Contrast Enhancement Techniques: A Brief and Concise Review

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 04 Issue: 07 | July -2017 www.irjet.net p-ISSN: 2395-0072
© 2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 1767
Contrast Enhancement Techniques: A Brief and Concise Review
Nikhil Verma1, Maitreyee Dutta2
1M.E. Scholar, Dept. of Electronics & Communication Engineering, NITTTR, Chandigarh, India
2Professor, Dept. ofComputer Science Engineering, NITTTR, Chandigarh, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The contrast enhancement is the most crucial
criterion to judge the quality of an image. The contrast levels
are needed to bring out the vital information that can be
extracted through an image. The contrast enhancement
becomes more significant when the images are affected with
the low illumination. The brightness levels, lightening
problems are addressed by the contrast enhancement
techniques. The subjective and objective image qualities of an
image are mainly accomplished by means of applying the
various contrast enhancement techniques. In this paper, a
number of these techniques are been discussedaccordingtoits
usage in various applications. This study will not only discuss
various advantages and disadvantages of different contrast
enhancement techniques but also demonstrate the different
aspects that are to be studied for analysis in an image.
Key Words: Contrast Enhancement, Histogram
Equalization, Image Processing, Illumination.
1. INTRODUCTION
The digital image processing is the field in which different
algorithms are been applied to perform the digital
processing on the different set of images. The digital image
processing is characterized by a number of different
methods namely, enhancement,detection,classification.The
image enhancement is the basis for the examination of the
image in all phases. The image enhancement is required for
improving the visual appearance or to obtain an improved
and transformed illustration of a particular image. The
processing of an image is defined by the domain analysis.
The domains that are widely used are frequency and spatial
domain. The spatial domain is the reference to the plane of
an image and in this pixels are manipulated directly in a
particular image [1]. There is a modificationinthepixelsthat
helps to obtain an improved appearance and visibility of
objects in an image [2]. The spatial domain techniques are
more popular than the frequency based methods, because
they are applied on the direct pixels values. Some of the
spatial domain techniques use either linear or non linear
level of intensities transformations functions, while others
use the composite study of the diverse features of image like
the edge detection and information of connected
components [3].
The image enhancement is done particularly for low light
and moderately low light conditions. [4] The purpose of
enhancement process is to obtain an image in which the
result is more appropriate than the original image for a
particular or definite application [5]. Nowadays, extend of
new image enhancement techniques is vital and desirable
problems in study of image processing. [6].
Fig. 1 Pre Processing Steps [7]
The Fig. 1 tries to explain the pre processing steps that are
applied to the input image and desired image is obtained by
application of these methods. It also helps to eliminate the
problem by increasing the dynamic range of digital valuesof
the pixel [5]. The known methods of image enhancement
techniques can be defined namely into two main branches
i.e. Global and local methods [6]. The paper is organized as
follows: Section 1 deals with the introduction to image
enhancement. The section 2 deals withthepreviouswork on
contrast enhancement. Section 3 deals with the different
histogram techniques. Section 4 givestheconclusionandthe
future work that might be useful in implementing the
different techniques on different set of images.
2. CONTRAST ENHANCEMENT
The Contrast Enhancement is the most significant and
essential technique of the spatial based imageenhancement.
The basic intend of the contrastenhancementtechniqueisto
adjust the local contrast in the image so as to bring out the
clear regions or objects in the image [1]. The contrast
enhancement tries to change the intensity of the pixel in the
image, particularly in the input image for the purpose to
obtain a more enhanced image. Contrast enhancement is
based on the number of techniquesnamelylocal,global,dark
and bright levels of contrast [8]. The contrast enhancement
is considered as the amount of color or gray differentiation
that lies among the different features in an image [9].
The contrast enhancement improves the quality of image by
increasing the luminance differencebetween theforeground
and backgrounds [5].
Histogram
Computatio
n
Thresholding
transformation
on
Illumination
Estimation

