International Journal of Engineering and Technical Research (IJETR)
ISSN: 2321-0869, Volume-2, Issue-5, May 2014
Bi-Histogram Equalization Using Frequency and
Spatial Technique for Digital Image Enhancement
Avinash Dhole, Tilak Raj Patel
modified are specific to a given task. Moreover,
Abstract Image Processing deals with analysis of image observer-specific factors, such as the human
using different techniques for different purpose. Image visual system and the observer's experience, will introduce a
Processing deals with the any action to change a image. Image great deal of subjectivity into the choice of image
processing has different methods like optical, analog and enhancement methods. There exist many techniques that can
digital image processing. Digital image processing is a part of
enhance a digital image without spoiling it. The
signal processing where we processes digital images using
computer algorithms. The computer algorithms can be enhancement methods can broadly be divided in to the
modified so that we can also change the appearance of the following two categories:
digital image easily and quickly. Many images such as medical A. Spatial Domain Methods
images, remote sensing images, electron microscopy images
and even real life photographic pictures, suffer from poor Spatial domain techniques directly deal with the image
contrast. Therefore it is necessary to enhance the contrast. pixels. The pixel values are manipulated to achieve desired
Today there are many image enhancement techniques are enhancement. Spatial domain techniques like the
available. There are some drawbacks of these techniques. logarithmic transforms, power law transforms, histogram
Adaptive histogram produces blurred and washed out images
equalization, are based on the direct manipulation of the
especially at the edges. Whereas alpha rooting produces
over-graying enhanced images so to overcome all these pixels in the image. Spatial techniques are particularly useful
drawbacks. for directly altering the gray level values of individual pixels
and hence the overall contrast of the entire image. But they
In this proposed method we are combining the frequency usually enhance the whole image in a uniform manner which
domain and spatial domain techniques for image enhancement in many cases produces undesirable results.
by using alpha rooting techniques with Bi-Histogram
Equalization in order to enhance the quality of the image. a) Contrast Stretching: -
There are many histogram equalization based methods have
been introduced, many of these methods are relatively
Contrast stretching enhances image by enhancing contrast
complicated in implementation, and most of Techniques
between various parts of the original image. The basic idea is
require a high computational time. Bi - Histogram
Equalization with Plateau Level (BHEPL) as one of the options to improve the image quality by increasing the dynamic
for the system that requires a short processing time image range of gray levels.
enhancement.
b) Grey Level Slicing
Index Terms Image enhancement, spatial domain
technique, Bi-Histogram Equalization technique, alpha rooting
A grey level slicing technique enhances all the gray levels in
(key words) the range of interest using high values and all other gray
levels using low values. This function is particularly useful to
I. INTRODUCTION enhance flaws in X-ray images and enhancing features such
as masses of water in satellite image.
Image enhancement techniques are used to improve the
image quality so that the resultant image quality is better c) Histogram Equalization (HE)
than the original image for a specific application or set of
objectives. Image enhancement is the task of applying certain Histogram equalization is a technique by which the dynamic
alterations to an input image like as to obtain a more visually range of the histogram of an image is increased. It flattens
pleasing image. The alteration usually requires and stretches the dynamic range of the image's histogram
interpretation and feedback from a human evaluator of the and resulting in overall contrast improvement. Histogram
output resulting image. The principal objective of image equalization assigns the intensity values of pixels in the input
enhancement is to modify attributes of an image to make it image such that the output image contains uniform
more suitable for a given task and a specific observer. During distribution of intensities. It improves contrast by obtaining a
this process, one or more attributes of the image are uniform histogram.
modified. The choice of attributes and the way they are
B. Frequency Domain Methods
Manuscript received May 19, 2014. The convolution theorem is the foundation of frequency
Avinash Dhole, Department of Computer Science and Engineering, Raipur domain techniques. Consider the following spatial domain
Institute of Technology, Raipur.
Tilak Raj Patel, Department of Computer Science and Engineering, Raipur
operation:
Institute of Technology, Raipur. g (x,y) = h (x,y) * f (x,y)
227 www.erpublication.org
Bi-Histogram Equalization Using Frequency and Spatial Technique for Digital Image Enhancement
The convolution theorem states that following frequency over-graying enhanced images so to overcome all these
domain relationship holds: drawbacks. In this proposed techniques we are combining the
G (u,v) = H (u,v) F (U,v) frequency domain and spatial domain for image
G, H and F are the Fourier transforms of g, h and f enhancement. For this we are using alpha rooting techniques
respectively. H is known as the transfer function of the with Bi-Histogram Equalization in order to enhance the
process. Many image enhancement problems can be quality of the image. There are many histogram equalization
expressed in the form of the above equation. The goal is to based methods have been introduced, many of these methods
select a transfer function that changes the image in such a are relatively complicated in implementation, and most of
way that some feature of the image is enhanced. Techniques require a high computational time. For fast
processing we are using BHEPL techniques. Bi-Histogram
a) General Concept: Equalization with a Plateau Level (BHEPL) as one of the
options for the system that requires a short processing time
The frequency filters process an image in the frequency image enhancement [5].
domain. Application of this type of filtering is easy: In proposed technique we first apply the Alpha rooting
algorithm on image. It is a simple but effective technique of
1. Transform the image into the Fourier domain image enhancement in the transform or frequency domain.
2. Multiply the image by the filter Alpha Rooting holds the flexibility of a parameter aided
3. Take the inverse transform of the image operation which is highly appreciated for experimenting
with different levels of enhancement. We first separate
magnitude and phase coefficients from Fourier Transform
Output then Apply Alpha Rooting and combine the
Magnitude and Phase coefficients. This Alpha Rooting
produces over-graying enhanced images. Output of the
Alpha rooting is passed as a input for the Bi-Histogram
Equalization with a Plateau Level (BHEPL). This BHEPL
method is enhanced the images without producing unwanted
artifacts. The method also is able to maintain the mean
brightness better than some well known mean brightness
preserving histogram equalization methods.
