3

An Enhancement of Radiography Images using Gamma Correction
CHAPTER 1
INTRODUCTION
1.1 MOTIVATION
In todays time health concern services or practices and the development of the
common platforms for radiography images have resultant data in efficient way. Digital
Image Processing is a rapidly evolving field with the growing technology in science &
engineering. Image Processing holds the possibility of developing an ultimate machine
that could perform visual activity of the entire living beings. An image can have low
contrast or bad quality due to a number of reasons like low quality of imaging tool,
adverse external environment at the time of image acquisition and many more [2]. The
image processing is a visual responsibility, the foremost step is to get an image i.e.
image acquisition then enhancement and lastly to process. In this, there is information
for image enhancement for the purpose of image processing. Image enhancement is
mostly used to improving the digital image quality. Image histogram is useful in image
enhancement. The histogram in the perspective of image processing is the action by
which the occurrences of each intensity significance in the image are shown and
Histogram Equalization is the technique by which the dynamic range of the histogram
of an image is enhanced.
Contrast is an essential feature in any objective evaluation of image quality.
Contrast is produced by the difference in luminance get back from two adjacent
surfaces. In additional words, contrast is the difference in visual properties that show an
object distinguishable from other objects and the background. In visual perception,
contrast is determined by the difference in the color and strength of the object with
other objects. Our visual system is more perceptive to contrast than complete
luminance; therefore, we can observe the world similarly regardless of the considerable
changes in illumination conditions. Contrast enhancement is a procedure in which
change the image value distribution to cover a wide range [3]. Contrast enhancement is
one of the mainly common methods of low-level digital image processing.
1.2 RADIOGRAPHY IMAGES

Radiography images are basically used in the medical science. Radiology is a branch of
medicine that uses imaging technology to diagnose and treat disease. Radiology may be
separated into two different areas, diagnostic radiology and interventional radiology.
The division of therapeutic radiation, which uses more powerful x-rays to treat cancer,
is now called radiation oncology. It is an imaging way that uses electromagnetic
radiation other especially X-rays to view the internal structure of a non-uniformly
composed and difficult object (i.e. a non-transparent object of unreliable density and
composition) such as the human body.
1.3 CONTRIBUTION AND ORGANISATION OF THESIS

1.3.1 SEGMENTATION
Image segmentation is the manner of isolating an image into different parts.
Segmentation defined to the process of dividing a digital image into the different
segments (set of pixels as known as super pixels). The aim of segmentation is primarily
segment an image into something that is more meaningful and easier to analyze [4].
This is typically used to distinguish objects or extra relevant information in digital
images. The following categories are used:
Threshold Based Segmentation: It is mostly used to segment an image on the
basis of their threshold value. Thresholding methods change each pixel in an
image with a black pixel if the picture intensity Ii, j is less than some fixed
constant T or a white pixel if the picture intensity is greater than that constant
[5].
Edge Based Segmentation: The edge-based segmentation can be used to the
study by significantly minimizing the amount of pixels from an image to be
processed, while still preserving sufficient object structures [6]. It is also used to
find the object boundaries and then locate the object itself by satisfying them.
Region Based Segmentation: Region growing is an easy region-based image
segmentation process. It is also divided as a pixel-based image segmentation
technique since it involves the selection of initial seed points. This approach to
segmentation determines neighboring pixels of initial starting points and
determines whether the pixel neighbors should be further to the region.
Clustering Techniques: Clustering is basically used where the gray-level
inputs are clustered in two parts as background and foreground (object), or
alternately are modeled as a combination of two Gaussians.
Here, we consider threshold based segmentation to find the threshold value to divide an
image into two parts contain low contrast and high contrast regions.
1.3.2 GAMMA CORRECTION
The Gamma Correction is related on a non-linear multiplication-based conversion and it
overcomes the effect of light [1]. Gamma Correction, Gamma Nonlinearity, Gamma
Encoding, or regularly simply Gamma, is the name of a nonlinear operation used to
code and decode luminance or inputs in video or still image system. Gamma Correction
is, in the simplest cases, defined by the following power law expressionVout = AVin
Where A is a constant and the input and output data are non-negative real values; in the
common case of A = 1, inputs and outputs are classically in the range 01. A Gamma
value < 1 is most of the time called an Encoding Gamma, and the procedure of
encoding with this compressive power-law nonlinearity is called Gamma Compression;
on the other hand a Gamma value > 1 is called a Decoding Gamma and the function
of the expansive power-law nonlinearity is called Gamma Expansion. The thought of
Gamma can be applied to any nonlinear relationship.
1.4 PROBLEM STATEMENT

In this research we justify the problem regarding to the uneven illumination of a
radiography image.
How to enhance the quality of an image and find out the PSNR and MSE values
to show the minimizing error of the proposed method.
1.5 CHALLENGES
The primary challenge of my thesis is on the radiography image to enhance the
quality using Gamma Correction.
This combination of low contrast and poor saturation of image is a very
common and it is a challenging task to obtain by capturing the images in
lowlight circumstances.
1.6 OUTLINE OF THE THESIS

The thesis is written chapter wise. Each chapter has its own data with an effort to make
the user to go through the thesis easily.
Chapter 1: gives the general motivation towards the work. Then it defines the problem
statement and the challenges posed by the research work.
Chapter 2: starts with the literature survey of the related work then moves towards the
basic concepts needed to understand the thesis.
Chapter 3: explains about the threshold value in segmentation with the help of this we
divide the image.
Chapter 4: it explains the working algorithm and the process followed to achieve the
desired goal.
Chapter 5: it gives the result and experimentation results after execution of the proposed
system.
Chapter 6: it concludes the thesis work and gives inspiration to extend the work with
new challenges as the future work.
1.7 OPERATING SYSTEM AND TOOL USED

The operating system and tools used in the thesis are as follows:
Operating system: Windows 10
Processor: Intel Core 2 Duo CPU
T5470 @ 1.60GHz
Installed memory (RAM): 2.00 GB

