International Journal of Computer Applications (0975 – 8887)

Volume 35– No.5, December 2011

Segmentation of Textile Textures using

Contextual Clustering

Shobarani Dr.S.Purushothaman
Research Scholar, Dean, Udaya School of Engineering
Department of Computer Science, Vellamodi, Ammandivilai,
Mother Teresa Women’s University, India-629204.
Kodaikanal, India-624 101.

ABSTRACT the K-means defined as centers. The weighting parameter is

This paper presents texture segmentation concept using used to adjust the relative contribution of the color and texture
supervised method in contextual clustering and fuzzy logic. The features. The proposed algorithm is applied to color images in
data set used is the textile textures. The image is split into 3 X 3 the (Red, Green, Blue -RGB), (Hue, Saturation, Value-HSV)
windows. The features of the windows are presented to the input and (Improved Hue, Saturation, and Lightness-IHLS) color
layer of the contextual clustering. The algorithm involves least spaces. [8]Soo et, al., (2007) extracts texture features using the
computation in the segmentation of textures. The output of fuzzy wavelet packet frame decomposition followed by a Gaussian-
logic depends upon the radii of the clusters used during mixture-based classifier, which assigns each pixel to a class.
segmentation. The implementation of the algorithm is made by This scheme shows high recognition accuracy in the
the fuzzy membership its probability indicates the spatial classification of Brodatz texture images. Lin Ma et. al., (2009)
influence of the neighboring pixels on the centre pixel. proposes Ant Colony Optimization (ACO) for texture
segmentation[3]. The study uses the direct thresholding method,
Keywords which is based on wavelet transform smooth sub-images from
Image segmentation, clustering, fuzzy logic, textures. the use of a quadrant mean filtering method, to attenuate the
texture background and preserve the anomalies. The images are
1. INTRODUCTION then segmented by threshold processing and noise filtering.
In this research work, the investigations are carried out on textile [7]Shoudong Han et. al., (2011) proposes a novel texture
textures and mosaic using fuzzy logic, contextual clustering for segmentation approach using independent-scale component-wise
improved texture segmentation. Texture is the result of spatial Riemannian-covariance Gaussian mixture model (ICRGMM) in
variation in pixel intensities in an image. Texture is useful in a Kullback-Leibler (KL) measure based multi-scale nonlinear
variety of applications and has been a subject of intense study by structure tensor (MSNST) space.
many researchers. Recognition of homogeneous image regions
using texture properties is an important application. The goal is 3. SCOPE OF PRESENT WORK
to do classification of input image with the existing textures In this work, a systematic approach has been developed to
template. Texture segmentation is achieved using texture extract textures from the given texture images. The features are
properties. extracted using contextual Clustering. The extracted features are
used to segment the image using supervised method, fuzzy
2. PREVIOUS WORK logic. The major combinations of the algorithms developed are
Yining Deng et. al., (2001) proposed a class map method for as follows:
texture segmentation[10]. Applying the criterion to local
windows in the class-map results in the image, in which high (i) Contextual clustering using supervised method
and low values correspond to boundaries and interiors of color- (ii) Fuzzy logic using supervised method.
texture regions. A region growing method is then used to
segment the image based on the multiscale J-images. Nikos
Paragios (2002) presented variational framework for boundary 4. SIMULATION OF DATA
and region-based segmentation under a curve-based
optimization objective function[5]. The task of supervised
4.1 Data Collection
texture segmentation is considered to demonstrate the potentials A sample database has been presented in Table 1. The table
of the proposed framework. Arivazhagan et. al., (2003) presents few textile textures.
describes multiple scales based on block by block comparison of
wavelet co-occurrence features with template features[1,2]. Sara
et. al., (2006) use Uniform Local Binary Pattern (ULBP) method
to extract texture features[6]. Image segmentation is carried out
using the K-means algorithm on feature vectors, including color
and texture features. The distance measure is defined as a
function of the color and texture feature vector distances from

45
 International Journal of Computer Applications (0975 – 8887)
Volume 35– No.5, December 2011

5. ALGORITHMS PROPOSED FOR

Table 1 Sample Database.
The table1 presents few textile textures IMAGE TEXTURE SEGMENTATION
5.1 Contextual Clustering ( CC)
Picture Segmentation is a subjective and context dependent
cognitive process. In mathematical terms, this process is ill-
posed. i.e. there is no single goal for picture partition algorithms.
To make a picture segmentation is a well posed problem, it is
regarded as an optimization task featuring a firm statistic
foundation. Recently there has been considerable interest in
statistical clustering techniques in image segmentation inspired
by the methods of statistical physics, which were developed to
study the equilibrium properties of large, lattice based systems
consisting of identical interacting components. In a clustering
technique for image segmentation each pixel is associated with
one of the finite number of categories to form disjoint regions.

