International Journal of Software Engineering Research & Practices Vol.
1, Issue 2, April, 2011
USE OF HISTOGRAM EQUALIZATION IN
IMAGE PROCESSING FOR IMAGE
ENHANCEMENT
Sapana S. Bagade Vijaya K. Shandilya
M.E ,Computer Engineering, Assistant Proffessor,Computer Science & Engg Dept,
Sipna’s C.O.E.T,Amravati, Sipna’s C.O.E.T,Amravati,
Amravati,India Amravati,India
[email protected] [email protected]Abstract— Digital Image Processing is a rapidly evolving field image. Therefore, the distinction of features in the scene can
with the growing applications in science & engineering. Image be easily performed by visualization. This will augment the
Processing holds the possibility of developing an ultimate efficiency of image classification and interpretation.
machine that could perform visual functions of all living beings. Generally, an image can be enhanced by spreading out the
The image processing is a visual task, the foremost step is to range of scene illumination. This procedure is called contrast
obtain an image i.e. image acquisition then enhancement and
stretch. If the ranges of gray values are uniformly prolonged,
finally to process.
In this paper there are details for image enhancement for the the process will be called linear contrast stretch. The
purpose of image processing. Image enhancement is basically disadvantage of the linear contrast stretch is that a number of
improving the digital image quality. Image histogram is helpful gray levels are equally assigned to the unusually appeared
in image enhancement. The histogram in the context of image gray levels as to the often appeared gray levels. This effect
processing is the operation by which the occurrences of each still causes the ambiguous distinction of the similar features.
intensity value in the image is shown and Histogram equalization To overcome the mention defect, the process of histogram
is the technique by which the dynamic range of the histogram of equalization is applied. The process tries to assigned more
an image is increased. number of gray levels to the frequency appeared gray levels.
The enhanced image obtained from the global area histogram
Keywords- Image processing, image enhancement, image equalization will cause an effect of intensity saturation in
histogram, Histogram equalization
darkness area and whiteness areas. The color image
enhancement will be obtained by encoding the color of red,
green and blue to three different spectral images.
1. Introduction
Image processing is in many cases concerned with taking one 2. Image Processing Technique
array of pixels as input and producing another array of pixels
as output which in some way represents an improvement to As has just been established, a number of factors can
the original array. The term Digital Image Processing adversely affect image quality. With the use of image
generally refers to the processing of a two-dimensional picture enhancement techniques, the difference in sensitivity between
by a digital computer i.e. altering an existing image in the image can be decreased. A number of image processing
desired manner. For example, this processing may remove techniques, in addition to enhancement techniques, can be
noise, improve the contrast of the image, remove blurring applied to improve the data usefulness. Techniques include
caused by movement of the camera during image acquisition, convolution edge detection, mathematics, filters, trend
it may correct for geometrical distortions caused by the lens. removal, and image analysis. The various image
Before going for image processing image enhancement is enhancements and image processing techniques will be
necessary. introduced in this section. Computer software programs are
We will not be considering every image processing and available, including some or all of the following programs:
enhancement technique in this section but we will see the
enhancement of image through image histogram or better way Enhancement programs make information more visible.
histogram equalization. If an image is low contrast and dark,
we wish to improve its contrast and brightness. The histogram Histogram equalization-Redistributes the intensities
equalization improve all parts of the image when the of the image of the entire range of possible intensities
(usually 256 gray-scale levels).
original image is irregularly illuminated. The enhancement
techniques are employed in order to increase the contrast of an
6
International Journal of Software Engineering Research & Practices Vol.1, Issue 2, April, 2011
Unsharp masking-Subtracts smoothed image from the input data value and y is the output data value. The non-linear
original image to emphasize intensity changes. contrast enhancement techniques have been found to be useful
for enhancing the color contrast between the nearly classes
Convolution programs are 3-by-3 masks operating on pixel and subclasses of a main class.
neighborhoods.
