...nooriasukmaningtyas
 
IRJET- Image Enhancement using Various Discrete Wavelet Transformation Fi...
IRJET-  	  Image Enhancement using Various Discrete Wavelet Transformation Fi...IRJET-  	  Image Enhancement using Various Discrete Wavelet Transformation Fi...
IRJET- Image Enhancement using Various Discrete Wavelet Transformation Fi...IRJET Journal
 
Image Enhancement using Guided Filter for under Exposed Images
Image Enhancement using Guided Filter for under Exposed ImagesImage Enhancement using Guided Filter for under Exposed Images
Image Enhancement using Guided Filter for under Exposed ImagesDr. Amarjeet Singh
 
A review on image enhancement techniques
A review on image enhancement techniquesA review on image enhancement techniques
A review on image enhancement techniquesIJEACS
 
IRJET- Contrast Enhancement of Grey Level and Color Image using DWT and SVD
IRJET- Contrast Enhancement of Grey Level and Color Image using DWT and SVDIRJET- Contrast Enhancement of Grey Level and Color Image using DWT and SVD
IRJET- Contrast Enhancement of Grey Level and Color Image using DWT and SVDIRJET Journal
 
IRJET- Contrast Enhancement of Grey Level and Color Image using DWT and SVD
IRJET-  	  Contrast Enhancement of Grey Level and Color Image using DWT and SVDIRJET-  	  Contrast Enhancement of Grey Level and Color Image using DWT and SVD
IRJET- Contrast Enhancement of Grey Level and Color Image using DWT and SVDIRJET Journal
 
IRJET- A Comparative Review of Satellite Image Super Resolution Techniques
IRJET- A Comparative Review of Satellite Image Super Resolution TechniquesIRJET- A Comparative Review of Satellite Image Super Resolution Techniques
IRJET- A Comparative Review of Satellite Image Super Resolution TechniquesIRJET Journal
 
IRJET - Underwater Image Enhancement using PCNN and NSCT Fusion
IRJET -  	  Underwater Image Enhancement using PCNN and NSCT FusionIRJET -  	  Underwater Image Enhancement using PCNN and NSCT Fusion
IRJET - Underwater Image Enhancement using PCNN and NSCT FusionIRJET Journal
 
A Review Paper on Low Light Image Enhancement Methods for Un- Uniform Illumin...
A Review Paper on Low Light Image Enhancement Methods for Un- Uniform Illumin...A Review Paper on Low Light Image Enhancement Methods for Un- Uniform Illumin...
A Review Paper on Low Light Image Enhancement Methods for Un- Uniform Illumin...IRJET Journal
 
Mammographic Image Enhancement using Digital Image Processing Technique
Mammographic Image Enhancement using Digital Image Processing TechniqueMammographic Image Enhancement using Digital Image Processing Technique
Mammographic Image Enhancement using Digital Image Processing TechniqueIJCSIS Research Publications
 
Image enhancement
Image enhancement Image enhancement
Image enhancement SimiAttri
 
A REGULARIZED ROBUST SUPER-RESOLUTION APPROACH FORALIASED IMAGES AND LOW RESO...
A REGULARIZED ROBUST SUPER-RESOLUTION APPROACH FORALIASED IMAGES AND LOW RESO...A REGULARIZED ROBUST SUPER-RESOLUTION APPROACH FORALIASED IMAGES AND LOW RESO...
A REGULARIZED ROBUST SUPER-RESOLUTION APPROACH FORALIASED IMAGES AND LOW RESO...cscpconf
 
Ijarcet vol-2-issue-7-2319-2322
Ijarcet vol-2-issue-7-2319-2322Ijarcet vol-2-issue-7-2319-2322
Ijarcet vol-2-issue-7-2319-2322Editor IJARCET
 
Ijarcet vol-2-issue-7-2319-2322
Ijarcet vol-2-issue-7-2319-2322Ijarcet vol-2-issue-7-2319-2322
Ijarcet vol-2-issue-7-2319-2322Editor IJARCET
 