System type: 32-bit operating system
Tool and Technology Used: MATLABR2009b (7.9.0.529)
CHAPTER 2
LITERATURE REVIEW
2.1 RELATED WORK
Karamjeet Kaur, et al. (2014) [1] has focused to improve the visual effects and
the contrast of image or to make the original image more informative or beneficial to
computer to process. Due to technical limitation, many imaging devices may not
display the actual appearance of objects. Even if nearly all image enhancement methods
can effectively improve the perception of information for human observer, however,
these proposed methods have their own limit in weakening the effect of uneven
illumination. They still have several deficiencies to give perfect image to viewer such as
loss of local contrast and the loss of details. The important factor which is directly
reducing the appearance of image is patchy illumination. In this paper to resolve
illumination effect on image quality, local gamma correction with adaptive threshold is
used to develop the visual appearance of the image or to provide enhanced transform
representation for future automated image processing. Resultant results are measured by
finding mean, standard deviation, average gradient and entropy values and are
compared with Local Gamma Correction techniques. The definitive reason behind the
proposed method is to permit higher level of brightness preservation to avoid unwanted
artifacts and unnatural enhancement due to extreme equalization while enhancing the
contrast of a particular image as much as possible.
Seema Rani, et al. (2014) [2] explained that the main idea behind contrast
enhancement techniques is to increase contrast and to preserve original brightness of
images. In this article a contrast enhancement technique is proposed that first segments
histogram of image recursively and after that applies Adaptive Gamma Correction with
Weighting Distribution (AGCWD) Technique. In this paper, author applied this
technique also on the colored images. The proposed technique is essentially an
improvement over AGCWD technique and objective to get better contrast enhancement
and brightness preservation than AGCWD technique. In this work recursive
segmentation of histograms is done and then weighting method is applied to smooth
down histogram. Further Gamma Correction is also applied that helps to control the
brightness of image. The work improves the Adaptive Gamma Correction using
Weighted Distribution (AGCWD) technique that had implemented Gamma Correction
and Weighted Distribution techniques with no segmentation of histogram. Thus this
work improves results by recursive segmentation of histograms that helps in better
enhancement of images.
S. Gayathri, et al. (2014) [7] has discussed that main objective of contrast
enhancement of the image. In this, it basically focuses to correct the local contrast in
different part of the image by which the feature invisible in the dark or bright area is
brought out and showing to the human views. It also preserves the brightness of the
image. So, it uses different type of techniques to enhance the contrast. Unwanted
environmental conditions in the captured view, such as the occurrence of clouds, short
of sunlight or indoor lighting, and some other condition. It improves the brightness of
the low contrast images and preservation of image feature using multilevel
thresholding, Gamma Correction and probability based entropy. The proposed method
conserves the feature of the image and brightness and also increases image quality.
Finally author concluded that the result of the enhanced image clear is measured using
in term of entropy value. The enhanced image using the proposed method brings out
more information of the original image and visual quality is improved well. This paper
explains different type of enhancement technique by which we can enhance the contrast
of an image. First one is Recursively divided and weighted histogram equalization
(RDWHE) for brightness preservation and contrast enhancement. In this, the essential
processes to divide input images into two are more sub-histograms recursively and
adjust the sub- histograms with the help of weighting process. Second is Image
Enhancement Based on Equal Area Dualistic Sub-Image Histogram Equalization
(DSIHE) Method. In this method decompose the original input image into two equal
sub-images on the basis of density function. Third is Recursive Sub-Image Histogram
Equalization (RSIHE) applied to gray scale images. It explains recursively separate the
image at several times to get multiple sub-histograms. Last one is Brightness Preserving
Histogram Equalization with Maximum Entropy (BPHEWME). The basic idea of this
approach to maximize the entropy by using target histogram.
Vivek Singh Rathore, et al. (2013) [8] in this paper author doing study of digital
filter on three types of noise like as Salt and Pepper (SPN), Gaussian noise, Speckle
(SPKN). Dissimilar noise variance will be removed between 10% to 60% by using
different types of filter as median and high pass digital filter. Filter methods are mostly
used for de-noising, smoothness and sharpening of images and extracting the useful
data for the analysis for image processing. The same idea is applied to the different
images and they are compared with one another. The study is proposed with the help of
Mean Square Errors (MSE) and Peak Signal to Noise Ratio (PSNR). So as to select the
appropriate noise for different filtering methods for any image. Results are simulated on
MATLAB R2007b.
H. Hassanpour, et al. (2011) [9] image enhancement is very essential preprocessing stage in most image processing applications. This paper explains a new
automatic image enhancement technique by modifying the Gamma value of its
individual pixels. Author locally measure the gamma values in an image with support
vector machine (SVM). A new Gamma Correction method that estimates image
Gamma values not including any calibration information or knowledge of the imaging
device. The method is able to Most of existing Gamma Correction methods applies a
uniform Gamma value across the image. To improve the Gamma values of an image the
aforementioned features are primarily computed in sliding windows, then SVM is
employed to estimate the Gamma value in each window. In this study, it is exposed that
the proposed method has performed well in improving the quality of images. The
proposed method compared to the existing methods in image quality enhancement
using image Gamma value. Experimental results in this research signify that the
proposed method improves image quality, and enhances the dynamic and information
of the image.
Rajeshwar Dass, et al. (2012) [10] paper based on the segmentation. This paper
explains the different segmentation techniques used in the field of ultrasound and SAR
Image Processing. The Research on Image segmentation for many years has been a high
degree of attention. Thousands of different segmentation methods are available, but
there is not a single way which can be considered good for different images, all
methods are not uniformly good for a particular type of image. Firstly this paper
investigates and compiles several of the technologies used for image segmentation. It
basically survey of current segmentation techniques and then finally general tendencies
in image segmentation are presented. This paper explains different type of segmentation
techniques by which we can segment an image into several regions. It explain
thresholding methods, region growing, region splitting and merging methods on the
basis of region based segmentation and segmentation based on clustering, etc. At last
there is no single method which can be considered good for neither all type of images,
nor all methods equally good for a particular type of image. Due to all over factors,
image segmentation remains a challenging problem in image processing and computer
visualization and is still a pending problem in the world.
Aman Kumar Sharma, et al. (2013) [11] the image segmentation is one of the
most important tasks in the field of image processing and pattern recognition. There are
different types of algorithms that are used for image segmentation such as clustering,
threshold, fuzzy logic, edge detection and watershed transformation. Here, author
basically used Watershed transform algorithm. Watershed transform is a way that
always generates closed contours for every region of an image and one of the best way
to represent for image segmentation but over segmentation is a major drawback of this
method for which present study was undertaken to create a new three step methodology
for image segmentation using watershed transformation working initially with presegmentation processing then detecting edges of image and then computing watershed
transformation. As a result of applying these segmentation techniques on image, better
results were observed than traditional methodology. Superior marker based watershed
transformation for image segmentation was done using MATLAB.
Er. Kritika Vaid, et al. (2014) [12] says that image processing plays a vital role
in phantasm application. It improves the visibility of poor images. Different techniques
have been estimated so far to get better image quality. Image enhancement can
selectively enhance and contain some information about image. Most of the techniques
are useful for altering the gray level values of individual pixels and therefore the overall
contrast of the entire image. It is a method which decreases image noise, eliminate
artifacts, and preserve information. Its purpose is to amplify certain image features for
analysis, diagnosis and exhibit. The overall objective of this article is to propose an
integrated technique which will integrate the nonlinear enhancement technique with the
Gamma Correction system. The proposed algorithm is implemented in MATLAB. And
the image enhancement quality can be explained by the Absolute Mean Brightness
Error (AMBE), the Discrete Entropy (H) and PSNR to assess the enhancement quality
between the low input image and the improved image.
Satbir Kaur, et al. (2014) [13] tells that there is no accepted universal image
enhancement algorithm or specific criteria which can fulfill user needs. The processed
image may be very dissimilar with the original image in the visual effects, but it also
may be similar to the original image. It will be a developing tradition to integrate the
advantage of various algorithms to practical application to image enhancements. Author
presents an adaptive image contrast enhancement method. It works locally to decrease
contrast as if we present enhancement techniques globally on portions which are
already bright then this gives poor results. Enhancement techniques are useful only to
those dark portions. We need accurate method that not only enhance the image but also
preserve the information. The proposed method is based on a Local Gamma Correction
piloted by histogram analysis. In this paper, to continue away from uneven luminance
image is divided into different segments. It works locally to reduce contrast as if we
perform enhancement techniques globally on portions which are already bright then this
gives poor results. Enhancement techniques are useful only to those dark portions. We
need accurate method that not only enhance the image but also preserve the
information.
S.S. Bedi, et al. (2013) [14] discussed about the aim of image enhancement is to
improve the visual appearance of an image, or to provide a better change representation
for future automated image processing. Many images like medical images, satellite
images, aerial images and still real life photographs suffer from reduced contrast and
noise. It is necessary to enhance the contrast and eliminate the noise to increase image
quality. One of the most important parts in medical images detection and analysis is
Image. An enhancement technique which improves the quality (clarity) of images for
human viewing, removing blurring and noise, increasing contrast, and informative
information are examples of enhancement operations. The enhancement procedure
differs from one field to another according to its purpose. The existing techniques of
image enhancement can be classified into two parts: Spatial Domain and Frequency
domain enhancement. In this paper, we present an outline of image enhancement
processing techniques in spatial domain. More specifically, we categories processing
methods based representative methods of Image enhancement. Thus, the contribution of
this paper is to classify and review image enhancement processing way, effort an
evaluation of shortcomings and general needs in this field of active research and in last
we will point out shows potential directions on research for image enhancement for
future research.
Amandeep Kamboj, et al. (2012) [15] discussed about this paper which presents
an efficient architecture for Image Segmentation. This architecture offers an alternative
during a Graphical User Interface tool MATLAB. Image segmentation can be obtained
by using various methods, but the problem of most of the methods is that they use a
high level language for coding. This paper focuses on processing an image pixel by
pixel and in adjustment of pixel neighborhoods that can be applied to the whole image.
All of the pixels in a region are similar with value to some characteristic or computed
property such as color, intensity, or texture. Some realistic applications of image
segmentation are: Image processing, computer visualization, Face recognition, Medical
Imaging, Digital libraries, figure and Video recovery. The objective show to the use of
a tool with a high-level graphical interface using the Matlab Simulink based blocks
which makes it very easy to feel with respect to other software. The different
applications where noise removal, enhancing edges and contours, blurring and so on.
This paper enhances the architecture of filtering images for edge detection with the help
of Video and Image Processing block set. The filters are intended by blocks and it even
supports MATLAB codes through user customizable blocks. It moreover offers an ease
of designing with GUI environment.
Salem Saleh Al-amri, et al. (2010) [16] this paper attempts to take on the study
of segmentation image techniques by using five threshold methods as Mean method, Ptile technique, Histogram Dependent Technique (HDT), Edge Maximization Technique
(EMT) and visual Technique and they are compared with one another so as to select the
best technique for threshold segmentation technique image. Mean is basically used to
find the mean value of the pixels as the threshold value and works well in severe cases
of the images that have just about half to the pixels belonging to the objects and the
other half to background. The p-tile technique uses information about the area size of
the desired object to the threshold an image. The P-tile method is one of the initial
threshold methods based on the gray level histogram. It assumes the objects in an image
are brighter than the background, and take up a fixed percentage of the picture area. The
histogram based techniques is dependent on the success of the assuming the threshold
value that separates the two homogonous region of the object and background of an
image. The threshold image by using edge maximization technique (EMT) is used when
there are more than one homogenous region in image or where there is an adjust on
illumination between the object and its background. In this paper, author also explains
the concept of how to select the threshold value to divide an image. These techniques
applied on three satellite images to decide base guesses for threshold segmentation
image.
Gajendra Singh Chandel, et al. (2012) [17] image segmentation has played a
significant role in computer vision mainly for human tracking. The result of image
segmentation is a set of segments that mutually cover the whole image or a set of
contours extracted from the image. Its accuracy but very indefinable is very crucial in
areas as medical, remote sensing and image retrieval where it may contribute to save,
maintain and protect human life. This paper presents the analysis and implementation
using MATLAB features and one finest result can be selected for any algorithm using
the subjective evaluation. We considered the techniques below the following five
groups: Edge-based, Clustering-based, Region-based, Threshold-based and Graphbased. After analysis of a variety of image segmentation algorithms and the comparison
of the results of each algorithm separately with different parameter's value using
MATLAB, the ending is that: In Edge Based Segmentation Algorithms, the Canny
Algorithm produced the best segmentation in comparison of Sobel, Prewitt and LoG. In
thresholding Based Algorithms, the Adaptive thresholding and Otsu thresholding
produced good outcome. The Adaptive thresholding produced the good edges and Otsu
thresholding recognized the object very fine. In Region Based Algorithms, the split and
merge method produced the better result. In Clustering Based Segmentation, the mean
shift method formed the good result. When the K-means and Fuzzy-C means methods
are compared, the Fuzzy C means is improved than the K-means method. In Graph
Based Algorithms, Normalized-cut is used to cut an image into specified number of
cuts. Additional methods are interactive methods. In N-cut, the foreground and
background area is selected by a user.
Chhaya Gautam, et al. (2015) [18] this research paper explain the application of
good quality of digital image which is face recognitions, computer vision applications,
satellite imaging and various security porpoises. Due to lack of light or insufficient
environment condition or some further causes many time captured images has bad
contrast issues. It is essential to improve contrast of the image for better Performances
in all image and video applications. These papers propose a well-organized method for
improving low contrast effects in colored images. This method uses hybridization of
Range Limited Bi-Histogram Equalization and Adaptive Gamma Correction technique.
Histogram Equalization is most popular method to enhancement of low contrast image,
Bi-Histogram equalization method enhanced the results of basic Histogram