A type of non linear contrast stretch involves scaling the input
Highpass filter-Emphasizes regions with rapid data logarithmically. This enhancement has greatest impact on
intensity changes. the brightness values found in the darker part of histogram. It
could be reversed to enhance values in brighter part of
Lowpass filter-Smoothes images, blurs regions with histogram by scaling the input data using an inverse log
rapid changes function.
2.1 Image Enhancement Technique Histogram equalization is another non-linear contrast
enhancement technique. In this technique, histogram of the
Image enhancement techniques improve the quality of an original image is redistributed to produce a uniform
image a perceived by a human. These techniques are most population density. This is obtained by grouping certain
useful because many images when examined on a color adjacent grey values. Thus the number of grey levels in the
display give inadequate information for image interpretation. enhanced image is less than the number of grey levels in the
There is no conscious effort to improve the fidelity of the original image
image with regard to some ideal form of the image. There
exists a wide variety of techniques for improving image 3. Histogram
quality. The contrast stretch, density slicing, edge
Histogram equalization is a widely used scheme for contrast
enhancement, and spatial filtering are the more commonly
enhancement in a variety of applications due to its simple
used techniques. Image enhancement is attempted after the
function and effectiveness. One possible drawback of the
image is corrected for geometric and radiometric distortions.
histogram equalization is that it can change the mean
Image enhancement methods are applied separately to each
brightness of an image significantly as a consequence of
band of a multispectral image. Digital techniques have been
histogram flattening. Clearly, this is not a desirable property
found to be most satisfactory than the photographic technique
when preserving the original mean brightness of a given image
for image enhancement, because of the precision and wide
is necessary. As an effort to overcome such drawback for
variety of digital processes.
extending the applications of the histogram equalization in
consumer electronic products, bi-histogram equalization has
Contrast
been proposed by the author which is capable of preserving
the mean brightness of an image while it performs contrast
Contrast generally refers to the difference in luminance or
enhancement. The essence of the bi-histogram equalization is
grey level values in an image and is an important
to utilize independent histogram equalizations separately over
characteristic. It can be defined as the ratio of the maximum
two subimages obtained by decomposing the input image
intensity to the minimum intensity over an image. Contrast
based on its mean. A simplified version of the bi-histogram
ratio has a strong bearing on the resolving power and detects
equalization is proposed, which is referred to as the quantized
ability of an image. Larger this ratio, more easy it is to
bi-histogram equalization. The proposed algorithm provides a
interpret the image.
much simpler hardware (H/W) structure than the bi-histogram
equalization since it is based on the cumulative density
Contrast Enhancement
function of a quantized image. Thus, the realization of bi-
histogram equalization in H/W is feasible, which leads to
Contrast enhancement techniques expand the range of
versatile applications in the field of consumer electronics
brightness values in an image so that the image can be
efficiently displayed in a manner desired by the analyst. The
density values in a scene are literally pulled farther apart, that 3.1 What is Histogram
is, expanded over a greater range. The effect is to increase the
visual contrast between two areas of different uniform The histogram in the context of image processing is the
densities. This enables the analyst to discriminate easily operation by which the occurrences of each intensity value in
between areas initially having a small difference in density. the image is shown. Normally, the histogram is a graph
showing the number of pixels in an image at each different
Non-Linear Contrast Enhancement intensity value found in that image. For an 8-bit grayscale
image there are 256 different possible intensities, and so the
In these methods, the input and output data values follow a histogram will graphically display 256 numbers showing the
non-linear transformation. The general form of the non-linear distribution of pixels amongst those grayscale
contrast enhancement is defined by y = f (x), where x is the values. Histogram modification is a classical method for
7
International Journal of Software Engineering Research & Practices Vol.1, Issue 2, April, 2011
image enhancement, especially histogram equalization. means 0 , 1 , 2 , 3 , ...... , 255. That is how new intensity levels
Histogram equalization method is a self-acting process since it are calculated for the previous intensity levels.
does not request any information, just only the probability of The next step is to replace the previous intensity level with the
each intensity level of image. However, the enhanced image is
new intensity level. This is accomplished by putting the value
obtained by the global area histogram equalization will cause
of Oi in the image for all the pixels, where Oi represents the
an effect of intensity saturation in some areas.
new intensity value, whereas i represents the previous
intensity level.
