International Journal of Advanced Engineering Research and Technology (IJAERT) 312

Volume 2 Issue 8, November 2014, ISSN No.: 2348 8190

Performance Analysis of Image Filtering Technique and Combined

Approach for Image Segmentation
Shradha P. Dakhare*, Manoj B. Chandak**
*Department of Computer Science & Engineering, Nagpur University
W.C.E.M. Dongargaon, Nagpur, India
**Department of Computer Science & Engineering, Nagpur University
S.R.C.O.E.M. Nagpur, India

ABSTRACT
Segmentation of an image entails the division or
separation of the image into regions of similar attribute.
The accurate and effective algorithm for segmenting
image is very useful in many fields, Many image
segmentation techniques have been developed over the
past two decades for segmenting the images, which help
for object recognition, occlusion boundary estimation
within motion or stereo systems, image compression,
image editing.
In this project that is combined approach for segmenting
the image, first we used histogram equalization to the
input image, from which we get contrast enhancement
output image. After that we applied median filtering which
will remove noise from contrast output image. At last we
applied fuzzy c-mean clustering algorithm to denoising
output image, which give segmented output image.
Keywords - Histogram Equalization(HE),
Filter(MF), Fuzzy C Means(FCM).

I.

Median

INTRODUCTION

Image segmentation refers to the major step in image

Processing in which the inputs are images and, outputs are
the attributes extracted from those images. The goal of
segmentation is typically to locate certain objects of
interest which may be depicted in the image. It is one of
the challenging tasks in image analysis.
Segmentation is the first essential and important step of
low level vision [9]. There are many application of image
segmentation. For example, in a vision guided car
assembly system, the Robot needs to pick up the
appropriate components from the bin .For this
segmentation followed by recognition is required. Its
application area varies from the detection of cancerous
cells to the identification of an airport from remote sensing
data, etc. In all this area, the quality of final output
depends largely on the quality of segmented output.

Segmentation is the process of partitioning an image into

non-intersecting regions such that each region is
homogeneous and the union of no two adjacent regions is
homogeneous. Formally, it can be defined as follows:
If F be the set of all pixels and P() be a uniformity
(homogeneity)predicate defined on groups of connected
pixels, then segmentation is a partitioning of the set F into
a set of connected subsets or regions(S1, S2, , Sn) such
that

The uniformity predicate

for all regions
and
when
is adjacent to
Note that this definition is applicable to all type of Images.
Clustering is powerful technique that has been research in
image segmentation. It is the process of organizing data
objects into a set of disjoint classes called clusters.
Clustering aims to analyze and organize data into groups
based on their similarity [6]. Clustering is an example of
unsupervised classification. This paper is divided in 3
section, in first section for each input image, HE , MF and
FCM are applied and compare their output on the basis of
time, mean square error (MSE), peak signal noise ratio
(PSNR), figure of merit (FOM) and data loss. In second
section HE & FCM, HE & MF and MF & FCM are
combined and compare there output. In third, combine
approach. for one input image HE is applied and the
output get from this HE, to the same output MF is applied
and the output get from MF ,to the same output FCM is
applied . The output get from this is the better segmented
output.

II.

HISTOGRAM EQUALIZATION

Histogram Equalization is a method that increases the

contrast of an image by increasing the dynamic range of
intensity given to pixels with the most probable intensity
values. One transformation function that accomplishes this
is a cumulative distribution function. In the histogram
equalization function that was used to produce the results
www.ijaert.org

International Journal of Advanced Engineering Research and Technology (IJAERT) 313

Volume 2 Issue 8, November 2014, ISSN No.: 2348 8190

shown in this report, the transformation is scaled such that

the least intense value in the original image is mapped to a
zero intensity value in the equalized image. As well, the
most intense value in the original image is mapped to an
intensity value that is equal to the maximum intensity
value determined by the bit depth of the image.

a)Normal HE

b)Adaptive HE

Fig. 2.1 This is an image of brain Tumor. Histogram

equalization increases the contrast of image in the image
a) is normal Histogram Equalization and b)is Adaptive
Histogram Equalization

2.1 ADAPTIVE HISTOGRAM EQUALIZATION

AHE is a computer image processing technique used to
improve contrast in images. It differs from ordinary
histogram equalization in the respect that the adaptive
method computes several histograms, each corresponding
to a distinct section of the image, and uses them to
redistribute the lightness values of the image. It is
therefore suitable for improving the local contrast of an
image and bringing out more detail. However, AHE has a
tendency to over amplify noise in relatively homogeneous
regions of an image. Ordinary histogram equalization uses
the same transformation derived from the image histogram
to transform all pixels. This works well when the
distribution of pixel values is similar throughout the image
[10].

III.

Median filtering is one kind of smoothing technique, as is

linear Gaussian filtering. All smoothing techniques are
effective at removing noise in smooth patches or smooth
regions of a signal, but adversely affect edges. Often
though, at the same time as reducing the noise in a signal,
it is important to preserve the edges. Edges are of critical
importance to the visual appearance of images. For small
to moderate levels of (Gaussian) noise, the median filter is
demonstrably better than Gaussian blur at removing noise
whilst preserving edges for a given, fixed window size.
However, its performance is not that much better than
Gaussian blur for high levels of noise, whereas, for
speckle noise and salt and pepper noise (impulsive noise),
it is particularly effective. Because of this, median filtering
is very widely used in digital image processing.