Equalization contrast enhancement but Histogram Equalization method does not show
the brightness improvement in enhanced image. In other hand Gamma Correction
method improves the brightness of low contrast image. The proposed hybridized
method Range Limited Bi-Histogram Equalization with Adaptive Gamma Correction is
explaining in low contrast color images and efficiently enhanced contrast better than by
means of Range Limited Bi-Histogram Equalization and Adaptive Gamma Correction
separately. In this paper result will be calculated absolute mean brightness error is also
found minimum for proposed method. Through all these results it is found that
proposed hybrid RLBHE with AGC is an efficient color image contrast enhancement
method.
Raman Maini, et al. (2010) [19] in this paper author review all the enhancement
techniques to improve quality of the image. The primary objective of image
enhancement is to adjust attributes of an image to make it more suitable for a given task
and a detailed observer. This paper will make available an overview of underlying
concepts, along with algorithms commonly used for image enhancement. The paper
focuses on spatial domain methods for image enhancement, with particular reference to
point processing methods and histogram processing. In spatial domain techniques, we
openly deal with the image pixels. The pixel values are manipulated to achieve
preferred enhancement. Image enhancement algorithms offer a wide variety of
approaches for modifying images to attain visually acceptable images. The variety of
such techniques is a function of the specific task, image content, observer
characteristics, and seeing conditions. The point processing methods are most primitive,
yet necessary image processing operations and are used primarily for contrast
enhancement. Image Negative is suitable for improving white detail embedded in dark
regions and has applications in medical imaging. Although we did not discuss the
computational cost of enhancement techniques in this article it may play a critical role
in choosing an algorithm for real-time applications. In spite of the effectiveness of each
of these algorithms when applied separately, in practice one has to develop a
combination of such methods to achieve more effective image enhancement.
Volodymyr Ponomaryov, et al. (2012) [20] in this article author tells about new
digital areas have brought in many new image and video applications and new
technologies in different fields, like as biology, medicine, engineering, and
entertainment. The images and videos are condensed, transmitted, captured, and stored
in various digital forms, with dissimilar types and amounts of impaired artifacts.
Although in bio-medical applications, speckle noise and coding artifacts are mainly
familiar, in the entertainment and engineering applications, the digital coding and show
artifacts are considered as dominant. Furthermore, each of the up-and-coming
applications and technologies has introduced different kinds of specific,
correlated/structured distortions. The elimination of such distortions, i.e., denoising, has
not adequately been investigated and needs to be further explored. This is particularly
true having in mind consistent growth of new up-and-coming digital multimedia
applications and services. This Special concern is dedicated to novel image and video
quality assessment and improvement techniques and systems which are designed for
currently challenging applications in which the visual quality is critical. Thirteen
wonderful research articles, regarding this topic, are published in this issue and cover
four important topics: Novel image and video processing optimization algorithms
10
depend on objective and subjective image quality assessment. In this article the author
has much acceptable SSIM Structural Similarity Index enthused image restoration with
sparse representation use the Structural Similarity (SSIM) index by combining it into
the structure of sparse signal representation and approximation. A gradient descent
algorithm is performed to achieve SSIM-optimal compromise in combining the input
and sparse dictionary reconstructed pictures. The performance of the proposed method
is established in image demising and super-resolution applications. The experimental
outcome explains that the proposed SSIM-based sparse representation algorithm
achieves better performance and superior visual quality than the related least squarebased method.
Ismail A. Humied, et al. (2012) [21] in this paper, a new proposed method is a
combination of Histogram Equalization (HE) and Fast Gray-Level Grouping (FGLG).
The basic process of this method is segments the original histogram of a low contrast
image into two sub-histograms on the basis of their locality of the highest amplitude of
the histogram components, and achieving contrast enhancement by equalizing the left
section of the histogram components using (HE) technique and using (FGLG) technique
to equalize the right segment of this histogram components. Some low contrast images
have certain characteristics makes it difficult to use traditional way to improve it. In the
proposed method, an automatic contrast enhancement technique is proposed that
performs efficiently with images that have the location of the highest amplitude
histogram components deceit in the left of NZHC region. It is an arrangement of
Histogram Equalization (HE) and Fast Gray-Level Grouping (FGLG). The proposed
method is carried out via various stages. An example of these characteristics, that the
amplitudes of images histogram mechanism are very high at one position on the gray
scale and very small in the rest of the gray scale. In this paper, a new system for
automatic contrast enhancement images is explained, which is a combination of the
Histogram Equalization (HE) method and the Fast Gray-Level Grouping (FGLG)
system. Experimental results demonstrate that the proposed method generated is
effective and robust enhancement and better enhancement for different images than
using each technique separately. The results have revealed that the proposed method
does not only produce better results than each individual contrast enhancement
procedure, but it is also fully automated. Moreover, it is applicable to a broad variety of
images that assure the properties mentioned above and suffer from low contrast.
K. Somasundaram, et al. (2011) [22] in this paper, author explain a new
automatic technique for medical image contrast enhancement based on Gamma
Correction. In this process, a global Gamma value is calculated depend on the image
cumulative histogram without any knowledge of the imaging device. Many images,
such as medical images, remote-sensing images, electron microscopy images and even
real photographic pictures, endure from poor contrast. Therefore, it is necessary to
enhance the contrast of such images before applying further processing techniques. We
evaluated our method using MR brain images and CT scan images. The performance of
the proposed system is compared with three popular contrast enhancement techniques
by means by PSNR measure. The luminance non-linearity introduced by much medical
imaging procedure often affects the performance of medical image processing
techniques. Image enhancement acting an important pre-processing step in medical
image processing methods. Automatic contrast enhancement by global histogram
processing is an essential tool for image enhancement. Histogram equalization (HE) is
11
an important technique for enhancing the contrast of an image. HE is related to contrast

stretching in that it attempts to enhance the dynamic range of the pixel values in an
image. Its necessary idea lies on mapping the gray levels based on the probability
division of the input gray levels. Adaptive histogram equalization (AHE) is an
automatic enhancement technique used to get better contrast in images. It differs from
ordinary histogram equalization in the value that the adaptive method computes
numerous histograms, each corresponding to a distinct section of the image, and uses
them to rearrange the intensity values of the image. AHE is measured an image
enhancement technique able of improving the local contrast of the image, bringing
exposed more detail in the image at the same time it generate significant noise. In this
paper, we proposed a new contrast enhancement method for medical images based on
gamma correction method. The gamma value is considered based on the cumulative
histogram. The quality of the enhancement is precise in term of PSNR values and
created higher PSNR than the existing method HE, IA and AHE. The proposed
technique is a simple and well-organized system it is directly forward to extent this
carry out to general gray scale images. Hence, this way can be used as a preprocessing
method for medical image processing.
Ankit Aggarwal, et al. (2013) [23], in this paper author discuss the improvement
of the image enhancement techniques and their application in the field of image
processing. The main purpose of image enhancement techniques is to procedure an
input image so that the resulting image is more appropriate than the original image for
specific application. Image enhancement, which is one of the major techniques in
digital image processing, plays a significant role in many fields, such as medical image
analysis, remote sensing, high description television, hyper supernatural image
processing, industrial X-ray image processing, microscopic imaging etc. Image
enhancement is a giving out an image in order to make it more suitable for specific
applications. It is generally utilized to improve the visual effects and the clarity of the
image or to make the original image more favorable for other automated processes. The
contrast enhancement is one of the frequently used image enhancement methods. Many
techniques for image contrast enhancement have been published and commonly used
which can be broadly categorized into two groups: direct method and indirect method.
Usually, an image may have poor dynamic range or distortion due to the poor quality of
the imaging devices or the unfavorable external conditions at the time of acquirement
and so on. Traditional global histogram equalization typically causes excessive contrast
enhancement while local histogram equalization may reason block effect. To overcome
these problems, a new method for image contrast enhancement is developed. The
originality of the proposed method is that the weighted average of the histogram
equalized, Gamma Corrected and the original image are joint to obtain the enhanced
processed image. The proposed algorithm not only attain contrast enhancement but also
conserve the brightness level. Experimental outcome show that the proposed algorithm
has good performance on enhancing contrast and visibility of images.
Amandeep Kaur, et al. (2014) [24], in this paper images are the most common
and suitable means of conveying or transmitting information. An image is implication a
thousand terms. Pictures in concise convey information on positions, sizes and interrelationships among objects. They explain spatial information that we can be familiar
with as objects. Human beings are better-quality at derive information from such
imagery, because of our inhabitant visual and mental abilities. About 75% of the
12
information received by human is in pictorial form. Histogram equalization is an