Qualitative and Quantitative Evaluation of Two New Histogram Limiting Binariz...
Qualitative and Quantitative Evaluation of Two New Histogram Limiting Binariz...Qualitative and Quantitative Evaluation of Two New Histogram Limiting Binariz...
Qualitative and Quantitative Evaluation of Two New Histogram Limiting Binariz...CSCJournals
 
Improved nonlocal means based on pre classification and invariant block matching
Improved nonlocal means based on pre classification and invariant block matchingImproved nonlocal means based on pre classification and invariant block matching
Improved nonlocal means based on pre classification and invariant block matchingIAEME Publication
 
Improved nonlocal means based on pre classification and invariant block matching
Improved nonlocal means based on pre classification and invariant block matchingImproved nonlocal means based on pre classification and invariant block matching
Improved nonlocal means based on pre classification and invariant block matchingIAEME Publication
 
An Inclusive Analysis on Various Image Enhancement Techniques
An Inclusive Analysis on Various Image Enhancement TechniquesAn Inclusive Analysis on Various Image Enhancement Techniques
An Inclusive Analysis on Various Image Enhancement TechniquesIJMER
 

Similar a G04302058063 (20)

A Comparative Study on Image Contrast Enhancement Techniques
A Comparative Study on Image Contrast Enhancement TechniquesA Comparative Study on Image Contrast Enhancement Techniques
A Comparative Study on Image Contrast Enhancement Techniques
 
Ijcatr04051016
Ijcatr04051016Ijcatr04051016
Ijcatr04051016
 
An image enhancement method based on gabor filtering in wavelet domain and ad...
An image enhancement method based on gabor filtering in wavelet domain and ad...An image enhancement method based on gabor filtering in wavelet domain and ad...
An image enhancement method based on gabor filtering in wavelet domain and ad...
 
IRJET- Image Enhancement using Various Discrete Wavelet Transformation Fi...
IRJET-  	  Image Enhancement using Various Discrete Wavelet Transformation Fi...IRJET-  	  Image Enhancement using Various Discrete Wavelet Transformation Fi...
IRJET- Image Enhancement using Various Discrete Wavelet Transformation Fi...
 
Image Enhancement using Guided Filter for under Exposed Images
Image Enhancement using Guided Filter for under Exposed ImagesImage Enhancement using Guided Filter for under Exposed Images
Image Enhancement using Guided Filter for under Exposed Images
 
A review on image enhancement techniques
A review on image enhancement techniquesA review on image enhancement techniques
A review on image enhancement techniques
 
IRJET- Contrast Enhancement of Grey Level and Color Image using DWT and SVD
IRJET- Contrast Enhancement of Grey Level and Color Image using DWT and SVDIRJET- Contrast Enhancement of Grey Level and Color Image using DWT and SVD
IRJET- Contrast Enhancement of Grey Level and Color Image using DWT and SVD
 
IRJET- Contrast Enhancement of Grey Level and Color Image using DWT and SVD
IRJET-  	  Contrast Enhancement of Grey Level and Color Image using DWT and SVDIRJET-  	  Contrast Enhancement of Grey Level and Color Image using DWT and SVD
IRJET- Contrast Enhancement of Grey Level and Color Image using DWT and SVD
 
IRJET- A Comparative Review of Satellite Image Super Resolution Techniques
IRJET- A Comparative Review of Satellite Image Super Resolution TechniquesIRJET- A Comparative Review of Satellite Image Super Resolution Techniques
IRJET- A Comparative Review of Satellite Image Super Resolution Techniques
 
IRJET - Underwater Image Enhancement using PCNN and NSCT Fusion
IRJET -  	  Underwater Image Enhancement using PCNN and NSCT FusionIRJET -  	  Underwater Image Enhancement using PCNN and NSCT Fusion
IRJET - Underwater Image Enhancement using PCNN and NSCT Fusion
 
A Review Paper on Low Light Image Enhancement Methods for Un- Uniform Illumin...
A Review Paper on Low Light Image Enhancement Methods for Un- Uniform Illumin...A Review Paper on Low Light Image Enhancement Methods for Un- Uniform Illumin...
A Review Paper on Low Light Image Enhancement Methods for Un- Uniform Illumin...
 