Contextual clustering algorithm segments a data into category 1

(ω0) and category 2 (ω1). They are assumed to be drawn from
standard normal distribution. The following steps are adopted
for implementing contextual clustering.

(i) Define decision parameter Tcc (positive) and weight

of neighborhood information β (positive). Let Nn be
the total number of data in the neighborhood. Let Zi
be the data itself, i.
(ii) Classify data with zi >Tα to ω1 and data to ω0. Store
the classification to C0 and C1.
(iii) For each data i, count the number of data u i,
belonging to class ω1 in the neighborhood of data i.
Assume that the data outside the range belong to ω0.
(iv) Classify data with

to ω1 and other data to ω0.

Store the classification to variable C2.
(v) If C2 ≠C1 and C2 ≠ C0, copy C1 to C0, C2 to C1
In this paper, for evaluating the proposed technique, here used
and return to step 3, otherwise stop and return to
two image databases that are called Tilda and Brodatz. Tilda
C2.
includes 8 groups of different fabric textile types and Each
group has 50 images with 512 x 768 pixels. Brodatz includes
112 types of fabric images with 640 x 640 pixels.
The implementing of the CC is given as follows:
4.1.1 Tilda
Tilda was developed within working group Texture Analysis of Step 1: Read a Pattern (texture image feature).
the DFGs (Deutsche Forschungs gemeinschaft). It is a textile Step 2: Sort the values of the Pattern.
texture database. This analyzed methods made it possible to
recognize and distinguish textures of varying kinds. In this Step 3: Find the Median of the Pattern Cm.
database contains the total of eight representative kinds of
Step 4: Find the number of values greater than the
textiles which is based on the analysis of textile atlases. There
Median Values, Um.
are 3200 TIF pictures with a storage capacity of 1.2 Gbyte in the
entire textile texture database., ASCII file exists in every picture Step 5: Calculate CC using Cm + (beta/Tcc) * (Um –
in the TILDA database. (bs/2)).
4.1.2 Brodtz Step 6: Assign CC as the segmented values.
For the Brodatz texture images, the page number of the
corresponding texture in the Brodatz book is in parenthesis. The
digital images from the texture book are not scanned the pages. Figure 1 shows a steady increase in the error as the beta value
The images are scans of a set of glossy black and white prints changes from 0.1 to 1. Hence, lower beta value is preferred for
were purchased by the author. The same textures of pictures are better estimation by CC.
prints as in the book, in most cases images are not same as the
one in the book.

46
 International Journal of Computer Applications (0975 – 8887)
Volume 35– No.5, December 2011

The segmentation analysis is shown in Figure 3.

14

12
Error between estimated

10
and target vakue

6 Series1
4

0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Beta vakue

Fig. 1 Impact of Beta value in CC estimation

(a) Original Image
Figure 2 shows a steady decrease in the error as the threshold
value changes from 10 to 100. Hence, higher threshold value is
preferred for better estimation by CC. (240) (210)

