Professional Documents
Culture Documents
Komal Vij,Yaduvir Singh Int. J. Comp. Tech. Appl., Vol 2 (2), 309-313
Abstract- Image enhancement is a mean as the This is also known as contrast enhancement.
improvement of an image appearance by increasing Intensity transformation functions based on
dominance of some features or by decreasing
ambiguity between different regions of the image.
Image enhancement processes consist of a collection information extracted from image such as
of techniques that seek to improve the visual enhancement, compression, segmentation
appearance of an image or to convert the image to a and description.
form better suited for analysis by a human or
machine. Many images such as medical images, The Histogram of digital image with the
remote sensing images, electron microscopy images intensity levels in the range [0,L-1] is a
and even real life photographic pictures, suffer from discrete function.
poor contrast. Therefore it is necessary to enhance the
h(rk) = nk
contrast.The purpose of image enhancement methods
is to increase image visibility and details. Enhanced Where
image provide clear image to eyes or assist feature • rk is the intensity value.
extraction processing in computer vision system.
• nk is the number of pixels in the
Numerous enhancement methods have been
proposed but the enhancement efficiency,
image with intensity rk.
computational requirements, noise amplification, user • h(rk) is the histogram of the digital
intervention, and application suitability are the image with Gray Level rk
common factors to be considered when choosing Histograms are frequently normalized by the
from these different methods for specific image total number of pixels in the image.
processing application.
Assuming a M × N image, a normalized
histogram.
𝑛𝑛 𝑘𝑘
Keywords: Enhancement, Histogram processing p(rk) = 𝑀𝑀𝑀𝑀 , K=0,1,2,3……….L-1
techniques, PSNR,MSE.
is related to probability of occurrence of rk in
the image.
INTRODUCTION
I. Where
Histogram processing is the act of altering • p(rk) gives an estimate of the
an image by modifying its histogram. probability of occurrence of gray
Common uses of histogram processing level rk..
include normalization by which one makes
the histogram of an image as flat as possible.
309
ISSN:2229-6093
Komal Vij,Yaduvir Singh Int. J. Comp. Tech. Appl., Vol 2 (2), 309-313
310
ISSN:2229-6093
Komal Vij,Yaduvir Singh Int. J. Comp. Tech. Appl., Vol 2 (2), 309-313
when the message has uniform distribution set are mapped into the range from the mean
property. t the maximum gray level. Means one of the
sub image is equalized over the range up to
the mean and the other sub image is
B. Problems In Histogram Equalization equalized over the range. From the mean
based on the respective histograms .Thus,
the resulting equalized sub images are
1. The Histogram Equalization method bounded by each other around the input
does not take the mean brightness of mean ,which has an effect of preserving
an image into account. mean brightness.
2. The HE method may result in over In this the computation unit counts and store
enhancement and saturation artifacts
the respective number of occurrences 𝑛𝑛𝑘𝑘 for
due to the stretching of the gray
𝑘𝑘 = 0,1, … . . , 𝐿𝐿 − 1,the Histogram Splitter
levels over the full gray level range.
the splits the number of occurrences as
3. Histogram equalization can be
(𝑛𝑛0 , 𝑛𝑛1 , … , 𝑛𝑛𝑚𝑚 ) and (𝑛𝑛𝑚𝑚+1 , … . . , 𝑛𝑛𝐿𝐿−1 )
found on the fact that the brightness
respectively and where the mapper outputs
of an image can be changed after the
Y(i,j) as
histogram equalization.
4. Nevertheless, HE is not commonly 𝑋𝑋0 + (𝑋𝑋𝑚𝑚 − 𝑋𝑋0 )𝑐𝑐𝐿𝐿 (𝑥𝑥 )
used in consumer electronics such as 𝑌𝑌 (𝑖𝑖, 𝑗𝑗) = �
𝑋𝑋𝑚𝑚+1 + (𝑋𝑋𝐿𝐿−1 − 𝑋𝑋𝑚𝑚+1 )𝑐𝑐𝑈𝑈 (𝑥𝑥 )
TV because it may significantly
change the brightness of an input Which is based on 𝑓𝑓𝐿𝐿 (𝑋𝑋𝐿𝐿 ) ∪ 𝑓𝑓𝑈𝑈 (𝑋𝑋𝑈𝑈 ).The
image and cause undesirable computation of Histogram and the mean
artifacts. typically need to be done during one frame
5. It can be observed that the mean period;thus a frame memory to store the
brightness of the histogram- image being processed is neccessary
equalized image is always the middle
gray level regardless of the input D. Contrast Limited Adaptive Histogram
mean. Equalization (CLAHE)
311
ISSN:2229-6093
Komal Vij,Yaduvir Singh Int. J. Comp. Tech. Appl., Vol 2 (2), 309-313
2552
𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 = 10𝑙𝑙𝑙𝑙𝑙𝑙
𝑚𝑚𝑚𝑚𝑚𝑚
312
ISSN:2229-6093
Komal Vij,Yaduvir Singh Int. J. Comp. Tech. Appl., Vol 2 (2), 309-313
V. CONCLUSION
Histogram Processing is an effective method for
image enhancement; we tested four techniques on a
low contrast image. Among these, BBHE (Brightness
Preserving Bi-Histogram Equalization Technique)
has the lowest MSE and hightest PSNR and hence
gave best results.
REFERENCES
[1] S. D. Chen, A. Rahman Ramli, “Contrast Enhancement
using RecursiveMean-Separate Histogram Equalization
for Scalable Brightness Preservation,” IEEE
Transactions on Consumer Electronics, 49(4),pp.1301-
1309, 2003..
[2] Hyunsup Yoon, Youngjoon Han, and Hernsoo Hahn
“Image contrast Enhancement based Sub-histogram
Equalization Technique without over-equalization
Noise,” World academy of Science,Engineering and
Technology 50 2009.
[3] Y. T. Kim, "Contrast Enhancement Using Brightness
Preserving Bi-Histogram Equalization," IEEE
Transactions on Consumer Electronics, 43(1), pp.1-8,
1997.
[4] J. A. Stark, “Adaptive Image Contrast Enhancement
Using Generalizations of Histogram Equalization,”
IEEE Transactions on Image Processing, 9(5), pp.889-
896, 2000.
[5] J. Y. Kim, L. S. Kim, S. H. Hwang, “An Advanced
Contrast Enhancement Using Partially Overlapped Sub-
Block Histogram Equalization,” IEEE Transactions on
313