Mammographic Image Enhancement using Digital Image Processing Technique
Mammographic Image Enhancement using Digital Image Processing TechniqueMammographic Image Enhancement using Digital Image Processing Technique
Mammographic Image Enhancement using Digital Image Processing Technique
 
Image enhancement
Image enhancement Image enhancement
Image enhancement
 
A REGULARIZED ROBUST SUPER-RESOLUTION APPROACH FORALIASED IMAGES AND LOW RESO...
A REGULARIZED ROBUST SUPER-RESOLUTION APPROACH FORALIASED IMAGES AND LOW RESO...A REGULARIZED ROBUST SUPER-RESOLUTION APPROACH FORALIASED IMAGES AND LOW RESO...
A REGULARIZED ROBUST SUPER-RESOLUTION APPROACH FORALIASED IMAGES AND LOW RESO...
 
Ijarcet vol-2-issue-7-2319-2322
Ijarcet vol-2-issue-7-2319-2322Ijarcet vol-2-issue-7-2319-2322
Ijarcet vol-2-issue-7-2319-2322
 
Ijarcet vol-2-issue-7-2319-2322
Ijarcet vol-2-issue-7-2319-2322Ijarcet vol-2-issue-7-2319-2322
Ijarcet vol-2-issue-7-2319-2322
 
Qualitative and Quantitative Evaluation of Two New Histogram Limiting Binariz...
Qualitative and Quantitative Evaluation of Two New Histogram Limiting Binariz...Qualitative and Quantitative Evaluation of Two New Histogram Limiting Binariz...
Qualitative and Quantitative Evaluation of Two New Histogram Limiting Binariz...
 
Improved nonlocal means based on pre classification and invariant block matching
Improved nonlocal means based on pre classification and invariant block matchingImproved nonlocal means based on pre classification and invariant block matching
Improved nonlocal means based on pre classification and invariant block matching
 
Improved nonlocal means based on pre classification and invariant block matching
Improved nonlocal means based on pre classification and invariant block matchingImproved nonlocal means based on pre classification and invariant block matching
Improved nonlocal means based on pre classification and invariant block matching
 
An Inclusive Analysis on Various Image Enhancement Techniques
An Inclusive Analysis on Various Image Enhancement TechniquesAn Inclusive Analysis on Various Image Enhancement Techniques
An Inclusive Analysis on Various Image Enhancement Techniques
 

Último

ChatGPT-and-Generative-AI-Landscape Working of generative ai search
ChatGPT-and-Generative-AI-Landscape Working of generative ai searchChatGPT-and-Generative-AI-Landscape Working of generative ai search
ChatGPT-and-Generative-AI-Landscape Working of generative ai searchrohitcse52
 
Phase noise transfer functions.pptx
Phase noise transfer      functions.pptxPhase noise transfer      functions.pptx
Phase noise transfer functions.pptxSaiGouthamSunkara
 
Quasi-Stochastic Approximation: Algorithm Design Principles with Applications...
Quasi-Stochastic Approximation: Algorithm Design Principles with Applications...Quasi-Stochastic Approximation: Algorithm Design Principles with Applications...
Quasi-Stochastic Approximation: Algorithm Design Principles with Applications...Sean Meyn
 
me3493 manufacturing technology unit 1 Part A
me3493 manufacturing technology unit 1 Part Ame3493 manufacturing technology unit 1 Part A
me3493 manufacturing technology unit 1 Part Akarthi keyan
 
Modelling Guide for Timber Structures - FPInnovations
Modelling Guide for Timber Structures - FPInnovationsModelling Guide for Timber Structures - FPInnovations
Modelling Guide for Timber Structures - FPInnovationsYusuf Yıldız
 
IT3401-WEB ESSENTIALS PRESENTATIONS.pptx
IT3401-WEB ESSENTIALS PRESENTATIONS.pptxIT3401-WEB ESSENTIALS PRESENTATIONS.pptx
IT3401-WEB ESSENTIALS PRESENTATIONS.pptxSAJITHABANUS
 