III. PROPOSED METHODS CONSISTS FOLLOWING STEPS
Frequency domain enables us to view the frequency content
of the image. In this frequency domain image enhancement
has been done via Fourier transform on image and then get
the appropriate transform function for image enhancement
and manipulating the image with transform function and
finally taken the inverse transformation on the Transformed
image.
b) Alpha rooting Technique: -
Alpha rooting is a simple but effective technique of image
enhancement in the transform or frequency domain. The
technique is applied on the orthogonal transforms of images.
It is used to augment the high frequency content in the image.
The method is based upon the fact that after applying an
orthogonal transform, high frequency coefficients of an
image, will have smaller magnitudes than low frequency
coefficients. By raising the magnitude of an image to some
value, , where 0 < < 1, the higher valued lower frequency
components of an image can be reduced more in proportion
to the lower valued high frequency components.
II. PROPOSED METHODS
There are so many image enhancement techniques are
available. They have some drawbacks on those Techniques.
Adaptive histogram produces blurred and washed out images
especially at the edges. Whereas alpha rooting produces
228 www.erpublication.org
International Journal of Engineering and Technical Research (IJETR)
ISSN: 2321-0869, Volume-2, Issue-5, May 2014
IV. EXPERIMENTAL RESULTS
The objective of the experiments was to establish the F. Enhanced Image with Propose Technique
algorithm for image enhancement. Experimental result on
image by using these proposed techniques of frequency and
spatial domain techniques of image enhancement has been
shown below.
A. Original Image
V. CONCLUSION
Image enhancement techniques such as spatial and transform
domain technique are important techniques. In this proposed
B. Enhanced Image with Alpha Rooting method we are combining the frequency domain and spatial
domain for image enhancement by using alpha rooting
techniques with Bi-Histogram Equalization in order to
enhance the quality of the image. The proposed technique
will produce highly quality and visually appealing results for
a diversity of images with different qualities of contrast and
edge information.
REFERENCES
[1] Rajesh Garg,Bhawna Mittal, Sheetal Garg, Histogram Equalization
Techniques for Image Enhancement, IJECT Vol. 2, Issue 1,
C. Enhanced Image with Propose Technique March2011, ISSN 2230-9543.
[2] Gonzalez, Rafael C., Richard E. Woods. : Digital Image Processing. Ed
III, Pearson Education Asia, New Delhi, 2007.
[3] Arun R, Madhu S. Nair, R. Vrinthavani and Rao Tatavarti. An Alpha
Rooting Based Hybrid Technique for Image Enhancement. Online
publication in IAENG, 24th August 2011.
[4] J. H. McClellan, Artifacts in alpha-rooting of images, in Proc. IEEE
Int. Conf. Acoustics, Speech, Signal Processing, Apr. 1980, pp.
449452.
[5] Chen Hee Ooi, Nicholas Sia Pik Kong, Bi-Histogram Equalization with
a Plateau Limit for Digital Image Enhancement. IEEE Transactions on
Consumer Electronics, Vol. 55, No. 4, NOVEMBER 2009
[6] Yeong-Taeg Kim, Contrast enhancement using brightness preserving
bi-histogram equalization, IEEE Trans. Consumer Electronics, vol. 43,
D. Another Original Image no. 1, pp. 1-8, Feb. 1997
[7] Haidi Ibrahim, Nicholas Sia Pik Kong, Brightness Preserving Dynamic
Histogram Equalization for Image Contrast Enhancement, IEEE Trans.
Consumer Electron., vol. 53, no. 4, pp. 1752-1758, November 2007.
[8] Rafael C. Gonzalez, and Richard E. Woods, Digital Image Processing,
2nd edition, Prentice Hall, 2002.
Avinash Dhole is The Head of Computer Science & Engineering Department
in Raipur Institute of Technology, Raipur. He received
his B.E. degree in Computer Science from Government
Engineering College, Bilaspur. He joined as an assistant
professor at Raipur Institute of Technology in 2005.
During this period he obtained Master of Technology
degree from C.S.V.T University, Bhilai in 2009. He has
worked on various researches such as automata theory,
software engineering and neural networks.
E. Enhanced Image with Alpha Rooting
Tilak Raj Patel, received the B.E. degree in
Information Technology Engineering from Pt.Ravi
Shanker Shukla University Raipur in 2009. He is now
pursuing Master of Technology in Computer Science
& Engineering at Chhattisgarh Swami Vivekananda
Technical University, Bhilai.
229 www.erpublication.org
