Image Enhancement Based on Histogram Equalization

Wahab Ahmad Taha Maryum

ID: F2017266065 ID: F2017065035
Dept.: BS Computer Sciences Dept.: BS Software Engineering
Lahore,Pakistan Lahore,Pakistan
F2017266065@umt.edu.pk F2017065035@umt.edu.pk

Abstract— Image enhancement is the process of adjusting II. HISTOGRAM

digital images so that the results are more suitable for
A. About Histogram
display or further image analysis. Image enhancement can
be done by Histogram equalization. Histogram equalization Histogram equalization is a method in image processing of
is a technique for adjusting image intensities to enhance contrast adjustment using the image's histogram. Histograms of
contrast. This method usually increases the global contrast an image before and after equalization. This method usually
of many images, especially when the usable data of the increases the global contrast of many images, especially when
the usable data of the image is represented by close contrast
image is represented by close contrast values. Through this
values. Through this adjustment, the intensities can be better
adjustment, the intensities can be better distributed on the
distributed on the histogram. This allows for areas of lower
histogram. This allows for areas of lower local contrast to local contrast to gain a higher contrast. Histogram equalization
gain a higher contrast. Histogram equalization accomplishes accomplishes this by effectively spreading out the most
this by effectively spreading out the most frequent intensity frequent intensity values. The method is useful in images with
values. This finds applications in Science and Engineering. backgrounds and foregrounds that are both bright or both dark.
In particular, the method can lead to better views of bone
Keywords— Image processing, image enhancement, structure in x-ray images, and to better detail in photographs
histogram equalization, contrast enhancement that are over or under-exposed. A key advantage of the method
is that it is a fairly straightforward technique and an invertible
I. INTRODUCTION
operator. So in theory, if the histogram equalization function is
Image is an important way for people to obtain external information. known, then the original histogram can be recovered.
With the development of technology, a variety of cameras have
come on the scene. The resolution and exposure time of cameras The calculation is not computationally intensive. A
have been improved greatly. However, in different working disadvantage of the method is that it is indiscriminate. It may
environments, the images acquired by cameras always have various increase the contrast of background noise, while decreasing
problems, which cannot meet people's requirements very well. the usable signal.
Therefore, we need to process, analyze and transform the acquired
image. General image processing technologies include image
compression, image enhancement, image reconstruction and so
On. Image enhancement is a very important application in image
processing technology. It can improve the visual effect of the image,
enhance the contrast and brightness of the image, and highlight some
information in the image, and meet the requirements of analysis.
Airspace method and frequency domain method are two commonly
used methods for improving image quality. Airspace method mainly
deals with the pixel value of a point in the image directly to achieve
the effect; Frequency domain method mainly uses Fourier transform
to transform the image into frequency domain, and then uses inverse
transform to get the final result. By comparing and analyzing the two
methods, we can find that airspace method is simple and effective,
B. Application Of Hisytogram
and it is a common processing method. Histogram equalization is a
kind of spatial method. It uses the different contrast of image In scientific imaging where spatial correlation is more
histogram to process, increases the local brightness of the image, and important than intensity of signal (such as separating DNA
does not affect the contrast of the whole image while increasing the fragments of quantized length), the small signal to noise ratio
local contrast. usually hampers visual detection. Histogram equalization
often produces unrealistic effects in photographs; however it is

Identify applicable funding agency here. If none, delete t\his text box.

XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE

very useful for scientific images like thermal, satellite or x-ray wiener filter gives the best result as regards the time and
images, often the same class of images to which one would efficiency. Besides, the demonstrated LFR is capable of
apply false-color. Also histogram equalization can produce tracking a pre specified colored line as long as it is different
undesirable effects (like visible image gradient) when applied from the surroundings.
to images with low color depth. For example, if applied to 8- B. In Flood Detection
bit image displayed with 8-bit gray-scale palette it will further
reduce color depth (number of unique shades of gray) of the Flood monitoring technique using adjustable histogram
image. Histogram equalization will work the best when equalization has been proposed by Nazir et al., .The technique
applied to images with much higher color depth than palette overcomes the limitations (over enhancement, artifacts, and
size, like continuous data or 16- bit gray-scale images. unnatural look) of existing technique by adjusting the contrast
of images. Nazir et al., have proposed a technique which takes
There are two ways to think about and implement histogram pre- and post-images and applies different processing steps for
equalization, either as image change or as palette change. The generating flood map without user interaction. The resultant
operation can be expressed as P(M(I)) where I is the original flood maps can be used for flood monitoring and detection.
image, M is histogram equalization mapping operation and P Simulation results show that the proposed technique provides
is a palette. If we define a new palette as P'=P(M) and leave better output quality compared to the state of the art existing
image I unchanged then histogram equalization is technique.
implemented as palette change. On the other hand if palette P C. In other Areas
remains unchanged and image is modified to I'=M(I) then the Design of products characterized by high stylistic content and
implementation is by image change. In most cases palette organic shapes in the form of bas-relief (e.g. fashion
change is better as it preserves the original data. accessories, commemorative plaques and coins) is
Modifications of this method use multiple histograms, called traditionally performed starting from handmade drawings or
sub histograms, to emphasize local contrast, rather than overall photographs that are manually reproduced by highly skilled
contrast. Examples of such methods include adaptive craftsmen such as sculptors and engravers and finally digitized
histogram equalization, contrast limiting adaptive histogram by means of 3D scanning. Several Computer-based procedures
equalization or CLAHE, multipeak histogram equalization have been devised with the aim of speeding up this process,
(MPHE), and multipurpose beta optimized bi-histogram which is considerably time consuming, subjective and costly;
equalization (MBOBHE). The goal of these methods, these are mainly based on image processing techniques such
especially MBOBHE, is to improve the contrast without as embossing, enhancement, histogram equalization or
producing brightness mean-shift and detail loss artifacts by dynamic range, also implemented in CAD-based commercial
modifying the HE algorithm. software. However, these approaches are characterized by
III. ADVANCEMENT IN HISTOGRAM several limitations preventing them from providing a "correct"
final geometry Image processing-based techniques and
A. In Indrustrial Area minimization SFS methods are first used in order to retrieve a
In a fast growing industrial world, carriers are required to rough version of the objective surface; successively, this is
carry products from one manufacturing plant to another which used as initialization for the final reconstruction algorithm.
are usually in different buildings or separate blocks. Elhady et Tested on a set of case studies, the method proved to be
al., have devised a study to automate this sector using vision effective in providing satisfactory digital bas-relief from
controlled mobile robots instead of laying railway tracks which single.
are both expensive and inconvenient. To achieve this purpose
an autonomous robot with computer vision as its primary IV. HISTOGRAM EQUALIZATION
sensor for gaining information about its environment for path
following is developed. The proposed Line Follower Robot Histogram equalization is a specific case of the more general
(LFR) Image enhancement by Histogram equalization. A PID class of histogram remapping methods. These methods seek to
control algorithm will be applied to adjust the robot on the line.
adjust the image to make it easier to analyze or improve visual
The proposed LFR is accomplished through the following
quality.
stages: Firstly, the image is acquired using the web cam. The
acquired RGB image is converted to another color coordinates Histogram equalization is a scheme that improves the contrast
for testing and comparing to choose the best color space. After of input image and stretch out the intensity range. In worst
that, the image contrast is enhanced using histogram case of equalization, contrast may also decrease. In discrete
equalization and then Wiener, Lee and Kuan filters are domain, the transformation function is:
implemented to decide the best filter to be implemented. 𝑠𝑘 = (𝐿 − 1) ∗ 𝑝𝑘 (𝑟) (4)
Subsequently, the basic morphological operations are carried
out to choose the suitable operation to be utilized. The results
are evaluated qualitatively and quantitatively from the points of
Peak Signal-to-Noise Ratio (PSNR), entropy and image
smoothness. The results show that the closing process is more
suitable for the vision enhancement purpose, as well as the
Where 𝑝𝑘 (𝑟) is probability density function. The values are applied. Nonlinear methods effectively preserve edges and
rounded off and then mapped in original image to obtain details of images while methods using linear operators tend to
blur and distort them. Additionally, nonlinear image
enhancement tools are less susceptible to noise. Noise is
always present due to the physical randomness of image
acquisition systems. For example, under-exposure and low-
light conditions in analog photography conditions lead to
images with film-grain noise, which, together with the image
signal itself, are captured during the digitization process.
A. Spatial Domain Image Enhancement
a) Gray Scale Enhancement:
Gray level transformation is an important technology
of image enhancement in airspace method. It can expand the
display width of the image and enhance the contrast of the
equalized image with improved contrast. whole image. Gray transformation improves every pixel of the
image by point operation, and analyses and adjusts the pixel
V. FLOW DIAGRAM OF HISTOGRAM EQUALIZATION value of the output point by the gray level of the input pixel.
Gray-level transformation does not change the spatial
relationship of the image itself, but uses some gray-level
transformation method to adjust the pixel of the image. It is an
operation of transforming from one pixel to another.