Fig. 2 The methodology for applying the enhancement
techniques
The Fig. 2 gives the methodology applied toobtainenhanced
image. The contrast enhancement is an importanttechnique
of image processing for the enhancement of an image in the
spatial domain [10]. The Contrast Enhancementdealswitha
number of techniques, mainly Histogram Equalization. The
histogram Equalization is the technique that assists in
improving the different ranges particularly the dynamic
range of the histogram of that of a particular image. The
histogram equalization depends on the global and local
variables for the aim to enhance the contrast of animage[5].
It is extensively available for processing of medical images
and acts as basis for pre processing move for the
applications in speech recognition, synthesis of texture in
images and other applications of video processing [11].3
3. HISTOGRAM EQUALIZATION
The histogram equalization is the basis for the modified
histogram techniques. The various histogram equalizations
techniques are available depending on their applications.
The techniques can be applied to the different image
datasets. The images may be of day or night conditions, find
its applications in medical or traffic applications. The image
after using histogram equalization gives brighter image as
compared to input image because it shows uniform
distribution [10]. But due toitslimitationofmean brightness
in histogram equalization, histogram matching takes place.
Histogram Equalizationworksbyhistogramflatteningandto
stretch the dynamic ranges of the graylevelsbytheusingthe
principle of the cumulative density function of the image.
The histogram equalization can be applied on the numberof
images. A sample image is taken from the dataset and the
histogram is showed in Fig. 3 and Fig. 4. The histogram
equalization technique executestheoperationbyremapping
the levels (gray level) of a particular image taken which is
based on the probability distribution of the gray levelsofthe
input image [11].
The histogram equalization method is most likely the best
known contrast enhancement methods for images
(especially the gray level images) duetoits easeofusageand
outcomes [12]. The histogram equalization preserves the
brightness on the output image but does not have a natural
look. The histogram equalization emphasize on the edges
and boundaries for the different set of images, but assist in
reducing the local details of the taken images and is not
suitable for enhancement of local details [3]. The classical
histogram equalization can well organized make use of the
intensities but it also likely to over enhance the contrast
levels if the histogram contains the high peak values, which
sometimes obtains a noisy and blurred emergence of the
desired image. [13]. Histogram is classified as the statistic
probability distribution of gray level in an image. Histogram
Equalization is one of the distinguished methods which aim
to enhance the contrast level of the taken images, producing
the desired image to have a distribution and more precisely
uniform distribution of the gray levels in an image. It also
performs the compressing and stretching the range of the
histogram of an image and consequence results particularly
for contrast enhancement. [14].
I. Global Histogram Equalization: is a technique that utilizes
the information of histogram of the original image by
applying the transformation function to the image. Though,
this method or approach is appropriate for the whole
enhancement procedure, but it falls short to adjust with the
features, specifically like the brightness levels of theoriginal
image. If there is an existence of a few numbersofgraylevels
in the image which has high frequencies, then they take
control of the remaining levels which have low frequencies.
In situation like this, Global Histogram Equalization tries to
remaps the gray levels in a desired method in which the
technique of contrast stretching turn out to be limited in
various dominating gray levels, which has the bigger
components of image histogram and results in major loss in
contrast levels for some small ones [11].
II. Local histogram equalization: method is the process that
develops the input image on the basis of block-by-block
method and makes use of the transformation function of
histogram equalization and theroleofthedevelopedblock is
to adjust the centre pixel of an image. The other part of this
technique is the Automatic Local Histogram Specification,
which is useful for the local details in a way that for every
pixel in its neighborhood/block of desired size is distinct
with the every other pixel being at the centre of the block.
This technique basically specifies the desired histogram
automatically which produces theenhancement whichisnot
only clear but also optimal enough to preserve the mean
brightness of the desired block about every pixel in a digital
image. This technique is proved by the results through
means of simulation in which a great balance is between the
enhancement details and the original image preservation in
a desired image that aims to attain by means of utilizing the
different histogram equalization methods [12].
Take the
sample image
from the
dataset
Apply pre
processing steps
like the gray
scale image and
filters.
Apply Contrast
Enhancement
Techniques