effective image enhancement technique, but brightness of an image can be
misrepresented after the histogram equalization, which is mainly due to the flattening
property of the histogram equalization. Image enhancement increases an image exterior
by increasing dominance of some features or by decreasing uncertainty between
different regions of the image. Histogram Equalization is extensively used to perform
contrast enhancement in images. As an outcome, such image creates side-effects such
as washed out appearance and false contouring due to the significant change in
brightness. In organize to overcome these problems, mean brightness preserving
histogram equalization based techniques have been proposed. Commonly, these
methods partition the histogram of the original image into sub histograms and then
independently make equal each sub histogram with Histogram Equalization. To
overcome this problem many modifications on histogram equalization have previously
been proposed such as BBHE and RMSHE etc. But these methods can generate
objectionable artifacts in images such as, over enhancement or unnecessary noise
generation in some cases. So a method should be evaluated that can enhance image
contrast as well as conserve image brightness and having low computational complexity
and having results recovered than existing techniques. So, we have projected a new
technique named Recursive Adaptive Gamma Correction. The point of this technique is
to calculate and to enhance the image contrast while preserving image brightness & to
get better the adaptive Gamma Correction with weighting distribution technique.
Results are analyzed on the basis of Calculation of the dissimilar image quality
measurement parameters such PSNR, MSE & AMBE. The proposed techniques give
better results analysis to BBHE, HE, RSWHE and AGCWD in all three parameters with
recursion value as 2 and Gamma co-relation value as zero which conserve brightness
and edges as an enhancement.
Wilfried Kubinger, et al. (2013) [25], in this article author explain the concept of
Gamma Correction. For todays computer vision systems it is a matter of course to
handle color cameras and to support color image processing. The main reason to add
color information to conventional intensity signals is to gain robustness and reliability
in visual applications. Commonly available CCD cameras represent the color data of a
pixel using the tristimuli ethics red, green and blue. This is referred to as the RGB color
space model. The model is strongly device dependent, since there is no widely accepted
definition of color wavelength and the characteristics of the filters to obtain these color
values. In this author analyze the influence of Gamma Correction of digital color
cameras on computer vision algorithms. This paper shows that tracking of colored
patches can be highly improved by correctly treating the Gamma Correction. Since
common cameras automatically use a Gamma Correction for purposes of viewing,
digital color image processing has to handle nonlinear signals which should be rendered
linear using the correct Gamma re-correction. Experiments of segmentation and
tracking based on different color space models confirm the claims.
Vijay A. Kotkar, et al. (2013) [26], author tells the objective of enhancement is
to process an image so that the result is more suitable than the original image for an
explicit application. Image enhancement is one of the most interesting and visually
appealing areas of image processing. Image enhancement is a processing on an image in
order to make it more appropriate for certain applications. It is used to get better the
13
visual effects and the clarity of image or to make the original image more conducive for
computer to process. Contrast enhancement altering the pixels intensity of the input
image to utilize maximum possible bins. We require to study and review the different
image contrast enhancement techniques because contrast losses the brightness in
enhancement of image. In view of this fact, the mixture of global and local contrast
enhancement techniques may enhance the contrast of image with conserving its
brightness. There are many image contrast enhancement methods like as HE, BBHE,
DSIHE, MMBEBHE, RMSHE, MHE, BPDHE, RSWHE, GHE, LHE and LGCS. This
paper based on the relative learning of contrast enhancement techniques with special
reference to local and global enhancement techniques. Also proposed solution is
acknowledged to apply to this enhancement technique. This novel method will use in
many fields, such as medical image investigation, remote sensing, HDTV, hyper
spectral figure processing, industrial X-ray image processing, microscopic imaging etc.
Generally in enhancement of image contrast will lose the brightness, the purpose of this
novel method is to enhance contrast with preserving natural visibility of image,
sharpening details in more contrasted manner, preserving the overall shape of
histogram, stretch histogram corresponding to their bins, utilize more bins by image
histogram showing better enhancement of image. Image contrast enhancement plays an
important role in image enhancement. In this paper, the different image contrast
enhancement techniques are analyzed. The mixture of global and local contrast
enhancement techniques is superior to other techniques because these techniques
improve the visual effects and clarity of the image with preserving its brightness. This
is proposed system, implementation work is going on. The major goal of image contrast
enhancement is to produce images without severe side effects at the same time maintain
input mean brightness.
Sai Sathish Konduri,, et al. (2014) [27], in this paper author explain the reason
due to the existence of noise and strength of in homogeneity in brain MR images many
segmentations from incomplete accuracy. It has been used to derive useful information
from the medical image data that get the most accurate and robust process for diagnosis.
The region based on level set of method is used for segmenting the medical images with
power in homogeneity. Intensity in homogeneity often occurs in real-world images due
to a range of factors like as spatial variations in illumination and imperfections of
imaging devices, which complicates frequent problems in image processing and
computer vision. In some aspects image segmentation may be considerably difficult for
images with amount in homogeneities due to the overlaps between the ranges of the
intensities in the regions to which are to be segmented. In previous techniques the flow
of the level set method for image segmentation can be depicted. A level set methods
have been extensively use in image processing and computer vision. Intensity in
homogeneity often occurs in existent world images which presents a substantial
challenge in image segmentation. MRI intensity in homogeneities can be attributes not
perfect in the RF. The result is slowly-varying shading artifact over the image that can
produce errors Therefore by minimizing this power of the level set method is able to
simultaneously segment the image and estimate the unfairness field, which can be used
for bias correction. Statistical analysis is performed by calculating probability, variance
and entropy for both the images. In this paper a variation level set framework for
segmentation and bias correction of images with concentration in homogeneities is
presented. A local clustering criterion function for the intensities in a district of each
point is established based on the local intensity clustering property, from a generally
14
established model of images with strength in homogeneities. The energy of the level set
functions represents a partition of the image field and a bias field that accounts for the
intensity in homogeneity. Segmentation and bias field estimation are therefore jointly
performed by decreasing the energy functional. This model efficiently utilizes the local
image information and therefore able to simultaneously divide and bias correct the
images with intensity in homogeneity. Experimental outcome prove the probability of
image (pixels) varies for both the images and is concentrated for certain pixel values,
indicating refinement of image. Entropy diversity of bias corrected image as compared
to original image is higher which indicates more raw data is available in the has
changed entropy is maximized and the source provides the greatest possible average
information, which in turn helps in obtaining better image segmentation results. Lastly
reduction in value of variance for individual pixel shows less deviation of intensity
value to that of mean value for the cluster of pixel in a given neighborhood.
Snehal O. Mundhada, et al. (2012) [28], in this article image enhancement is
among the simplest and most appealing areas of digital image processing. Image
processing technology is used by terrestrial scientists to enhance images of mars, Venus
or other planets. One of division of the image processing is the image enhancement.
Image enhancement is the improvement of digital image quality, without understanding
about the source of degradation. Image enhancement is the technique to improve the
interpretability or insight of information in images for human viewers. It is improving
the image quality so that the resultant image is better than the original image for a
detailed application. The main purpose of image enhancement is to bring out detail that
is concealed in an image or to increase contrast in a low contrast image. Whenever an
image is converted from one form to other such as digitizing the picture some form of
degradation occurs at output. Image enhancement is among an easy and most appealing
area of digital image processing. Basically, the idea behind enhancement techniques is
to bring out detail that is obscured. Enhancement techniques such as alpha rooting
operate on the transform domain. The transform domain open an operation on the
frequency content of the image, and therefore high frequency content such as edges and
other delicate information can easily be enhanced. However, these techniques bring
about tonal changes in the images and can also generate discarded artifacts in many
cases, as it is not possible to enhance all parts of the image in balanced manner. Image
enhancement is used for enhancing a quality of images. The applications of image
improvement are Aerial imaging, Satellite imaging, Medical imaging, Digital camera
application, Remote sensing. Most of the techniques are helpful for altering the gray
level values of individual pixels and hence the overall contrast of the entire image. But
they typically enhance the whole image in a uniform manner which in many cases
produces undesirable results. There are a range of techniques available which produce
highly balanced and visually appealing results for a diversity of images with diverse
qualities of contrast and edge information and it will produce satisfactory result.
Rajulath Banu A.K., et al. (2015) [29], in this paper author tells about the MRI
images. MRI is an extremely developed medical imaging technique used to generate
high quality images of human body and different parts. It gives a lot of information to
analyze the diseases. Magnetic resonance imaging gives detailed information of living
tissues in the human body. This informations can be used for developing the tissue
deformities. We can also obtain detailed anatomical information to discover
abnormalities. In this article, author study and compare different histogram based
15
Magnetic Resonance Imaging (MRI) enhancement techniques. Image enhancement is

dealing out an image in order to make it more suitable for certain applications. The
main purpose of the image enhancement is to carry out the concealed part in an image
or to enhance the low contrast images. The quality of the image gets improved by
contrast manipulation. Histogram Equalization is the traditional contrast enhancement
method, which has good performance for MR images. It is used to get better the visual
effects and the clarity of image or to create the original image more conducive for
computer to procedure. Contrast enhancement altering the pixels intensity of the input
image to develop maximum possible bins. We require to study and review the
dissimilar image contrast enhancement techniques for the reason that contrast losses the
brightness in enhancement of image. In view of this fact, the combination of global and
local contrast enhancement methods may enhance the contrast of image with preserving
its brightness. Normally, MRI images are having low contrast. It may be difficult for
analyzing because of lack of detailed information. Contrast of MR images can be
enlarged by number of ways in image processing. Histogram based techniques are used
to enhance all types of medical images. This paper explains comparative analysis of
unusual enhancement methods based on histogram processing. Depending ahead
original nature of an image, appropriate method should be used for enhancement from
mentioned above. There are two criteria that are used extensively to calculate image
quality the visibility of artifacts and the preservation of edge details. It is usually
measured by illustration inspection. The difference between the original image and its
low contrast image shows the noise removed by the algorithm, which is called as
method noise or residual image. Some techniques for MR image enhancement and their
effectiveness are analyzed.
R. Somas Kandan, et al. (2015) [30], here, author discuss the Panoramic
radiography also called Panoramic X ray is a two dimensional (2-D) dental X-ray that
captures the intact mouth in a single image, together with the teeth, upper & lower jaws,
surrounding structures & tissues. Unlike other X-rays, where the film is located inside
the patient's mouth, the panoramic film is enclosed in a machine that moves in the
region of the patients head. Large region of the jaws that cannot be seen with intraoral
radiography can be seen with panoramic. These X-ray images undergo from poor
contrast on the basis of the luminance non-linearity introduced by medical imaging
techniques. Hence an automatic system of image enhancement of panoramic X-ray
image is necessary. A novel technique of enhancing the quality of these images is
determined depend on Gamma Correction using GLCM (Gray level co-occurrence
matrix) by minimizing the homogeneity feature and by generating a global Gamma
value based on cumulative histogram without any knowledge of the imaging device.
The performance of both the way is compared in conditions of metrics like MSE, PSNR
& AMBE and it is explained that Gamma Correction based on GLCM by minimizing
the homogeneity yields enhanced results in terms of low MSE, high PSNR and low
AMBE. In this approach, the image enhancement is based on Gamma Correction based
on minimizing the homogeneity in GLCM matrix and Gamma Correction based on the
computing cumulative histogram. The experimental results and performance analysis
explain the effectiveness of Gamma Correction method based on GLCM matrix
compared to the Gamma Correction method based on the cumulative histogram as it
produces low MSE, high PSNR and low AMBE.
16
Pooja Kaushik, et al. (2012) [31], in this paper author basically work on
estimation for image quality is a traditional need. The conformist method for measuring
quality of image is MSE & PSNR. In this article we compared the different image
enhancement techniques by using their quality terms (MSE & PSNR) & proposed a
new generated enhancement technique. This technique gives enhanced result than
different techniques and their PSNR value is high & MSE is low. The experimental
results give that the proposed enhancement method gives better results.
17
CHAPTER 3
ENHANCEMENT TECHNIQUES
3.1 ENHANCEMENT
The concepts of enhancement of images are very important in day to day life. Some
low contrast images have certain characteristics makes it not easy to use traditional
methods to enhanced it. The principal objective of image enhancement is performing a
given image so that the result is more suitable than the original image for a specific
application. It accentuates or sharpens figure features such as edges, boundaries, or
contrast to make a graphic display more helpful for display and analysis. The
enhancement doesn't enhance the inherent information content of the data, but it
increases the dynamic range of the selected features so that they can be generated
easily.
Image Enhancement
Point operation
Spatial operation
Transform operation
-Contrast stretching
- Noise smoothing
- Linear filtering
-Noise clipping
- Median filtering
-Root Filtering
-Window Slicing
- LP, HP & Bp filtering
-Histogram modeling
Pseudocoloring
-False coloring
-Pseudocoloring
-Homomorphic Filtering
-Zooming
Fig 3.1: categorization of an Image enhancement
The supreme difficulty in image enhancement is quantifying the criterion for

enhancement and, therefore, a large number of image enhancement techniques are
experiential and require interactive procedures to obtain satisfactory results. Image
enhancement methods can be depend on either spatial or frequency domain techniques.
3.1.1 SPATIAL DOMAIN ENHANCEMENT METHODS:
Spatial domain methods are based to an image plane itself and they are based on direct
manipulation of pixels in an image. The operation can be generated as g(x, y) = T [f(x,
y)], where g is the output, f is the input image and T is a function on f defined over
some neighborhood of (x, y). According to the operations on the image pixels, it can be
further separated into 2 categories: Point operations and spatial operations (including
linear and non-linear operations).
18
3.1.2 FREQUENCY DOMAIN ENHANCEMENT METHODS:

These techniques enhance an image f(x, y) by convoluting the image with a linear,
position Invariant operator. The 2D convolution is explained in frequency domain with
DFT.
Spatial domain: g(x, y) = f(x, y)*h(x, y)
Frequency domain: G(u, v) = F(u, v) H(u, v)
f(x, y)
h(x, y)
g(x, y)
(a)
F(u, v)
H(u, v)
G(u, v)
(b)
Fig 3.2: Diagram for (a) Spatial Domain and (b) Frequency Domain
Enhancement is an alteration of an image to alter contact on the viewer. Generally
enhancement distorts the original digital values; therefore enhancement is not complete
until the restoration processes are completed.
Contrast Enhancement: There is a strong way of contrast ratio on resolving
power and detection capacity of images. Techniques for improving image
contrast are among the most generally used enhancement processes. The
sensitivity range of any remote sensing detector is premeditated to record a wide
range of terrain brightness from black basalt plateaus to white Sea beds under a
wide collection of lighting conditions. Few individual scenes have a brightness
range that generated the full sensitivity range of these detectors. To create an
image with the optimum contrast ratio, it is important to utilize the entire
brightness range of the display medium, which is generally film. Histogram of
the number of pixels that communicate to each DN of an image with no
modifications of original DNs. The middle 92 percent of the histogram has a
series of DNs from 49 to 106, which useful only 23 percent of the available
brightness range [(106 - 49)/25 = 22.3%]. This limited series of brightness
values balance sheet for the low contrast ratio of the original image.
Linear Contrast Stretch: The easy way of contrast enhancement is called a
linear contrast stretch. A DN rate in the low end of the original histogram is
declare to extreme black and a value at the high end is assigned to extreme
19
white. In this example the poorer 4 percent of the pixels (DN < 49) are declared
to black, or DN = 0, and the upper 4 percent (DN > 106) are declared to white,
or DN = 255. The remaining pixel values are distributed linearly between these
extremes. The better contrast ratio of the picture with linear contrast stretch will
enhance different features on the map. Mainly of the image processing software
displays an image only after linear stretching by default. For color images, the
individual bands were stretched previous to being combined in color. The linear
contrast stretch greatly improves the contrast of most of the unique brightness
values, but there is a loss of contrast at the excessive high and low end of DN
values. In evaluation to the overall contrast improvement, these contrast losses
at brightness extremes are acceptable unless one is critically interested in these
basics of the scene.
Nonlinear Contrast Stretch: Nonlinear contrast enhancement is prepared in
different ways. A uniform distribution stretch (or histogram equalization) in
which the original histogram has been regenerated to produce a uniform
population density of pixels beside the horizontal DN axis. This stretch trying
the greatest contrast enhancement to the most populated range or brightness
values in the original image. In the middle range of brightness values are
preferentially stretched, which results in higher contrast. The uniform
distribution stretch powerfully saturates brightness values at the sparsely
populated light and dark tails of the original histogram. The resulting defeat of
contrast in the light and dark ranges is similar to that in the linear contrast
stretch but not as severe.
Gaussian Stretch: it is a nonlinear stretch that improves contrast within the tails
of the histogram. This stretch hysterics the original histogram to a normal
distribution curve between the 0 to 255 limits, which enhanced contrast in the
light and dark ranges of the images. This enhancement occurs at the rate of
contrast in the middle gray image. A significant step in the method of contrast
enhancement is for the user to examine the original histogram and determine the
elements of the scene that are of highest interest. The user then chooses the
optimum stretch for his requirements. Experienced operators of image
processing way bypass the histogram examination stage and regulate the
brightness and contrast of images that are displayed on a screen. For a number
of scenes a variety of stretched images are required to display fully the original
data. It should be famous that contrast enhancement should not be done until
other processing is completed, since the stretching distorts the original values of
the pixels.
3.2 EXISTING ENHANCEMENT TECHNIQUES

Image enhancement is a procedure which is usually used to enhance the quality of an
image. Image enhancement method consists of a collection of techniques that look for
to improve the visual appearance of an image or to change the image to a form
enhanced suited for analysis by a human or machine. The major objective of image
enhancement is to adjust attributes of an image to make it additional suitable for a given
task and specific observer. These enhancement operations are performed in regulate to
alter the image brightness, contrast or the allocation of the gray levels. There are an
20
amount of techniques which is used to improve the quality of an image. These existing
algorithms are:
3.2.1 AVERAGE SMOOTHING:
Smoothing is to decrease noise and improve the visual quality of the image [32].
Average smoothing can be used to diminish the image noise and suppress the noise.
This technique defined to arrangement with transitions in the image pixel value of the
images. This is the necessary form of image smoothing. A filter can be useful to reduce
the amount of unwanted noise in an exacting image. An additional type of filter can be
used to reverse the effects of blurring on an exacting picture. Nonlinear filter have
rather different behavior compared to linear filters. For nonlinear filters, the filter
production or response of the filter does not mind the principles outlined before,
particularly scaling and shift invariance. Furthermore, a nonlinear filter can create
results that vary in a non-intuitive mode. In this technique the neighborhood pixel is
equal, the general slanted model make point (x, y) as the center, the 33 neighborhood
of the point are involved to decide the new image (x, y) points pixel value, and the
coefficient is 1.
3.2.2 HOMOMORPHIC FILTERING:
Homomorphic filtering is a universal technique for signal and image processing,
connecting a nonlinear mapping to a dissimilar domain in which linear filter techniques
are useful, followed by mapping back to the original field. Homomorphic filtering is a
technique is a frequency field operation applied on low contrast images. Homomorphic
filter is used for image enhancement. It concurrently normalizes the dullness across an
image and increases contrast. Here homomorphic filtering is used to take away
multiplicative noise. Illumination and reflectance are not divisible, but their
approximate locations in the frequency field may be located. Since illumination and
reflectance join multiplicatively, the components are made additive by delightful the
logarithm of the image intensity, so that these multiplicative mechanism of the image
can be divided linearly in the frequency domain. Illumination variations can be
consideration of as a multiplicative noise, and can be reduced by filtering in the record
field. To construct the illumination of an image more even, the high-frequency
components are enlarged and low-frequency components are slow down, for the reason
that the high-frequency components are unspecified to represent mostly the reflectance
in the scene (the amount of light reflected off the object in the scene), while the lowfrequency components are unspecified to represent mostly the illumination in the scene.
That is, high-pass filtering is way to suppress low frequencies and amplify high
frequencies, in the log-intensity field.
The homomorphic filtering designed for high pass filter; high pass filter is
removing the low frequency components. After that the image looks like enhanced
which are mostly used improve the image uneven illumination.
3.2.3 HISTOGRAM EQUALIZATION:
Histogram is one of the significant features which are very related to image
enhancement. The histogram does not only gives us a general summary on some useful
image statistics (e.g. mode, and dynamic range of an image), but it also can be used to
predict the look and intensity characteristic of an image [33]. The objective is to map an
input image to an output. Image is basically based on histogram is uniform after the
mapping.
21
Histogram equalization is a technique in image processing of contrast

adjustment with the image's histogram. Histogram equalization can improve the
contrast of the image but the area looking too bright enhancement and used for contrast
enhancement in a variety of application due to its simple function and effectiveness.
Histogram indicates that most of the pixels are too dark only the minority of pixel is
light. This technique work by flattening the histogram and stretching the dynamic range
of the gray level by using the cumulative density purpose of the image. Histogram
equalization treated from the problem of being poorly suitable for retaining local detail
due to its global treatment of the image. It is also common that the equalization will
over enhance the image, resulting in an unwanted loss of visual data, of quality and of
intensity scale. There is a still one difficulty of the histogram equalization is that
brightness of an image is changed after the histogram equalization, therefore it is not
suitable for consumer electronic products, where conserving the original brightness and
enhancing contrast are essential to avoid annoying artifact. Histogram equalization
often produces unrealistic effects in photographs; still it is very useful for scientific
images like thermal, satellite or x-ray images, frequently the same class of images to
which one would relate false-color. Also histogram equalization can generate
undesirable effects (like visible image gradient) when useful to images with low color
depth.
Fig 3.3: Diagram of histogram equalization

3.2.4 ADAPTIVE HISTOGRAM EQUALIZATION:
Adaptive histogram equalization (AHE) is a machine image processing method used to
improve lightning in images. It differs from ordinary histogram equalization in the
value that the adaptive method computes several histograms, each equivalent to a
distinct section of the image, and uses them to reorganize the lightness values of the
image. It is therefore fitting for improving the local contrast. Adaptive histogram
equalization is an excellent contrast enhancement technique for both natural images and
medical and other firstly no visual images [34]. In medical imaging its automatic
operation and helpful presentation of all contrast available in the image data make it a
competitor to the standard contrast enhancement method, interactive intensity
windowing.
22
3.2.5 GAMMA CORRECTION:

The Gamma Correction is defined for a non-linear multiplication-based conversion and
it overcomes the effect of light. It explains the involvement between a pixels numerical
value and its actual luminance. That has been used to gray-value image study. For
alternating the pixel value to improve the image, the projection relationship between
pixel value and Gamma value is established according to the internal map [1]. Gamma,
Gamma Correction or Nonlinear Encoding is an operation on images that adjust
luminance for show on a computer monitor. Computer graphic area can process images
in each linear or Nonlinear Gamma. Inaccurate use of Gamma will result in poor image
quality and decrease predictability of colors and textures. Therefore, the correct use of
Gamma is very significant for a computer graphics workflow.
Gamma Correction is an included printer feature that permits users to adjust the
lightness/darkness position of their prints. The amount of correction is precise by a
single value ranging from 0.0 to 10.0. Gamma Correction may be particular on both a
printer default and user-specific basis across the network and on a printer default basis
through the printers front panel. Gamma Correction allows users to better match the
intensity of their prints to what they see on their computer screen (CRT). For case, an
image that appears just fine on the CRT might print out darker on the printer. This is as
the printer Gamma (the characteristic traversal from dark to light) is different from
that of the monitor.
3.2.5.1 THE GAMMA CURVES

To fix this problem, the user can select a Gamma curve to be useful to the image
before printing that will lighten or darken the overall tone of the image without moving
the dynamic range. The shape of the Gamma curve is determined by a number ranging
from 0.0 to 10.0 identified as the Gamma value. Fig 3.4 tells numerous Gamma
curves indicating the effect that the Gamma value has on the form of the Gamma
curves.
Fig 3.4: Gamma Curves
23
3.2.5.2 SETTING GAMMA VALUES

The amount and direction of Gamma correction is specified by a single number as
follows:
Gamma Values
0.0
darker
1.0
no change
10.0
lighter
Fig 3.5: Draw to setting Gamma Values

The printer Gamma location can be adjusted two ways. First, the printer default Gamma
value may be set with the printers front panel. Second, users may locate their
individual Gamma value favorite on the printer by sending a text file to Logical Device
0.
A big deal of work has been devoted to diminish the effect of uneven
illumination on images, like that average smoothing, homomorphic filtering, histogram
equalization (HE), Adaptive histogram equalization and Gamma Correction technique.
Compared to the above methods, Gamma Correction method has some advantages to
overcome the effects of light. A proper estimation of Gamma value enhances the
contrast of the images. Gamma Correction method has a number of compensation to
overcome the effect of light. Gamma Correction controls overall brightness of an
image. Thats why I consider Gamma Correction technique in my thesis to enhance the
image and get more information than the original image.
3.3 SEGMENTATION
THRESHOLD VALUE: Threshold value is basically used to find from the
histogram. We can use the maxima (peaks) of the histogram to establish a
segmentation threshold. This threshold t may be halfway between two peaks
with gray values p1 and p2:
t = (p1 + p2) / 2
It may be the gray value at the minimum between the two peaks.
Fig 3.6: Threshold value concept
24
CHAPTER 4
PROPOSED METHOD
4.1 PROPOSED WORK
In this section, we have explained a proposed method work flow for image
enhancement using Gamma Correction technique for the radiography images. Image
enhancement is basically used to improve the quality and preserve the brightness of an
image. The algorithm is fundamentally explaining the low contrast part of the image to
reach the high contrast. When the low contrast part of the image is enhanced in high
contrast then resultant image will give better information. The steps of presented
algorithm are:
BEGIN
Step 1: Firstly, initialize a low contrast radiography image. Then apply preprocessing
technique.
Step 2: Plotting histogram of an original image. After that we will find threshold value
of the image.
Step 3: With the help of segmentation, we will segment an image into low contrast and
high contrast on the basis of their threshold value.
Step 4: Apply high Gamma values only on the low contrast of an image.
Step 5: Show resulting output image and histogram of an image.
Step 6: Calculate mean, standard deviation, variance and entropy for the resulted image
and compare with Gamma Correction technique.
END
The flow chart of the proposed work of our thesis is shown in fig 4.1. We follow
these steps very carefully.
25
4.2 FLOW CHART OF THE PROPOSED WORK

LOW CONTRAST RADIOGRAPHY
ORIGINAL IMAGE
PREPROCESSING OF AN IMAGE
IMAGE HISTOGRAM
SEGMENTATION OF AN IMAGE
LOW CONTRAST IMAGE
HIGH CONTRAST IMAGE
APPLY GAMMA CORRECTION
GAMMA CORRECTED LOW CONTRAST

IMAGE
JOIN OF AN IMAGE
JOINED ENHANCED IMAGE
ANALYSIS OF HISTOGRAM OF AN
IMAGE
Fig 4.1: Proposed Flow Diagram for Image Enhancement
26
First, we consider the original image to further operation.