b) Linear Gray Scale Enhancement:

Linear gray enhancement is a kind of gray
transformation enhancement. It transforms every pixel of an
image according to a certain gray transformation relationship.
If the camera is under-exposed or over-exposed, the gray level
of the whole image will be limited to a very small gray level
range. At this time, the collected image will be blurred. It is
possible to use a linear function to transform each pixel in the
image linearly, which can effectively improve the quality of
the image. If the gray scale range before an image
transformation is [a, b], and you want the gray scale range
after the transformation to be expanded or compressed to [c,
d], the gray scale linear transformation function expression is:

c) Non-Linear Gray Scale Enhancement:

When the nonlinearity function is used to map the image
VI. ENHANCEMENT BY HISTOGRAM EQUALIZATION
grayscale, the nonlinear grayscale enhancement of the image
Image enhancement tools are often classified into point can be realized. Commonly used nonlinear gray enhancement
operations and spatial operations. Point operations include methods are logarithmic function nonlinear transformation and
contrast stretching, noise clipping, histogram modification, exponential function nonlinear transformation.
and pseudo-coloring. Point operations are, in general, simple When doing a logarithmic nonlinear transformation on an
nonlinear operations that are well known in the image image, the transformation function is:
processing literature and are covered elsewhere. Spatial
operations used in image processing today are, on the other
hand, typically linear operations. The reason for this is that
spatial linear operations are simple and easily implemented.
Although linear image enhancement tools are often adequate
in many applications, significant advantages in image
enhancement can be attained if nonlinear techniques are
 VII. CUMULATIVE SUM HISTOGRAM
Histogram equalization redistributes the pixel intensity values
evenly by using cumulative (sum) histogram as a transfer
function or as a look-up table. Image enhancement by
When performing an exponential function nonlinear
Histogram equalization
transformation on an image, the transformation function is:

By adjusting a, b, c, you can adjust the position and shape of

the curve. With this transformation, the low gray scale range
of the input image can be expanded, and the high gray scale
range is compressed to make the image distribution uniform.

d) Piece-Wise Gray Scale Enhancement:

Piecewise linear gray enhancement does not transform all the
pixels of the image equally, but expands the gray level of the The projected intervals get wider or narrower based on the
pixels needed and compresses the unnecessary pixels. If the slope of cumulative histogram. The idea of histogram
equalization comes from the cumulative histogram. Take a
look at the left image of an example of cumulative histogram.
The frequencies (the number of pixels) at each intensity value
are accumulated. The higher pixel density is, the greater
steepness is. There are 2 intervals with same width, but
different slopes; one interval has slow steepness, and the other
interval has steep slope. Let's project the intervals of the
gray scale range of the input image is 0~M level, the gray intensity values onto the other axis. Once they are projected,
scale range of the enhanced image is 0~N level, and the they have different intervals on the projected axis. If the slope
interval [a, b], [c, d] are respectively a gray level interval of is less than 1, the interval gets narrower. If the slope is greater
the source image and the enhanced image. Then the piecewise
linear transformation function is:

B. Histogram Modification Enhancement

The histogram of the image is to analyze the gray scale range
of image and understand the distribution of the gray level.
From the distribution of the gray level, we can see many than 1, then the interval is going to be wider. In other words,
features of the image. Therefore, image enhancement can be we are going to spread the intensity range if the pixel density
achieved by changing the histogram characteristics of the is high, and shrink the interval if the density is low. Histogram
image, modifying the histogram of the image, improving the equalization works best on an over or under exposed image,
different gray levels and enhancing the contrast of the image. which has narrow contrast range. In histogram equalization,
Histogram equalization and histogram specification are the the input pixel intensity, x is transformed to new intensity
two most commonly used methods. Before performing value, x′ by T. The transform function, T is the product of a
histogram related operations, we must know what the cumulative histogram and a scale factor. The scale factor is
histogram looks like, that is, histogram statistics. needed to fit the new intensity value within the range of the
Mathematically, the image histogram is actually a function of intensity values,
the statistical properties of the gray values of the image and Notice that the cumulative histogram after histogram
the gray value of the image. It counts the number or equalization. It is almost linear, which means the number of
probability of occurrence of each gray level in an image. intensity values are evenly distributed.
Graphically speaking, it is a two-dimensional map, with the
abscissa indicating the gray level of each pixel; the ordinate
indicates the number of pixels or probability of the
corresponding gray level.
 image, the size of histogram bin is 256, because the range of
the intensity of 8-bit image is from 0 to 255.

Histogram is a useful tool to analyze the brightness and

contrast of an image. It shows how the intensity values of an
image is distributed and the range of brightness from dark to
bright. An image can be enhanced by remapping the intensity
values using the histogram. Also, histogram is used to segment
an image into the several regions by thresholding. For
example, if the image intensities in the histogram are divided
into 2 groups, the threshold value can be chosen at the middle
of 2 peaks in the histogram
a) 8-bit Gray scale Image Enhancement by Histogram
Equalization
This image is the enhancement of the gray scale 8-bit image
by using the technique of the Histogram Equalization. . It can
improve the visual effect of the image, enhance the contrast
and brightness of the image, and highlight some information in
the image. In this image the low values of gray scale is
enhanced and adjust the contrast and equally distribute the
gray scale vales. As we can see in this image the image is

enhanced and brighter the value of image

Grayscale and color component histograms extracted from the
image after equalization the histogram obtained by equalizing
the color components of the image. From the histogram shown
in the figure, it can be observed that the display width of the
image is significantly widened, and the gray value of the
region is significantly enhanced. The three images are
synthesized using the cat command in Matlab. Compared with
VIII.IMPLEMENTATION OF HISTOGRAM the original image and linear change in pixel the contrast of
A. Histogram of 8-bit Gray scale Image the image after equalization is improved, and the brightness of
the image is obviously enhanced.
In Digital Signal Processing (DSP), we store the number of
pixels (frequencies) of same intensity values into a histogram
array, which is commonly called "bin". For an 8-bit grayscale