Guardians and Glitches: Navigating the Duality of Gen AI in AppSec
Guardians and Glitches: Navigating the Duality of Gen AI in AppSecGuardians and Glitches: Navigating the Duality of Gen AI in AppSec
Guardians and Glitches: Navigating the Duality of Gen AI in AppSecTrupti Shiralkar, CISSP
 
Summer training report on BUILDING CONSTRUCTION for DIPLOMA Students.pdf
Summer training report on BUILDING CONSTRUCTION for DIPLOMA Students.pdfSummer training report on BUILDING CONSTRUCTION for DIPLOMA Students.pdf
Summer training report on BUILDING CONSTRUCTION for DIPLOMA Students.pdfNaveenVerma126
 
Basic Principle of Electrochemical Sensor
Basic Principle of  Electrochemical SensorBasic Principle of  Electrochemical Sensor
Basic Principle of Electrochemical SensorTanvir Moin
 
Power System electrical and electronics .pptx
Power System electrical and electronics .pptxPower System electrical and electronics .pptx
Power System electrical and electronics .pptxMUKULKUMAR210
 
Design of Clutches and Brakes in Design of Machine Elements.pptx
Design of Clutches and Brakes in Design of Machine Elements.pptxDesign of Clutches and Brakes in Design of Machine Elements.pptx
Design of Clutches and Brakes in Design of Machine Elements.pptxYogeshKumarKJMIT
 
Multicomponent Spiral Wound Membrane Separation Model.pdf
Multicomponent Spiral Wound Membrane Separation Model.pdfMulticomponent Spiral Wound Membrane Separation Model.pdf
Multicomponent Spiral Wound Membrane Separation Model.pdfGiovanaGhasary1
 
Technology Features of Apollo HDD Machine, Its Technical Specification with C...
Technology Features of Apollo HDD Machine, Its Technical Specification with C...Technology Features of Apollo HDD Machine, Its Technical Specification with C...
Technology Features of Apollo HDD Machine, Its Technical Specification with C...Apollo Techno Industries Pvt Ltd
 
Lecture 1: Basics of trigonometry (surveying)
Lecture 1: Basics of trigonometry (surveying)Lecture 1: Basics of trigonometry (surveying)
Lecture 1: Basics of trigonometry (surveying)Bahzad5
 
Test of Significance of Large Samples for Mean = µ.pptx
Test of Significance of Large Samples for Mean = µ.pptxTest of Significance of Large Samples for Mean = µ.pptx
Test of Significance of Large Samples for Mean = µ.pptxHome
 

Último (20)

Lecture 4 .pdf
Lecture 4                              .pdfLecture 4                              .pdf
Lecture 4 .pdf
 
計劃趕得上變化
計劃趕得上變化計劃趕得上變化
計劃趕得上變化
 
ChatGPT-and-Generative-AI-Landscape Working of generative ai search
ChatGPT-and-Generative-AI-Landscape Working of generative ai searchChatGPT-and-Generative-AI-Landscape Working of generative ai search
ChatGPT-and-Generative-AI-Landscape Working of generative ai search
 
Phase noise transfer functions.pptx
Phase noise transfer      functions.pptxPhase noise transfer      functions.pptx
Phase noise transfer functions.pptx
 
Quasi-Stochastic Approximation: Algorithm Design Principles with Applications...
Quasi-Stochastic Approximation: Algorithm Design Principles with Applications...Quasi-Stochastic Approximation: Algorithm Design Principles with Applications...
Quasi-Stochastic Approximation: Algorithm Design Principles with Applications...
 
me3493 manufacturing technology unit 1 Part A
me3493 manufacturing technology unit 1 Part Ame3493 manufacturing technology unit 1 Part A
me3493 manufacturing technology unit 1 Part A
 
Modelling Guide for Timber Structures - FPInnovations
Modelling Guide for Timber Structures - FPInnovationsModelling Guide for Timber Structures - FPInnovations
Modelling Guide for Timber Structures - FPInnovations
 