The Histogram Equalization is applied to the digital images
by various techniques:
I. Adaptive Histogram Equalization: It is a histogram
technique in which the various are used to improve the
levels of contrast values. It varies from the
ordinary equalization in the way that this method attempts
to compute the several histograms,inwhich everyhistogram
will correspond to a different sections of the image, and
usage is found to restructure the values of the lightness in
the processed image [5]. Therefore, it is suitable for
recovering the value of the local contrast level and improves
to enhance the boundaries or edges in every region of an
image.
II. Contrast limited Adaptive Histogram Equalization: This
method works on the principle that the image is being
divided into number of parts and the enhancement on each
and every part is analyzed individually. The resultant part is
the combined result of the part taken all together. The
application part is the sky images and underwater images.
[10]. The Contrast limited Adaptive histogram Equalization
is an efficient algorithm for enhancement of the local details
in an original image. The main disadvantage is that it
exploited to the effects of overstretching of the contrast
levels and various problems of noise. [13]
(i). CLAHE – DWT – The sample image taken from the
dataset is decomposed into the components of both low and
high frequency by the principle of DWT. The coefficients of
low frequency that make use of contrast limited adaptive
histogram equalization are enhanced and by keeping the
coefficients of high frequency unaffected which limits noise
enhancement. This method reconstructsthedigital image by
taking the reverse DWT of the obtained coefficients [14].
III. Dynamic Histogram Equalization: This method assists in
the control of the effect of an image without craftinganyloss
of any information in the image. This method partitions the
histogram of an image that has the foundation on the
principle on local minima and allocates thedesiredrangesof
gray level for every partition ahead of applying equalization
to them separately. These dividing levels go further through
one more partitioning level to make sure the lack of any
ruling portions [11]. The dynamic histogram equalization
utilizes a better partitioning method over the histogram of
the input image for chopping the histogram some sub-
histograms which does nothaveanydominatingcomponent.
Then every sub-histogram passes through histogram
equalization and is approved to engage in a desired range of
gray levels obtained in the improved image. So,ingeneral an
improved contrast enhancement level is gained by the
dynamic histogram with the help of dynamic gray levels
ranges which are controlled and to eliminate the option of
components of low histogram as they arecompressedwhich
might cause few parts of the image which have washed out
appearance. The main advantage of this technique is that it
ensures the consistencyinpreservingimageinformationand
it is protected from any severe side effects. The whole
procedure can be defined into the three main parts namely,
partition the levels in the histogram, assigning the ranges of
gray levels for every sub-histogram and assigning the
histogram equalization technique on every partitioned part
[11].
(i). Brightness Preserving Dynamic Histogram Equalization:
In this technique, the original image is decayed into many
sub parts or sub images by means of using the concept of
local maxima, then this equalization technique is functional
to the every sub part and resulting the combination of sub
images.. It separates the histogram which is directed on the
basis of local maxima. It generates the desired image with
the aim to find out the mean intensities which are almost
equivalent to the mean intensity of the input image and by
that it accomplishes the main condition of preserve the
brightness of the digital or desired image [8].
IV. Multi Histogram Equalization: In order to improve the
natural image that is desired after applying the histogram is
provided by this technique. This technique also uses the
method of decomposing the image into several sub images
and technique of histogramequalizationisprocessedto each
sub-image. [3].The intensity value that divides the images to
sub images is the optimal threshold set. It uses two
algorithms by which the threshold values are calculated.
The first algorithm calculates the optimal threshold set
based within class variance and separates theinputimage in
to different sub images based on the threshold set. The
second part separates the input image into several sub
images that form basis to the optimal threshold setusingthe
Otsu Method. The main advantage is that improves image
contrast by brightness preserving and generates natural
looking images. It has also been found out that it preserves
the brightness more efficiently than other methods and also
the error is also reduced. [3].
V. Brightness Bi-Histogram Equalization: The traditional
histogram equalization has the main drawback in terms of
not preserving the original brightness and the essential
contrast levels. The drawback is overcome the brightness.
The basic theory behind the BBHE is that the image is
decomposed into the sub-images. This technique generates
an image which has the brightness values placed in the
middle of the input image mean [8].
(i). Minimum Mean Brightness Error Bi-Histogram
Equalization: This method is the extension of the BBHE, in
the respect that in this method the threshold level is
searched and then the decomposition of the images takes
places into the sub-images. The aim behind this is that the
minimum brightness between the images of input and
output is achieved [8]. It also provides the maximum
brightness preservation. It also used to carry out the
separation which has threshold level as its source, which
further obtains the value of minimum Absolute Mean
Brightness Error.