PREPROCESSING OF AN IMAGE: Image pre-processing can extensively
increase the regularity of an optical inspection. A number of filter operations
which intensify or reduce convinced image details enable an easier or sooner
evaluation. Users are able to optimize a camera image with just a few clicks.
Preprocessing is the step in which we use to reduce the noise, unwanted artifacts
using any filter. Pre-processing means we remove the entire noisy element on an
image using Weiner filter etc.
HISTOGRAM OF AN IMAGE: When the preprocessing is done then we
make the histogram of the original image to analyze the result.
SEGMENTATION OF AN IMAGE: Segmentation is a process in which to
segment an image into two parts low contrast and high contrast on the basis of
their threshold concept.
LOW CONTRAST IMAGE: Left side or white side of an image has low
contrast which is divided on the basis of their threshold value.
HIGH CONTRAST IMAGE: Right side or black side of an image has high
contrast which is divided on the basis of their threshold value.
APPLY GAMMA CORRECTION: Gamma correction is a way to control the
brightness and enhance the quality of an image.
JOIN OF AN IMAGE: After performing operation we apply join operation to
combine both the images Gamma Corrected low contrast and high contrast
image.
27
CHAPTER 5
IMPLEMENTATION AND EXPERIMENTAL RESULTS
For implementation we have to take the radiography images. In this we have to

take 4 images for performing the experiments. In this thesis to improve contrast and
preserve brightness of the image from the proposed method.
1. Image (a)
3. Image (c)
2. Image (b)
4. Image (d)
Fig. 5.1: 1, 2, 3, 4 are the four original radiography images
5.1 IMPLEMENTATION
Firstly, we take the image (a) for the implementation to get that enhanced image in
MATLAB. At the first, the original image of the histogram will be generated. Using
histogram we can find the threshold value to divide the image. After that local Gamma
technique will be applied and get the histogram of the Gamma Corrected image. Then,
28
the proposed method will be applied only on the low contrast of the image and generate
the histogram of the proposed method. When we get both the result then analyze with
the help of histogram.
Fig 5.1.1: Original Image (a)
Fig 5.1.2: Histogram of an original Image (a)

As we have seen in figure 5.1.1, this is the original radiography image which is used to
improve the contrast and preserve brightness.
As we above seen figure 5.1.2, shown the histogram of the image (a) radiography
image. By which we can see the difference between resultant images. Histogram is a
bar chart, whose X-axis represents the tonal scale (black at the left and white at the
right), and Y-axis explain the number of pixels in an image in a definite area of the
tonal scale.
29
Fig 5.1.3: Gamma Corrected Image (a)
Fig 5.1.4: Histogram of Gamma Corrected Image (a)

As we above seen in fig 5.1.3, this is Gamma Corrected image which is get when we
apply Gamma Correction to the original image. In the traditional Gamma Correction
algorithm get better the low contrast images. In fig 5.1.4 this is the histogram of the
Gamma Corrected Image. By which we use this to find out the results and compare with
the proposed method.
30
Fig 5.1.5: Proposed Method Image (a)
Fig 5.1.6: Histogram of Proposed Method Image (a)

In Fig 5.1.5 the proposed method get the result where low contrast radiography image
regions increases. Algorithm basically enhances the contrast and gets the visual quality
of radiography images that captured under insufficient lighting conditions. It enhances
the quality in terms of their factors which is mean, standard deviation, variance and
entropy. In Fig 5.1.6 this is the histogram of proposed method which is like to locate
out the results comparison.
31
Fig 5.1.7: Snapshot of the Gamma Corrected Image (a)
32
Fig 5.1.8: Snapshot of Proposed Method Image (a)
33
Table 5.1: Showing mean, standard deviation, variance and entropy for image (a)
(T=22)
Technique
Mean
Standard
Deviation
Variance
Entropy
Gamma Correction
51.7882
62.1971
85.9247
6.1178
Proposed Method
52.5420
65.1953
86.1976
6.2262
The Table 5.1 explains the result of the proposed method will be compared with the
traditional method on the basis of performance measurement. A mean is factor can be
used to evaluate the image contrast. Over the region what is the contrast value after the
reconstruction of image. Standard deviation is the term which is used to provides
smaller quantity rounds off noise and greater computational efficiency. Entropy can be
used to estimate the clarity of the image, the greater its value, the more clear that the
image. A variance factor can be used to analyze of the spread between numbers in data
set. The variance calculates how far each no. in the set is from mean. If the value of
variance is greater. It means the quality or pixel intensity is clearer and blurriness of
image is also reduced.
100
Gamma Correction
Proposed Method
90
80
70
60
50
40
30
20
10
0
Mean
Standard
Deviation
Variance
Entropy
Fig 5.1.9: Bar graph of Table 5.1

This bar diagram is basically shows the comparison between the traditional Gamma
Correction technique and proposed method for the radiography images.
34
Fig 5.1.10: Snapshot of Gamma Corrected Image
Fig 5.1.11: Snapshot of Proposed Method Image
Table 5.2: Showing MSE and PSNR values for Image (a)
Techniques
MSE
PSNR
Gamma Correction
49.99
50.03
Proposed Method
49.43
49.05
In Table 5.2 explain the performance of the proposed method is compared with Gamma
Correction technique by means PSNR and MSE measure. PSNR is stand for peak signal
to noise ratio and MSE known for mean square error. PSNR and MSE value is measure
the performance of the image. If the value of PSNR is high then generally indicates that
the reconstruction of higher quality of image. MSE is used to measures the average of
the squares of the errors, i.e., the separation between the estimator and what is
estimated. MSE is a risk function, resultant to the expected data of the squared error
loss or quadratic loss.
Now, Consider the second image (b) are taken for the implementation. Firstly,
we find the histogram of the image. From the histogram we find the threshold value for
segmenting an image into two parts in low contrast and high contrast. The threshold
value for image (b) is 116 as we seen in fig 5.1.12 and fig 5.1.13.
35
Fig 5.1.12: Original Image (b)
Fig 5.1.13: Histogram of original Image (b)

As we have seen in figure 5.1.12, it shows original radiography image which is used to
improve the contrast and preserve brightness. As we above seen figure 5.1.13, this is the
histogram of the original radiography image. By which we can see the difference
between resultant images. A histogram is a graphical depiction of the distribution of
numerical data.
36
Fig 5.1.14: Gamma Corrected Image (b)
Fig 5.1.15: Histogram of Gamma Corrected Image (b)

As we above seen in fig 5.1.14, it is Gamma Corrected image which is generated when
we apply Gamma Correction to the original image. In the traditional Gamma Correction
method improve the low contrast images. In fig 5.1.15 shows the histogram of the
Gamma Corrected Image. By which we use this to find out the results and compare with
the proposed method.
37
Fig 5.1.16: Proposed Method Image (b)
Fig 5.1.17: Histogram of Proposed Method Image (b)

Fig 5.1.16 shows the proposed method shows the result where low contrast radiography
image regions increases. Algorithm basically improves the contrast and enhanced the
visual quality of radiography images that captured under insufficient lighting
conditions. It develops the quality in terms of their features which is mean, standard
deviation, variance and entropy. In Fig 5.1.17 shows the histogram of proposed method
which is used to find out the results comparison.
38
Fig 5.1.18: Snapshot of Gamma Corrected and Proposed Method Image (b)
39
Table 5.3: Showing mean, standard deviation, variance and entropy for image (b)
(T=116)
Technique
Mean
Standard
Deviation
Variance
Entropy
Gamma correction
87.2023
85.2363
117.2273
5.9099
Presented Algorithm
88.2606
85.7390
117.8063
6.1392
The Table 5.3 shows the result of the proposed method will be evaluated with the
traditional method on the basis of performance measurement. A mean is factors can be
re-enactment of image. Standard deviation provides smaller quantity rounds off noise
and greater computational efficiency. Entropy can express the ability of small details. It
can also be used to evaluate the clarity of the image, the greater its value, the more
understandable that the image. A variance is function can be used to measurement of
the spread between information in data set. The variance measure how far each no. in
the set is from mean. If the value of variance is greater. It means the quality or pixel
intensity is clearer and blurriness of image is also reduced.
140
Gamma Correction
120
Proposed Method
100
80
60
40
20
0
Mean
Standard
deviation
Variance
Entropy

This bar chart shows the comparison between the traditional Gamma Correction
technique and proposed method for the radiography images.
40
Table 5.4: Showing MSE and PSNR value for image (b)
Technique
MSE
PSNR
Gamma Correction
51.65
51.48
Proposed Method
51.11
51.49
In Table 5.4 shows the performance of the proposed method is compared with Gamma
Correction technique by means by PSNR and MSE measure. PSNR and MSE value is
basically measure the performance of the image. Higher PSNR generally indicates that
the reconstruction of higher quality. MSE is used to measures the average of the squares
of the "errors", that is, the difference between the estimator and what is estimated. MSE
is a risk factor, corresponding to the expected value of the squared error loss or
quadratic loss.
Now, Consider the second image (c) are taken for the implementation. Firstly,
we find the histogram of the image. From the histogram we find the threshold value for
segmenting an image into two parts in low contrast and high contrast. The threshold
value for image (c) is 48 as we seen in Fig 5.1.22 and Fig 5.1.23.
41
Fig 5.1.22: Original Image (c)
Fig 5.1.23: Histogram of Original Image (c)