IT3401-WEB ESSENTIALS PRESENTATIONS.pptx
IT3401-WEB ESSENTIALS PRESENTATIONS.pptxIT3401-WEB ESSENTIALS PRESENTATIONS.pptx
IT3401-WEB ESSENTIALS PRESENTATIONS.pptx
 
Lecture 2 .pptx
Lecture 2                            .pptxLecture 2                            .pptx
Lecture 2 .pptx
 
Présentation IIRB 2024 Chloe Dufrane.pdf
Présentation IIRB 2024 Chloe Dufrane.pdfPrésentation IIRB 2024 Chloe Dufrane.pdf
Présentation IIRB 2024 Chloe Dufrane.pdf
 
Guardians and Glitches: Navigating the Duality of Gen AI in AppSec
Guardians and Glitches: Navigating the Duality of Gen AI in AppSecGuardians and Glitches: Navigating the Duality of Gen AI in AppSec
Guardians and Glitches: Navigating the Duality of Gen AI in AppSec
 
Summer training report on BUILDING CONSTRUCTION for DIPLOMA Students.pdf
Summer training report on BUILDING CONSTRUCTION for DIPLOMA Students.pdfSummer training report on BUILDING CONSTRUCTION for DIPLOMA Students.pdf
Summer training report on BUILDING CONSTRUCTION for DIPLOMA Students.pdf
 
Basic Principle of Electrochemical Sensor
Basic Principle of  Electrochemical SensorBasic Principle of  Electrochemical Sensor
Basic Principle of Electrochemical Sensor
 
Power System electrical and electronics .pptx
Power System electrical and electronics .pptxPower System electrical and electronics .pptx
Power System electrical and electronics .pptx
 
Design of Clutches and Brakes in Design of Machine Elements.pptx
Design of Clutches and Brakes in Design of Machine Elements.pptxDesign of Clutches and Brakes in Design of Machine Elements.pptx
Design of Clutches and Brakes in Design of Machine Elements.pptx
 
Litature Review: Research Paper work for Engineering
Litature Review: Research Paper work for EngineeringLitature Review: Research Paper work for Engineering
Litature Review: Research Paper work for Engineering
 
Multicomponent Spiral Wound Membrane Separation Model.pdf
Multicomponent Spiral Wound Membrane Separation Model.pdfMulticomponent Spiral Wound Membrane Separation Model.pdf
Multicomponent Spiral Wound Membrane Separation Model.pdf
 
Technology Features of Apollo HDD Machine, Its Technical Specification with C...
Technology Features of Apollo HDD Machine, Its Technical Specification with C...Technology Features of Apollo HDD Machine, Its Technical Specification with C...
Technology Features of Apollo HDD Machine, Its Technical Specification with C...
 
Lecture 1: Basics of trigonometry (surveying)
Lecture 1: Basics of trigonometry (surveying)Lecture 1: Basics of trigonometry (surveying)
Lecture 1: Basics of trigonometry (surveying)
 
Test of Significance of Large Samples for Mean = µ.pptx
Test of Significance of Large Samples for Mean = µ.pptxTest of Significance of Large Samples for Mean = µ.pptx
Test of Significance of Large Samples for Mean = µ.pptx
 