(ii) Brightness Preserving Histogram Preserving with
Maximum Entropy: The concept of maximum entropy
attempts to obtain the values by the variational approach,
the objective histogram thosemaximstheentropy,underthe
values of constraints in which there is a fixed value of mean
brightness and then alters the histogram originallyobtained
which is desired histogram by using the histogram
specification [14]. Thus, an optimized brightness preserving
method under histogramtransformationismaximizedto the
target entropy of histogram under the constraints values of
brightness levels.
Fig. 3 Original image
Fig.4 The histogram equalised image
Techniques
(Application
Specific)
Main points
Histogram
Equalization
Easy to use.
Loss of mean brightness.
Adaptive Histogram
Equalization
Improvestoenhanceboundaries
in an image.
Contrast Limited
Adaptive Histogram
Equalization
Efficient algorithm for
enhancement of local details.
Applied to sky images and
underwater images [10].
Exploited to the effects of
overstretching and noise issues
[13].
Dynamic Histogram
Equalization
Consistency in preserving image
information [11].
Bi-Histogram
Equalization
Overcome brightness levels [8].
4. CONCLUSIONS & FUTURE WORK
This paper tries to sum up all the techniques that can be
applied in the image. An analysis of these entire contrast
enhancement techniques leaves a scope of more
improvement in the histogram techniques based on the
different applications. The future work should notonly beto
identify which enhancement technique is better for a
particular application but also that enhancement results
should be further used in detection and classification of the
data.
REFERENCES
[1] Y. Rao, L. Chen, “A Survey of Video Enhancement
Techniques”, Journal of Information Hiding and Multimedia
Signal Processing”, Vol. 3, No. 1, pp. 71-99, January, 2012.
[2] P. Mamoria, D. Raj, “An analysis of images using fuzzy
contrastenhancementtechniques”,International Conference
on Computing for SustainableGlobal Development”,pp.288-
291, 2016.
[3] A. S. Krishna, G. S. Rao, M. Sravya,“ContrastEnhancement
techniques using histogram equalization methods on color
images with poor lightning”, International Journal of
Computer Science, Engineering and Applications (IJCSEA),
Vol. 3, No. 4, pp. 15-24, 2013.
[4] P. H. Yawalkar, P. N. Pusdekar, “A Review on Low Light
Video Enhancement using image processing technique”,
International Journal ofAdvanced ResearchinComputerand
Communication Engineering”, Vol. 4, No. 1, 2015.
[5] A. Singh, S. Yadav, N. Singh, “Contrast Enhancement and
Brightness Preservation using Global Local Image
Enhancement Techniques”, 4th International Conference on
Parallel, Distributed and GridComputing,pp.291-294,2016.
[6] E. Yelmanova, Y. Romanyshyn,“HistogramBasedMethod
for Image Contrast Enhancement”, 14th International
Conference in Experience of Designing and Application of
CAD Systems in Microelectronics (CADSM),pp. 165-169,
February, 2017.
[7] K. Akila, S. Chitrakala, S. Vaishnavi, “Survey on
IlluminationConditionofVideo/ImageunderHeterogeneous
Environments for Enhancement”, 3rd International
Conference on Advanced Computing and Communication
Systems (ICACCS), pp. 412-418, January, 2016.
[8] V. A. Kotkar, S. S. Gharde, “Review of various image
Contrast Enhancement Techniques”,International Journal of
Innovative ResearchinScience,Engineeringand Technology,
Vol. 2, No. 7, 2013.
[9] S. Lal, M. Chandra, “Efficient algorithm for contrast
enhancement of Natural Images”, International ArabJournal
of Information Technology, Vol. 11, No. 1, pp. 95-102, 2014.

[10] V. Yadav, M. Verma, V. D. Kaushik, “Comparative
analysis of contrast enhancement Techniques of Different
Image”, 2nd International Conference on Computational
Intelligence & Communication Technology, pp. 76-81, 2016.
[11] M. A. Al Wadud, Md. H. Kabir, M. A. A. Dewan, O. Chae,
“A Dynamic Histogram Equalization for image contrast
enhancement”, IEEE TransactionsonConsumerElectronics,
Vol. 53, No. 2, pp. 593-600, 2007.
[12] I. Jafar, H. Ying, “A New Method for Image Contrast
Enhancement based on Automatic Specification of Local
Histograms”, International Journal of Compute Science and
Network Security (IJCSNS), Vol. 7, No. 7, pp. 1-10, July 2007.
[13] H. Lidong, Z. Wei, W. Jun, S. Zebin, “Combination of
contrast limited adaptive histogram equalization and
discrete wavelet transform for image enhancement”, IET
Image Processing, Vol. 9, Issue No. 10, pp. 908-915, April,
2015.
[14] K. Murahira, T. Kawakami, A. Taguchi, “Modified
Histogram Equalization for Image Contrast Enhancement”,
Proceedings of the 4th International Symposium on
Communications,Control andSignal Processing(ISCCSP),pp.
1-5, March 2010.
[15] S. D. Chen, A. R. Ramli, “Minimum Mean Brightness
Error Bi-Histogram Equalization in Contrast Enhancement”,
IEEE Transactions on Consumer Electronics, Vol. 49, No. 4,
pp. 1310-1319 November 2003.
[16] D. Menotti, L. Najman, J. Facon, A. de A. Araujo, “Multi-
Histogram Equalization methods for contrast enhancement
and brightness preserving”, IEEE TransactionsonConsumer
Electronics, Vol. 53, No. 3, pp. 1186-1194, August, 2007.
[17] C. Wang, Z. Ye, “Brightness Preserving: Histogram
Equalization with maximum entropy: A Variational
Perspective”, IEEE Transactions on Consumer Electronics,
Vol. 51, No. 4, pp. 1326-1334, November, 2005.
[18] P. Kaushik, Y. Sharma, “Comparison Of Different Image
Enhancement Techniques Based Upon Psnr & Mse”,
International Journal of Applied Engineering Research, Vol.
7, No. 11, pp. 2010-2014, 2012.

Contrast Enhancement Techniques: A Brief and Concise Review

More Related Content

What's hot (18)

Similar to Contrast Enhancement Techniques: A Brief and Concise Review (20)

More from IRJET Journal (20)

Recently uploaded (20)

Contrast Enhancement Techniques: A Brief and Concise Review