3.2 What is Histogram Equalization
Histogram equalization is the technique by which the dynamic To understand the working of the histogram equalization, take
range of the histogram of an image is increased. Histogram the example of the following image: the dynamic range of
equalization assigns the intensity values of pixels in the input image intensities is shown by the following histogram:
image such that the output image contains a uniform
distribution of intensities. It improves contrast and the goal of
histogram equalization is to obtain a uniform histogram. This
technique can be used on a whole image or just on a part of an
image.
Histogram equalization redistributes intensity distributions. If
the histogram of any image has many peaks and valleys, it will
still have peaks and valley after equalization, but peaks and
valley will be shifted. Because of this, "spreading" is a better
term than "flattening" to describe histogram equalization. In
histogram equalization, each pixel is assigned a new intensity
value based on its previous intensity level.
3.3 General Working
The histogram equalization is operated on an image in three
step:
1). Histogram Formation
2). New Intensity Values calculation for each Intensity Levels
Fig1: Dark image
3). Replace the previous Intensity values with the new
intensity values
For the first step see the article on histogram.
In step 2, new intensity values are calculated for each intensity
level by applying the following equation:
The meaning of Max. Intensity Levels maximum intensity
level which a pixel can get. For example, if the image is in the
grayscale domain, then the count is 255. And if the image is of
size 256x256 then, the No. of pixels is 65536. And the
expression is the bracket means that the no. of pixels having
the intensity below the output intensity level or equal to it. For
example, if we are calculating the output intensity level for 1
input intensity level, then it means that the no. of pixels in the
image having the intensity below or equal to 1 means 0 and 1.
If we are calculating the output intensity level for 5 input
intensity level, then the it means that the no. of pixels in the Fig4: High contrast image
image having the intensity below or equal to 5 means 0 , 1 , 2 ,
3 , 4 , 5. Thus, if we are calculating the output intensity level
for 255 input intensity level, then the it means that the no. of
pixels in the image having the intensity below or equal to 255
8
International Journal of Software Engineering Research & Practices Vol.1, Issue 2, April, 2011
3.4 Histogram Equalization: Regional Adaptive for the source image. Also, areas of low intensity and low
contrast in the source image were correspondingly remapped
The Histogram Equalization: Regional Adaptive algorithm, a in the reference image to areas of higher contrast and
high-pass filter, enhances the contrast in an image by generally overall brighter values.
reevaluating the gray-scale, or intensity, value of each pixel
based on the values of nearby pixels in the same region. Notice that different low areas of the reference image are more
Background distinguishable depending on the number of rows and columns
that divide the image. The contrast changes are due to which
Features in some images require a larger range of intensity pixel is considered the brightest and which the darkest in a
values than are displayed. Histogram equalization is a method particular region. The brightest pixel in a 1 x 1 separated
to improve the contrast of an area in an image by distributing image is the brightest of the image. However, when the image
an equal number of pixels across the range of intensities in the is divided into more regions, the brightest pixel in the region
image. This algorithm tabulates the histogram for each region, gets remapped as one of the brightest pixels in the image
then assigns the pixel to the new histogram level. regardless of its absolute intensity in the image. Likewise, the
darkest pixel in the region is remapped as one of the darkest in
This algorithm acts in regions of an image, dividing the image the image regardless of its absolute intensity.