Fig. 3.1 Original Image is image of brain tumor having

noise and Filter image is noise removed image by Median
Filter.

IV.

FUZZY C MEANS

In fuzzy clustering (also referred to as soft clustering),

data elements can belong to more than one cluster, and
associated with each element is a set of membership
levels. These indicate the strength of the association
between that data element and a particular cluster [4].
Fuzzy clustering is a process of assigning these
membership levels, and then using them to assign data
elements to one or more clusters. One of the most widely
used fuzzy clustering algorithms is the Fuzzy C-Means
(FCM) Algorithm (Bezdek 1981).
In the FCM approach, the same given datum does not
belong exclusively to a well-defined cluster, but it can be
placed in a middle way as shown in the diagram below. In
this case, the membership function follows a smoother line
to indicate that every datum may belong to several clusters
with different values of the membership coefficient [5].

MEDIAN FILTER
www.ijaert.org

International Journal of Advanced Engineering Research and Technology (IJAERT) 314

Volume 2 Issue 8, November 2014, ISSN No.: 2348 8190

Input Image Histogram Equalization

Contrast
Output
image

Median Filter

Noise
removed
Output image
Fuzzy C-means

Segmented
Output image

Fig. 5.1 Working of combined approach

Fig. 4.1 Membership function of fuzzy clustering

4.1. ALGORITHM
CLUSTERING

OF

FUZZY

MEAN

Step1. Choose a number of clusters in a given image.

Step2. Assign randomly to each point coefficients for
being in a cluster.
Step3. Repeat until convergence criterion is met.
Step4. Compute the center of each cluster.
Step5.For each point, compute its coefficients of being in
the cluster [4-5].

Figure 5.2: show the image formed by combined

Approach

Fig.4.2 Shows the image segmented by C means algorithm

V.

COMBINED APPROACH

In this for one input image HE is applied and the output

get from this HE, to the same output MF is applied and the
output get from MF, to the same output FCM is applied.
The output get from this is the better segmented output.

Fig. 5.2 this shows, combine output of all techniques.

5.1. RESULT

By combining HE & FCM, data loss is improved.

By combining HE and MF none of the factor is
improved
www.ijaert.org

International Journal of Advanced Engineering Research and Technology (IJAERT) 315

Volume 2 Issue 8, November 2014, ISSN No.: 2348 8190

By combining MF & FCM none of the factor is

improved.
By combined approach accuracy is improved.

VI.

CONCLUSION

The techniques are developed in MATLAB lab for

analysis and comparisons. Combined approach produces
fairly higher accuracy as compare to FCM. The future
scope of combined approach is to decrease the
computation time and increase a PSNR, which generally
indicates that the reconstruction is of higher quality.

Karel Zuiderveld ,Adaptive Histogram Equalization

and its variation, 1987 by Academic press.
Books:
[11]
Gonzalez, Woods, and Eddins, Digital Image
Processing Using MATLAB, Prentice Hall 2004. (Ch.
3, Ch. 5).

REFERENCES
Journal Papers:
[1] M. M. Abdelsamea ,An Enhancement Nieghborhood
connected segmentation for 2D cellular Image,
International Journal of Bioscience, Biochemistry and
Bioinformatics, Vol. 1, No. 4, November 2011.
[2] Gunwanti S. Mahajan, Kanchan S. Bhagat, Medical
image segmentation using enhanced k-means and
kernelized fuzzy c- means , (ijecet) Volume 4, Issue 6,
November December, 2013.
[3] Rupinder Singh, Jarnail Singh, Preetkamal Sharma,
Sudhir Sharma, ---Edge based region growing,rupinder
Singh et al, Int. J. Comp. Tech. Appl., Vol 2 (4), 11221126 IJCTA | JULY-AUGUST 2011.
[4] Mrs. Bharati R.Jipkate, Dr. Mrs.V.V.Gohokar, A
Comparative Analysis of Fuzzy C-Means Clustering
and K Means Clustering Algorithms, IJCER , MayJune 2012 , Vol. 2 .Issue No.3, ISSN: 22503005 .
[5] B.Sathya, R.Manavalan, Image Segmentation by
Clustering
Methods:performance
Analysis,
International Journal of Computer Applications (0975
8887) Volume 29 No.11, September 2011.
[6] Jian-Ping Mei and Lihui Chen An Enhanced Fuzzy cMeans Clusterings using Relational Information,
2011 IEEE.
[7] Bei Yan, Mei Xie, Jing-Jing Gao, Weizhao ,A fuzzy
c-means based algorithm for bias field estimation and
segmentation of MR images , 2010 IEEE.
[8] R. Krishnapuram and J. Keller, A possibilistic
approach to clustering, IEEE Trans. Fuzzy Syst., vol.
1, no. 2, pp. 98110, Apr. 1993.
[9] Nikhil R.Pal and Sankar K. Pal, Review on Image
Segmentation Techniques , vol 26 March 1993.
[10]
Stephen M.Pizer, E. Philip Amburn, John D.
Austin, Robert Cromartie, John B. Zimmerman and

www.ijaert.org