In fig 5.1.24 and fig 5.1.25 shows the Gamma Corrected image and histogram of the
Gamma Corrected image. Gamma Correction is basically preserves the brightness of
the image. Through the number of histogram we plot the number of pixels for each
tonal value.
42
Fig 5.1.24: Gamma Corrected Image (c)
Fig 5.1.25: Histogram of Gamma Corrected Image (c)
In Fig 5.1.26 shows the proposed method and in Fig 5.1.27 shows the histogram of the
Proposed method. It shows the overall enhance quality of image.
43
Fig 5.1.26: Proposed Method Image (c)
Fig 5.1.27: Histogram of Proposed Method Image (c)
44
Fig 5.1.28: Snapshot of Gamma Corrected Image (c)
45
Fig 5.1.29: Snapshot of Proposed Method Image (c)
46
Table 5.5: Showing mean, standard deviation, variance and entropy for image (c)
(T=48)
Technique
Mean
Standard
Deviation
Variance
Entropy
Gamma Correction
83.9758
57.4357
90.4844
6.6405
Proposed Method
88.5139
59.1995
91.7294
6.7134
Table 5.5 shows the result of the Gamma Corrected technique and proposed method to
calculate the value of Mean, Standard deviation, Variance and Entropy.
160
Gamma Correction
140
Proposed Method
120
100
80
60
40
20
0
Mean
Standard
Deviation
Variance
Entropy
This bar diagram shows the comparison between the traditional Gamma Correction
technique and proposed method for the radiography images.
47
Table 5.6: Showing MSE and PSNR values for Image (c)
Technique
MSE
PSNR
Gamma Correction
65.31
64.83
Proposed Method
64.92
64.28
In Table 5.6 shows the performance of the proposed method is compared with Gamma
Correction technique by means by PSNR and MSE measure. PSNR and MSE value is
basically measure the performance of the image. The Mean Square Error (MSE) and the
Peak Signal to Noise Ratio (PSNR) are the two error measurement used to compare
image compression quality. The MSE represents the increasing squared error between
the compressed and the original image, whereas PSNR represents a measure of the peak
error.
Here, we consider the fourth image (d) to find the result to enhance the image in
terms of mean, standard deviation, variance and entropy.
48
Fig 5.1.33: Original Image (d)
Fig 5.1.34: Histogram of original Image (d)
49
Fig 5.1.35: Gamma Corrected Image (d)
Fig 5.1.36: Histogram of Gamma Corrected Image (d)
50
Fig 5.1.37: Proposed Method Image (d)
Fig 5.1.38: Histogram of Proposed Method Image (d)
51
Fig 5.1.39: Snapshot of Gamma Corrected Image (d)
52
Fig 5.1.40: Snapshot of Proposed Method Image (d)
53
Table 5.7: Showing mean, standard deviation, variance and entropy for image (d)
(T=16)
Technique
Mean
Standard
Deviation
Variance
Entropy
Gamma Correction
101.5631
67.7217
132.6552
7.1110
Proposed Method
106.6609
70.9356
134.2920
7.2499
The Table 5.7 shows the result of the proposed method will be evaluated with the
traditional method on the basis of performance measurement. A mean factors can be
re-enactment of image. Standard deviation provides smaller quantity rounds off noise
and greater computational efficiency. Entropy can express the ability of small details. It
can also be used to evaluate the clarity of the image, the greater its value, the more
understandable that the image. A variance is function can be used to measurement of
the spread between information in data set. The variance measure how far each no. in
the set is from mean. If the value of variance is greater. It means the quality or pixel
intensity is clearer and blurriness of image is also reduced.
160
Gamma Correction
140
Proposed Method
120
100
80
60
40
20
0
Mean
Standard
Deviation
Variance
Entropy
54
Table 5.8: Showing MSE and PSNR values for Image (d)
Technique
MSE
PSNR
Gamma correction
59.14
59.12
Proposed Method
58.71
58.71
In Table 5.8 explain the performance of the proposed method is compared with Gamma
Correction technique by means by PSNR and MSE measure. The term peak signal-tonoise ratio (PSNR) is a way for the ratio between the maximum possible value (power)
of a signal and the power of distorting noise that harm the quality of its image. PSNR
and MSE value is basically measure the performance of the image. The Mean Square
Error (MSE) and the Peak Signal to Noise Ratio (PSNR) are the two error measurement
used to compare image quality. The MSE represents the cumulative squared error
between the condensed and the original image, whereas PSNR represents a measure of
the peak error.
From the above result analysis, the proposed method is provided much better
result than the Gamma Correction technique. In proposed method, we apply high
Gamma value only on the low contrast image and result will be calculated in terms of
Image factors i.e. Mean, Standard Deviation, Entropy and Variance. When we compare
the result on the basis of their image factor proposed method is good then the Gamma
Correction technique. After that we calculate PSNR and MSE value. It also gives
improved result then the Gamma Correction technique.
55
CHAPTER 6
CONCLUSION AND FUTURE ADVANCEMENT
6.1 CONCLUSION
Image Enhancement is an important way to modify or improve the image to the
better form. By which we can get better result and more informative for the human
viewer. The aim of image enhancement is to or to provide better input for other
enhanced the interpretability or perception of information in images for human viewing
automated image processing techniques.
Here, the proposed method can be used to reduce the loss of low contrast in restricted
regions by applying high gamma values to that portion. The main aim behind the
proposed method is providing the higher level of brightness preservation and contrast
improvement to keep away from dimmed images and unnatural improvement. The
concept of image enhancement is basically to improve the digital image quality.
From the result analysis, the proposed method provides much better result than
the Gamma Correction technique. In proposed method, we apply high Gamma value
only on the low contrast image and result will be calculated in terms of image factors
i.e. Mean, Standard Deviation, Entropy and Variance. When we compare the result on
the basis of their image factor the proposed method gives better results than the Gamma
Correction technique. Then we calculate PSNR and MSE value. After that it also gives
improved result then the Gamma Correction technique.
6.2 FUTURE ADVANCEMENT

Further, Proposed algorithm can be used other type of images like satellite,
astrophotography and colored images to get better result.
56
REFERENCES
1. Kaur K., Kanwal N., Bhullar J. S. 2014. A Technique for Enhancement of Gray Image
using Local Gamma Correction. International Journal of Computer Applications (0975 8887),
Volume 105 No. 5, pp. 36-39.
2. Rani S., Kumar M. 2014. Contrast Enhancement using Improved Adaptive Gamma
Correction with Weighting Distribution Technique. International Journal of Computer
Applications (0975 8887), Volume 101 No.11, pp. 47-53.
3. Yao Wang. Contrast Enhancement Image Processing. Polytechnic University, Brooklyn, NY