into m rows and n columns. Each region is the source when
remapping the histogram, adapting the value of the pixel to its The histogram equalization process can enforce certain
location in the histogram of the region. characteristics about the image called clipping. As the
algorithm processes the image, it counts out the number of
Although, depending on image type, there are some minor pixels of certain intensity. Clipping cuts that counting short; it
variations in how this algorithm proceeds, generally the stores the excess pixels of a particular intensity and
algorithm first calculates the bin width of each region. It then redistributes that number equally among all intensities. For
processes each region by judging each pixel's location within instance, if a particular intensity must be clipped at 100 and
this new histogram. Regardless of pixel value, the algorithm the count is 145, the algorithm redistributes 45 pixels to other
remaps the brightest pixel in each region to the brightest value brightness. It adds to the count of each intensity level an equal
in the histogram (i.e., 256) and remaps the darkest pixel to the number of pixels if there are enough excess, but it attempts to
lowest value in the histogram (i.e., 0) of the pixels' spread the excess over as many pixels as possible.
neighboring region, recording the new values as pixel-
intensity values. The algorithm then processes each of the Gamma correction
remaining regions in the original, or source, image in the same
way and builds a new, or reference, image using the new Gamma correction operation performs nonlinear brightness
pixel-intensity values. adjustment. Brightness for darker pixels is increased, but it is
almost the same for bright pixels. As result more details are
Performing histogram equalization makes it possible to see visible.
minor variations within portions of the reference image that
appeared nearly uniform in the source image. Arithmetically, 4. Conclusion
the process in a region is rather simple:
Histogram equalization is powerful method for image
enhancement and it will increase the contrast of image. The
enhanced image will give the full dynamic range of histogram.
However, histogram equalization process tries to merge the
adjacent gray levels together in order to force the uniformity
of number of pixels in each appeared gray levels.
where Ni is the intensity of the ith pixel, and T is the total
Consequently, the intensity saturation will be presented in
number of pixels in the region.
darkness regions and whiteness region. Histogram
The reference image using the new histogram is similar to the equalization assigns the intensity values of pixels in the input
source image; however, in areas where the source image had image such that the output image contains a uniform
similar values, the reference image is enhanced. The reference distribution of intensities. It improves contrast and obtain a
histogram now stretches over a greater proportion of the uniform histogram. This technique can be used on a whole
possible image intensities image or just on a part of an image.
The histogram for the source image is very compact in 5. References
structure; that is, it has a great number of darker pixels, but
very few brighter pixels. In fact, the bottom-most intensity
[1]. Bob Cromwell, "Localized Contrast Enhancement: Histogram
values make up almost all of the pixels' values in the image. Equalization," paper on his web site at http://www.cromwell-
Note, however, in the histogram of the reference image, the intl.com/3d/histogram/Index.html.
look-up table is more evenly spread than that of the histogram
9
International Journal of Software Engineering Research & Practices Vol.1, Issue 2, April, 2011
[2]. John Russ, The Image Processing Handbook (Boca Raton, Florida:
CRC Press LLC, 2003).
[3]. J. Alex Stark, "Adaptive Image Contrast Enhancement Using
Generalizations of Histogram Equalization," in IEEE Transactions
on Image Processing (May 2000).
[4]. http//www.codersource.net/MicrosoftNet/CImageProcessing/Histo
gramandNormalizedHistograminC.aspx
[5]. Dr A D Marshall,‖ Vision Systems ‖, Vision_lecture_caller.tex.
[6]. ROHAN KAR―IMAGE PROCESSING – OPTICAL
CHARACTER RECOGNITION‖ Seminar Report 2004 Monsoon
[7]. http//www.ndt
ed.org/EducationResources/CommunityCollege/Radiography/Adv
ancedTechniques/Real_Time_Radiography/ImageProcessingTechn
iques.htm
[8]. Minakshi Kumar‖ DIGITAL IMAGE PROCESSING‖,
Photogrammetry and Remote Sensing Division Indian Institute of
Remote Sensing, Dehra Dun
[9]. V.V. Starovoitov, D.I Samal, D.V. Briliuk‖ Image Enhancement
for Face Recognition ―,Submitted to International Conference on
Iconics, 2003, St.Petersburg, Russia.
[10]. YuWang; QianChen; BaeominZhang; ―Image enhancement based
on equal area dualistic sub-image histogram equalization method
―06 August 2002 IEEE Consumer Electronics Society
10