G04302058063

  • 1. International Journal of Computational Engineering Research||Vol, 04||Issue, 3|| ||Issn 2250-3005 || || March || 2014 || Page 58 Image Performance Tuning Using Contrast Enhancement Method with Quality Evaluation Alphy George1, S John Livingston2 1 PG Student, Department of Computer Science and Engineering, Karunya University, Tamilnadu, India 2 Assistant Professor, Department of Computer Science and Engineering, Karunya University, Tamilnadu, India I. INTRODUCTION Image quality has an important role in many image processing applications mainly in the case of recognition. Quality of an image can be degraded because of the presence of various distortions like blur, contrast change, noise, blocking artifacts etc. A low quality or degraded image may not match with the existing image in the database. So image quality measurement is important. In the above case if some preprocessing is done to remove some distortions, then it will increase the image quality. A quality measurement method is the used to measure the image quality. For measuring the quality, image quality assessment (IQA) algorithms are used. These IQA methods are classified into three, 1) no reference 2) reduced reference and 3) full reference [1]. In no reference method, only distorted image will be available. No information about the original image is given. Because of this, blind calculation of the image quality is possible. In the case of reduced reference method, a portion of the original reference image will be available whereas, in full reference, a full original image will be given. The quality metric can be calculated by comparing distorted image with the original image. In full reference approach, many quality metrics are available for calculating the quality. Structural similarity index measure (SSIM) [1], visual information fidelity (VIF) [2], visual signal to noise ratio (VSNR) [3] etc. is some of the full reference methods. This paper investigating a full reference image quality assessment metric based on kolmogorov complexity and normalized information distance (NID) [7]. These two concepts are rarely been studied in the field of image processing. As a new method of introducing kolmogorov complexity in image quality assessment, it shows comparable performance with the other existing methods. One particular type of distortion usually present in images is contrast change. This may occur due to the bad lighting condition or because of the camera problem. In this paper we combined a preprocessing method for image contrast enhancement with the quality metric calculation. There are many methods available for contrast enhancement. Here, contrast enhancement based on gamma correction and a weighting distribution function is used. A normalization function is also applied to the enhanced image to improve the quality of the image. ABSTRACT: Quality of an image plays a very crucial role in various image processing applications such as, recognition, identification, transmission etc. Therefore identifying the quality of an image is very necessary in such areas. Restoration of the good image from a degraded image will improve the quality of the image. The purpose of this paper is to find the quality value of an image using a new metric after some preprocessing steps. Here one particular type of image distortion taken into account that is contrast change and enhancing the contrast using an additive gamma correction method. After this preprocessing the quality value is found using a new image quality assessment metric called normalized perceptual information distance. For this metric, the main concept used is kolmogorov complexity and normalized information distance. Keywords: Contrast enhancement, Image quality assessment, gamma correction, normalized information distance, kolmogorov complexity.
  • 2. Image Performance Tuning Using Contrast Enhancement… ||Issn 2250-3005 || || March || 2014 || Page 59 This paper is organized as follows. In section 2, contrast enhancement method using additive gamma correction (ACG) [4] is explained. This contrast enhancement method is taken as a preprocessing step and embedded to the final image quality assessment framework. In section 3, a full reference image quality measurement based on kolmogorov complexity and normalized information distance [6] is explained. In section 4, the results after implementing this method are given. Finally, in section 5, conclusions are described. II. CONTRAST ENHANCEMENT USING ACG The problem with most of the contrast enhancement methods is their high computational time and cost. The proposed approach is resolving this problem. In ACG method [4], the densely associated pixels in the input dimmed image are smoothened by using a weighting distribution. The flow chart for ACG is given in