11201, pp. 1-46.
4. Daniel P., Raju R., Neelima G. 2012. Image Segmentation by using Histogram Thresholding.
IJCSET, Vol 2, Issue 1,776-779, pp. 776-779.
5. Al-amri S. S., Kalyankar N.V. and Khamitkar S.D 2010. Image Segmentation by
Using Threshod Techniques. Journal of computing, volume 2, issues 1, pp. 83-86.
6.
Image
segmentation
methods
are
taken
http://www.bioss.ac.uk/people/chris/ch4.pdf, pp. 1-25.
from
pdf
file,
7. Gayathri S., Mohanapriya N., Kalaavathi B. 2014. Efficient Contrast Enhancement using
Gamma Correction with Multilevel Thresholding and Probability Based Entropy. IJRET:
International Journal of Research in Engineering and Technology, Volume: 03 Special Issue:
07, pp. 781-786.
8. Rathore V. S., Dubey V., Varshney S. 2013. A Study to Improve the Quality of Image
Enhancement Based on Different Filtering Design Techniques. IJRET: International Journal of
Research in Engineering and Technology, Volume: 02 Issue: 08, pp. 11-16.
9. Hassanpour S. and Asadi Amiri S. 2011. Image Quality Enhancement using Pixel-wise
Gamma Correction via SVM classifier. Department of Computer Engineering, University of
Shahrood Technology, Shahrood, Iran, Vol. 24, No. 4, pp- 301-311.
10. Dass R., Priyanka, Devi S. 2012. Image Segmentation Techniques, IJECT, Vol. 3, Issue 1,
pp. 66-70.
11. Sharma A. K., Bala A. 2013. Marker Based watershed transformation for image processing.
International Journal of Computer Science Engineering and Information Technology Research
(IJCSEITR), Vol. 3, Issue 4, pp. 187-191.
12. Vaid K., Dua B. 2014. Image Enhancement of Aerial Images Using Adaptive Gamma
Correction: A Review. IJEDR | Volume 2, Issue 2, pp. 1617-1619.
13. Kaur S., Walia N. K. 2014. Image Enhancement in Case of Uneven Illumination using
Variable Thresholding Method. IJSRD - International Journal for Scientific Research &
Development Vol. 2, Issue 09, 201, pp. 160-164.
57
14. Bedi S.S., Khandelwal R. 2013. Various Image Enhancement Techniques- A Critical
Review. International Journal of Advanced Research in Computer and Communication
Engineering, Vol. 2, Issue 3, pp. 1605-1609.
15. Kamboj A., Gupta A. 2012. Simulink Model Based Image Segmentation. International
Journal of Advanced Research in Computer Science and Software Engineering, Volume 2,
Issue 6, pp. 146-149.
16. Al-amri S. S., Kalyankar N.V. and Khamitkar S.D. 2010. Image Segmentation by Using
Thershod Techniques. Journal of Computing, Volume 2, ISSUE 5, ISSN 2151-9617, pp. 83-87.
17. Chandel G. S., Kumar R., Khare D., Verma S. 2012. Analysis of Image Segmentation
Algorithms Using MATLAB. International Journal of Engineering Innovation & Research,
Volume 1, Issue 1, ISSN 2277 5668, pp. 51-55.
18. Gautam C., Tiwari N. 2015. Efficient Color Image Contrast Enhancement Using Range
Limited Bi-Histogram Equalization With Adaptive Gamma Correction. International
Conference on Industrial Instrumentation and Control (ICIC) College of Engineering Pune,
India, pp. 175-180.
19. Maini R., Aggarwal H. 2010. A Comprehensive Review of Image Enhancement
Techniques. Journal of computing, VOLUME 2, ISSUE 3, MARCH 2010, ISSN 2151-9617,
pp. 8-13.
20. Ponomaryov V., Herfet T., Lukin V., Smolka B. and Zlokolica V. 2012. Image and video
quality improvement techniques for emerging applications. EURASIP Journal on Advances in
Signal Processing, 2012:3, pp. 33-35.
21. Humied A. I., Fatma E.Z. Abou-Chadi, Magdy Z. Rashad. 2012. A new combined technique
for automatic contrast enhancement of digital images. Egyptian Informatics Journal 13, 2737,
pp. 27-37.
22. Karuppanagounder S. 2011. Image Contrast Enhancement based on Gamma Correction, pp.
1-5.
23. Aggarwal A., Chauhan R.S. and Kaur K., 2013. An Adaptive Image Enhancement
Technique Preserving Brightness Level Using Gamma Correction. Advance in Electronic and
Electric Engineering, pp. 1097-1108.
24. Kaur A., Singh M. 2014. Image Enhancement Using Recursive Adaptive Gamma
Correction. International Journal of Innovations in Engineering and Technology (IJIET), pp. 19.
25. Kubinger W., Vincze M., Ayromlou M. The Role of Gamma Correction in Colour Image
Processing. Institute of Flexible automation, Vienna university of technology, Gusshausstr, pp.
1-4.
26. Kotkar V. K., Gharde S. S. 2013. Review of various image contrast enhancement
techniques. International Journal of Innovative Research in Science, Engineering and
Technology, Vol. 2, Issue 7, pp. 2786-2793.
27. Konduri S. S., Thirupathaiah A. 2014. A level set Image Articulations in presence of
disagreement with MRI Scanning. International Journal of Research in Computer and
Communication Technology, Vol 3, Issue 1, pp. 1-4.
58
28. Mundhada S. O., Shandilya Prof. V. K. 2012. Image Enhancement and Its Various
Techniques. International Journal of Advanced Research in Computer Science and Software
Engineering, Volume 2, Issue 4, pp. 370-372.
29. Banu A.K. R., Ram Dr. A. R. 2015. Contrast Enhancement of MRI Images: A Review.
International Journal of Emerging Technology and Advanced Engineering, Volume 5, Issue 6.
30. Kandan R. S. and John A., Kumar S. 2015. An Improved Contrast Enhancement Approach
for Panoramic Dental X-ray Images. ARPN Journal of Engineering and Applied Sciences VOL.
10, NO. 4, pp. 99-103.
31. Kaushik P. and Sharma Y. 2012. Comparison of Different Image Enhancement Techniques
Based Upon Psnr & Mse. International Journal of Applied Engineering Research, ISSN 09734562 Vol.7 No.11, pp. 1-5.
32. Chandel R., Gupta G. 2013. Image Filtering Algorithms and Techniques: A Review.
International Journal of Advanced Research in Computer Science and Software Engineering,
Volume 3, Issue 10. pp. 198-202.
33. Kong N. S. P., Ibrahim H., and Hoo S. C. 2013. A Literature Review on Histogram
Equalization and Its Variations for Digital Image Enhancement. International Journal of
Innovation, Management and Technology, Vol. 4, No. 4, pp. 386-389.
34. Pizer S. P., Amburn E. P., Austin J. D., Cromartie R. 1987. Adaptive histogram equalization
and its variance. computer vision, graphics and image processing 39, pp. 355-368
59
APPENDIX I
Introduction to Platform
MATLAB
MATLAB is a high-level idiom and interactive surroundings for numerical
visualization, calculation and programming. Using MATLAB you can look at data,
expand algorithms, and create models and submission. The tools are built-in math
functions enable you to search multiple approaches and reach a solution quicker than
with traditional programming languages or spreadsheets such as C/C++ or Java. You
can use MATLAB for a variety of applications, together with signal processing and
interactions, image and video processing, control systems, and measurement,
computational and computational finance. More than a million scientists and engineers
in commerce and academia use MATLAB, the verbal statement of technical computing.
Key Features:
High-level language intended for numerical computation, application development
and visualization.
Interactive environment for design, problem solving and iterative exploration.
Arithmetical functions for statistics, filtering, linear algebra, optimization, numerical
integration, and solving ordinary differential equations.
Integrated graphics for visualizing information and tools for creating custom plots.
Expansion of tools for improving code maintainability excellence and maximizing
performance.
Apparatus for building application by means of custom graphical interface.
Functions for integrating MATLAB based on algorithms by means of external
applications and languages such as C, .NET, Java and Microsoft Excel.
60
PUBLICATION
61
CURRICULUM VITAE
TOSHI PATEL
Proficiency
Subjective: Software Engineering, Operating System, Data Mining, Computer Organization,
Design and Analysis of Algorithm.
Linguistics: English and Hindi
Operating Systems: Windows NT/2000/XP/7/8
Programming Languages: Basic C, Core Java, HTML, XML
Packages: Microsoft Office v2007, v2010
Technical Software: MATLAB R2009b, Open Office
Educational Qualification
M. Tech (Software
Engineering)
Address:
Sonakpur Sports Stadium,
Moradabad
Phone: +91-8533882642
Examination
M. Tech in
Software
Engineering
E-Mail:
Board
Year of
University
Passing
Shri Ram Murti

Smarak College of
Engineering &
Technology,
Bareilly (U.P.)
Dr. A.P.J. Abdul

Kalam
Technical
University,
Lucknow (U.P.)
Pursuing
80.48
Bharat Institute of
Technology,
Meerut (U.P.)
Gautam Buddha
Technical
2013
72.98
College/Institute
Percentage
toshipatel16@gmail.com
Personal Data:
Date of Birth:
4,1991
Sex:
B. Tech. in
Computer
Science
Engineering
Dec
Female
Fathers Name: Rajesh

Singh
Mothers Name: Seema
Patel
Nationality:
University,
Lucknow (U.P.)
Indian
62
12TH
10TH
Karan Public
School, Meerut
CBSE
2009
71.45
Dayawati Modi
Academy, Rampur
CBSE
2007
65.56
Marital Status:
Unmarried
Research Paper Publications
A paper titled Biomedical Image Compression by DCT is published in an International

Journal of Electrical and Electronics Engineers in Vol. 07, Issue 01, Jan- June 2015, ISSN No.
2321-2055 (E).
Hobbies:
Reading
Newspaper &
A paper titled Blur Detection Methods and Measurement in image Compression using
DWT is published in an International Journal of Electrical and Electronics Engineers in Vol.
07, Issue 01, Jan- June 2015, ISSN No. 2321-2055 (E).
Current Affair
Magazines
Strengths:
Willingness to
take higher
responsibilities.
Interact
effectively at all
levels.
Creative, Sincere
and Hard
A paper titled Mining Web Graphs for Recommendations using Collaborative filtering
and Query suggestion Techniques is published in an International Journal of Advanced
Technology in Engineering and Science in Vol. 03, Issue 01, February 2015, ISSN No. 2348
7550.
A paper titled Survey on Green Computing is published in an International Journal of
Advanced Technology in Engineering and Science in Vol. 03, Issue 01, February 2015,
ISSN No. 2348 7550.
A paper titled Statistical Performance based on different Edge Detection Techniques in
International Journal of Advance Research in Science and Engineering, Vol. 4, Issue 1
March 2015, ISSN No. 2319-8354(E).
A paper titled Analysis based on Statistics and Histogram Edge Detection Techniques in
International Journal of Advance Research in Science and Engineering, Vol. 4, Issue 1 April
2015, ISSN No. 2319-8354(E).
Working.
A paper titled Contrast Enhancement and Brightness Preservation of Radiography

Images using Gamma Correction is published in an International Journal on Recent and
Innovation Trends in Computing and Communication in Vol.3 Issue 11, ISSN No. 23218169.
Declaration
I hereby declare that all the information mentioned above is true to the best of my knowledge.
Signature:
TOSHI PATEL
63
Date:
19 December, 2015
Place:
BAREILLY

3

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

3

Uploaded by

Copyright:

Available Formats

An Enhancement of Radiography Images using Gamma Correction

1.2 RADIOGRAPHY IMAGES

An Enhancement of Radiography Images using Gamma Correction

1.3 CONTRIBUTION AND ORGANISATION OF THESIS

An Enhancement of Radiography Images using Gamma Correction

1.4 PROBLEM STATEMENT

1.6 OUTLINE OF THE THESIS

An Enhancement of Radiography Images using Gamma Correction

1.7 OPERATING SYSTEM AND TOOL USED

Installed memory (RAM): 2.00 GB

An Enhancement of Radiography Images using Gamma Correction

An Enhancement of Radiography Images using Gamma Correction

An Enhancement of Radiography Images using Gamma Correction

An Enhancement of Radiography Images using Gamma Correction

An Enhancement of Radiography Images using Gamma Correction

An Enhancement of Radiography Images using Gamma Correction

Bi-Histogram equalization method enhanced the results of basic Histogram

An Enhancement of Radiography Images using Gamma Correction

An Enhancement of Radiography Images using Gamma Correction

an important technique for enhancing the contrast of an image. HE is related to contrast

An Enhancement of Radiography Images using Gamma Correction

information received by human is in pictorial form. Histogram equalization is an

An Enhancement of Radiography Images using Gamma Correction

An Enhancement of Radiography Images using Gamma Correction

An Enhancement of Radiography Images using Gamma Correction

Magnetic Resonance Imaging (MRI) enhancement techniques. Image enhancement is

An Enhancement of Radiography Images using Gamma Correction

An Enhancement of Radiography Images using Gamma Correction

- LP, HP & Bp filtering

Fig 3.1: categorization of an Image enhancement

The supreme difficulty in image enhancement is quantifying the criterion for

3.1.1 SPATIAL DOMAIN ENHANCEMENT METHODS:

An Enhancement of Radiography Images using Gamma Correction

3.1.2 FREQUENCY DOMAIN ENHANCEMENT METHODS:

An Enhancement of Radiography Images using Gamma Correction

3.2 EXISTING ENHANCEMENT TECHNIQUES

An Enhancement of Radiography Images using Gamma Correction

An Enhancement of Radiography Images using Gamma Correction

Histogram equalization is a technique in image processing of contrast

Fig 3.3: Diagram of histogram equalization

An Enhancement of Radiography Images using Gamma Correction

3.2.5 GAMMA CORRECTION:

3.2.5.1 THE GAMMA CURVES

Fig 3.4: Gamma Curves

An Enhancement of Radiography Images using Gamma Correction

3.2.5.2 SETTING GAMMA VALUES

Fig 3.5: Draw to setting Gamma Values

Fig 3.6: Threshold value concept

An Enhancement of Radiography Images using Gamma Correction

An Enhancement of Radiography Images using Gamma Correction

4.2 FLOW CHART OF THE PROPOSED WORK

LOW CONTRAST IMAGE

HIGH CONTRAST IMAGE

APPLY GAMMA CORRECTION

GAMMA CORRECTED LOW CONTRAST

JOINED ENHANCED IMAGE

Fig 4.1: Proposed Flow Diagram for Image Enhancement

An Enhancement of Radiography Images using Gamma Correction

First, we consider the original image to further operation.

An Enhancement of Radiography Images using Gamma Correction

For implementation we have to take the radiography images. In this we have to

Fig. 5.1: 1, 2, 3, 4 are the four original radiography images

An Enhancement of Radiography Images using Gamma Correction

Fig 5.1.1: Original Image (a)