B. Histogram of Color Image

Image captured by camera in MATLAB is an X×Y×3 array.
Each data in the array is the color information components
corresponding to the color image in a specific spatial position.
By combining these three components, a color image can be
obtained. In matlab, read a color image, its data type is unit8,
that is, 8-bit unsigned integer data. The image stored in this
way is called 8-bit image, which can save data space very
well. The imread command in MATLAB is to store the image [1] Yuanmin Xie et al 2019 J. Phys.: Conf. Ser. 1314 012161
into an 8-bit matrix. If the read image is a color image, it will [2] http://in.mathworks.com/discovery/image-enhancement.html
store the data into an 8-bit RGB matrix. For example, if the [3] https://en.wikipedia.org/wiki/Histogram_equalization
pixels of the color image read are 800×1248, then the matrix [4] A new method for fusion, denoising and enhancement of x-ray images
retrieved from Talbot-Lau grating interferometry, Scholkmann, F.,
stored in MATLAB is 800×1248×3, and each channel value of Revol, V.,Kaufmann, R., Baronowski, H.,Kottler, C., Physics in
the color image is between 0 and 255. The images saved by Medicine and Biology,2014,59 (6), pp. 1425-1440
MATLAB are all uint8 data types, but the double type is used [5] Image enhancement based on contourlet transform, Asmare, M.H.,
in matrix operation. This method can ensure the accuracy very Asirvadam, V.S.,Hani, A.F.M. Signal, Image and Video Processing, 20
well. In matrix operation, uint8 type arrays can operate with March 2014, DOI10.1007/s11760-014-0626-7
each other, and the result is still uint8 type. [6] Y. Wang, Q. Chen, B. Zhang, “Image enhancement based on equal area
dualistic sub-image histogram equalization method,” IEEE Trans. on
R1=RGB(:,:,1); Consumer Electronics, vol. 45, no. 1, pp. 68-75, Feb. 1999.
G2=RGB(:,:,2); [7] Abdullah-Al-Wadud, M., M.H. Kabir, M.A.A. Dewan and C. Oksam,
B3=RGB(:,:,3); 2007. A dynamic histogram equalization for Image contrast
enhancement. IEEE T. Consum. Electr., 53(2): 593-600.
It can be seen from the figure that the dynamic range of the [8] Nilsson M, Sallberg B and Claesson I 2013 IEEE Transactions on Image
Processing 22 644-656.
original image histogram is limited to a small range. The
[9] Flood detection/monitoring using adjustable histogram equalization
degree of aggregation is high and the gray value of the area is technique, Nazir, F., Riaz, M.M., Ghafoor, A., Arif, F., The Scientific
low. The contrast of the original image is not high and the World Journal, 2014, 2014, 809636
brightness is low, some important information in the image [10] http://www.songho.ca/dsp/histogram/histogram.html
can't show up. [11] A simple and effective histogram equalization approach to image
enhancement Author links open overlay panelH.D.ChengX.J.Shi
[12] Image Enhancement Based on Histogram EqualizationTo cite this
article: Yuanmin Xie et al 2019 J. Phys.: Conf. Ser. 1314 012161

[13] Image contrast enhancement using histogram equalization with

maximum intensity coverage Chin Yeow Wong,Shilong Liu,San Chi
Liu,Md Arifur Rahman,Stephen Ching-Feng Lin,Guannan Jiang,

[14] ] Rafael C. Gonzalez, Richard E. Woods, “Digital Image Processing”,

2nd edition, Prentice Hall, 2002.

[15] 5] A. K. Jain, “Fundamentals of Digital Image Processing”. Englewood

Cliffs, NJ: Prentice-Hall, 1991.
IX. CONCLUSION
Image enhancement is an indispensable video processing [16] J. Alex Stark “Adaptive Image Contrast Enhancement Using
method. Histogram equalization is the most commonly used Generalizations of Histogram Equalization”, IEEE Transactions on
method of image enhancement. It can improve the quality of Image Processing, Vol. 9, No. 5, May 2000
image very well. This paper mainly discusses the principle of
histogram equalization, and carries out image enhancement
experiments through MATLAB experiments. The results show [17] Yu Wan, Qian Chen, Bao-Min Zhang, “Image enhancement based on
equal area dualistic sub-image histogram equalization method,” IEEE
that the histogram equalization method has a good processing Transactions Consumer Electron., vol. 45, no. 1, pp. 68-75, 1999.
effect. It will increase the contrast of the image and highlight [18] Histogram Equalization Techniques For image Enhancement 1Rajesh
the elements of the image. Garg, 2Bhawna Mittal, 3Sheetal Garg1,2H.I.T., Sonepat, Haryana, India
3S.M.Hindu Sr.Sec.School, Sonepat, Haryana, India
[19] De La Torre, A., Peinado, A. M., Segura, J. C., Prez-Crdoba, J. L.,
Bentez, M. C., Rubio, A. J., “Histogram equalization of speech
ACKNOWLEDGMENT representation for robust speech recognition,” in IEEE Trans. Speech
Audio Processing, vol. 13, pp. 355–366, May 2005.
The authors are thankful to their respective managements and
[20] A. Zagzebski, “Essentials of Ultrasound Physics”. St. Louis, Missouri:
appropriate websites. The material regarding the topic and its Mosby, 1996.
implementation is mainly support this article to describe it
easily.
REFERENCES