Fig. 1. Fig. 1 Flow chart of the AGC method Firstly, a dimmed image is given as input to this method. Histogram analysis can be done to see the densely oriented pixels. The gamma correction function used here is, where gives the maximum intensity of the dimmed input image. represents the transformed intensities of the dimmed image after gamma correction. This gamma correction method includes cdf calculation. For cdf calculation, probability density function (pdf) has to be calculated. A weighting distribution function is using here for pdf calculation. This will modify the histogram and decrease any adverse effects. The weighting distribution function is, where is the maximum pdf and is the minimum pdf. is the adjusting parameter. This can be used to find cdf as follows, sum of is calculated as follows, (1) (2) (3) (4) Degraded input image Histogram analysis Weight distribution function Gamma correction using calculation of cdf Normalization of intensities Enhanced image
  • 3. Image Performance Tuning Using Contrast Enhancement… ||Issn 2250-3005 || || March || 2014 || Page 60 This can be applied in Eq. 1 to get the transformed intensities. While implementing this algorithm, first the image is converted into HSV. Here H and S contain the color information. We are preserving the color information and changing only value V. Sometimes the output enhanced image may be appeared as whitish after gamma correction. In order to solve this problem, a normalization of the intensity levels of the output image is used. Equation for that is given in Eq. (2.6). where are original intensities of the input image. are the intensities produced after gamma correction. w is the non negative weight. w is selected based on the improvement in the quality. III. IMAGE QUALITY ASSESSMENT USING NPID The problem with most of the contrast enhancement methods is their high computational time and cost. The overall architecture of the proposed approach is given in Fig. 2. Figure 2. Flow chart for the proposed system Contrast enhancement method is described in section 2. For quality evaluation using NPID [6], important mathematical concepts used are kolmogorov complexity and normalized information distance (NID). These two concepts are rarely studied in the context of image processing. Kolmogorov complexity [8] is a non computable concept. General equation for NID is, where is the kolmogorov complexity of two objects x and y. For the implementation of NPID metric, first transform the original and the reference image into the wavelet domain. The two images are decomposed into subbands in that domain. Particular type of wavelet transform used here is 5 scale laplacian pyramid transformation [11]. Then kolmogorov complexity can be formulated as, where and are the subbands of reference image and the distorted image. Using this, NID can be calculated as, Because of the non computable nature of kolmogorov complexity, NID is also non computable. To solve this problem, kolmogorov complexity can be replaced with Shannon entropy. Shannon entropy gives how much information contained in an image. And also for finding information content, subband can be modeled using a gaussian scale mixture model [9]. So the final formula for finding NPID image quality metric is, Let are the N elements of subband , and , are defined in the same way. Information content between the subbands can be calculated as [10], (5) (6) (7) (8) (9) (10) Input dimmed image Contrast enhancement using additive gamma correction Quality evaluation using NPID Quality value Enhanced image
  • 4. Image Performance Tuning Using Contrast Enhancement… ||Issn 2250-3005 || || March || 2014 || Page 61 To calculate the information content estimation of the parameters g, s, , are important. These parameters are calculated based on the information from [6]. IV. RESULT ANALYSIS The proposed algorithm is applied on the images in CSIQ (Categorial Image Quality) database [5]. This database is publicly available for download. The database contains 866 distorted images for 30 reference images. Different types of distortions include contrast change, jpeg compression, noise, blur etc. Each distortion is at four to five distortion levels. Figure. 3. Some of the reference images present in CSIQ database Figure 4. Five levels of contrast change for child_swimming image in CSIQ database and its corresponding contrast enhanced images. For the proposed approach, contrast change distortion is taken into account. For each image, five levels of contrast change is there in the database. For each level, contrast enhancement is done, and to the enhanced image, the quality metric is applied. The result obtained after performing the contrast enhancement for some of the images in the CSIQ database is given in Fig. 4. (11) (12)
  • 5. Image Performance Tuning Using Contrast Enhancement… ||Issn 2250-3005 || || March || 2014 || Page 62 The quality metric values obtained for the enhanced images after performing the image quality assessment algorithm based on normalized perceptual information distance (NPID) is given in Table 1. Table 1. Results obtained before and after performing the algorithm to the images in CSIQ database Image name 1st level 2nd level 3rd level 4th level 5th level before after before after before after before after before after 1600 0.9057 0.7675 0.8963 0.8037 0.8419 0.8454 0.8194 0.8477 0.8194 0.8473 Aerial_city 0.8886 0.7273 0.8488 0.7712 0.7495 0.8261 0.7083 0.8222 0.7083 0.8222 boston 0.8921 0.7043 0.8433 0.755 0.7084 0.8059 0.6277 0.7868 0.6277 0.7868 bridge 0.9115 0.6655 0.8948 0.6812 0.8582 0.7071 0.8398 0.7207 0.8398 0.7207 butter_flower 0.9014 0.7324 0.8653 0.7815 0.7245 0.7909 0.6913 0.7241 0.6913 0.7241 child_swimmin g 0.8934 0.6295 0.8461 0.6854 0.7107 0.7567 0.646 0.7489 0.646 0.7489 couple 0.9201 0.6138 0.9013 0.6301 0.8568 0.6538 0.8377 0.6607 0.8377 0.6607 family 0.8824 0.7255 0.8476 0.7668 0.7739 0.8239 0.724 0.8275 0.724 0.8275 lake 0.9145 0.7257 0.8861 0.7673 0.8182 0.8253 0.7893 0.8363 0.7893 0.8363 log_seaside 0.9161 0.5775 0.8969 0.5946 0.8523 0.6202 0.835 0.6269 0.835 0.6269 monument 0.9154 0.651 0.8958 0.6729 0.8518 0.7116 0.8378 0.7216 0.8378 0.7216 native_america n 0.9069 0.6664 0.8839 0.6863 0.8325 0.7156 0.8098 0.7229 0.8098 0.7229 redwood 0.9032 0.7592 0.8761 0.7904 0.8099 0.8351 0.7869 0.8413 0.7869 0.8413 roping 0.9031 0.7203 0.8765 0.7557 0.8172 0.8049 0.782 0.8147 0.782 0.8147 rushmore 0.9054 0.4891 0.8814 0.507 0.8282 0.5347 0.8064 0.543 0.8064 0.543 Results from the above figure shows that the contrast enhancement function works well if the image quality lies in the range less than 0.8. If the quality value is greater than 0.8 then enhancing the image may decrease the image quality. V. CONCLUSIONS In this paper, we combined the performance tuning of images with quality evaluation. Major two steps performed in the proposed method is contrast enhancement using additive gamma correction and quality metric calculation using normalized perceptual information distance. The algorithm is implemented on publicly available CSIQ database and the results are obtained. If the quality is very poor, the contrast enhancement works well and quality is improved in a good manner. If the image is of good quality, then no change is needed. This work can be extended for finding the quality of video after performance tuning of video frames. REFERENCES [1]. Wang. Z., Bovik. A.C., Sheikh, H.R. and Simoncelli. E.P., “Image quality assessment: from error visibility to structural similarity”, IEEE Trans. Image Process. 13(4), 600–612, 2011. [2]. Sheikh. H.R. and Bovik. A.C., “ Image information and visual quality”, IEEE Trans. Image Process. 15(2), 430–444, (2006) [3]. Chandler. D. M. and Hemami, S. S., “Vsnr: a wavelet-based visual signal-to-noise-ratio for natural images”, IEEE Trans. Image Process. 16, 2284–2298, 2007. [4]. Huang S., H.,Cheng F., Chiu Y., “Efficient contrast enhancement using adaptive gamma correction with weighting distribution” IEEE Trans. Image Process. Vol.22, 2013. [5]. Larson. E. C., Chandler. D.M.: Categorial image quality (CSIQ) database. (Online) Available http://vision.okstate.edu/csiq. [6]. Nikwand M. and Wang Z., “Image distortion analysis based on normalized perceptual information distance”, Springer, SlViP 7:403-410, 2013. [7]. Li. M., Chen. X., Li. X., Ma. B., and Vitanyi. P.M.B., “The similarity metric”, IEEE Trans. Inf. Theory 50,3250-3264, 2004. [8]. Li. M., and Vitanyi. P., “An introduction to kolmogorov complexity and its applications”, 2nd edn. Springer, Berlin, 1997. [9]. Wainwright, M.J., Simoncelli, E.P.: Scale mixtures of gaussians and the statistics of natural images. In: Solla, S.A., Leen, T.K., Müller, K.-R. (eds.) Advances in Neural Information Processing Systems, vol. 12, pp. 855–861. MIT Press, Cambridge, 2000. [10]. Wang, Z., Li, Q.: Information content weighting for perceptual image quality assessment. IEEE Trans Image Process. 20(5), 1185–1198 (2011) [11]. Burt, P.J., Adelson, E.H.: The laplacian pyramid as a compact image code. IEEE Trans. Commun. 31, 532–540 (1983)
  • 6. Image Performance Tuning Using Contrast Enhancement… ||Issn 2250-3005 || || March || 2014 || Page 63 BIOGRAPHIES Alphy George received her B.Tech (Computer Science and Engineering) degree from SCMS school of Engineering and Technology affiliated by Mahatma Gandhi University in the year 2012. She is pursuing her M. Tech degree in Computer Science and Engineering at Karunya University. S John Livingston received his M.E. degree from karunya university. He is currently working as an Assistant Professor at Karunya University.