14

12
Error between estimated

10
and target value

6 (225) (230)
Series1
4

0
10
20
30
40
50
60
70
80
90
100

Threshold (b) CC segmented image

Fig. 2 Impact of Threshold value in CC estimation

47
 International Journal of Computer Applications (0975 – 8887)
Volume 35– No.5, December 2011

150 140

210 190

240 225

Fig. 4 Impact of threshold in segmentation

5.2 Fuzzy Logic

Weina Wang et al (2006) used the Many clustering strategies
such as the crisp clustering scheme and the fuzzy clustering
scheme, each of which has its own special characteristics[13].
Zadeh L.A.,(1965) proposed the conventional crisp clustering
method restricts each point of the data set to exclusively just one
cluster[14], which involves the idea of partial membership
described by a membership function, Bezdek J.C, Ferahta N
and Zhang D.Q used the fuzzy clustering as a soft segmentation
method has been widely studied and successfully applied to
image segmentation[16-22]. Bezdek J.C extracts among the
fuzzy clustering methods, fuzzy c-means (FCM) algorithm[15]
is the most popular method used in image segmentation because
it has robust characteristics for ambiguity and can retain much
(c) Histogram more information than hard segmentation methods discussed by
Fig. 3 CC segmentation analysis Bezdek J.C and Zhang D.Q et al[16,22]. Although the
conventional FCM algorithm works well on most noise-free
images, it has a serious limitation: it does not incorporate any
Impact of threshold in segmentation information about spatial context, which cause it to be sensitive
Original(6.bmp) Original(7.bmp) to noise and imaging artifacts. To compensate for this drawback
of FCM, the obvious way is to smooth the image before
segmentation. The training and testing fuzzy logic is to map the
input pattern with target output data. For this, the inbuilt
function has to prepare membership table and finally a set of
number is stored. During testing, the membership function is
used to test the pattern.
50 50

75 75

48
 International Journal of Computer Applications (0975 – 8887)
Volume 35– No.5, December 2011

Training Fuzzy logic more rules. When RADII is a scalar it is applied to all input and
output dimensions.
Read the pattern and its target value

Radii=0.1
Create Fuzzy membership function 1

0.8
Create clustering using K-Means algorithm

Outputs
0.6

0.4 Actual
Process with target values
output
0.2
Fuzzy
Obtain final weights 0 output

1
11
21
31
41
51
61
71
81
91
Fig. 5 Training Fuzzy Logic Window
Testing Fuzzy logic for texture segmentation

Input a pattern Fig. 7 Output of Fuzzy logic ( Radii=0.1)

Process with Fuzzy membership function

Radii=0.6
Find the cluster to which the pattern belongs
1.2
1
Obtain estimated target values
0.8
Output

Fig. 6 Testing Fuzzy logic Actual

0.6 output
Table 2 presents the segmented texture outputs for two different
0.4 Fuzzy
radii.
output
0.2
Radii in Fuzzy logic Segmented output
0.1(Figure 7) 0
1
10
19
28
37
46
55
64
73
82
91

Window

Fig.8 Output of Fuzzy logic (Radii= 0.6)

0.6(Figure 8) 6. CONCLUSIONS
In this work, texture segmentation for textile textures and
mosaic is considered. The conclusions of this research work are

(i)
The supervised contextual clustering extracts
features of the texture image which represents
information in a given window. The algorithm
involves least computation in the segmentation of
Table 2 Outputs of Fuzzy logic textures.
(ii) The output of fuzzy logic depends upon the radii
RADII specifies the range of influence of the cluster center for of the clusters used during segmentation.
each input and output dimension, assuming the data falls within In this work, suitable smoothing schemes have also been used to
a unit hyperbox (range [0 1]). Specifying a smaller cluster radius overcome some of the earlier reported limitations of the crisp
will usually yield more, smaller clusters in the data, and hence clustering scheme.

49
 International Journal of Computer Applications (0975 – 8887)
Volume 35– No.5, December 2011

7. REFERENCES [15] Bezdek J.C., Pattern Recognition with Fuzzy Objective

[1] Arivazhagan S., Ganesan L., “Texture segmentation using Function Algorithms, Plenum Press, NewYork 1981.
wavelet transform”, Pattern Recognition Letters, Volume [16] Bezdek J.C., Hall L.O., Clarke L.P., Review of MR image
24 Issue 16, pp. 3197 –3203, 2003. segmentation techniques using pattern recognition, Medical
[2] Arivazhagan S., Ganesan L., “Texture classification using Physics 20(4), 1993, pp. 1033–1048. Image segmentation
wavelet transform”, Pattern Recognition Letters, v.24 n.9- based on fuzzy clustering ... 147
10, p.1513-1521, 2003. [17] Ferahta N., Moussaoui A., Benmahammed K., Chen V.,
[3] Lin Ma, Kuanquan Wang, David Zhang, “A universal texture New fuzzy clustering algorithm applied to RMN image
segmentation and representation scheme based on ant segmentation, International Journal of Soft Computing
colony optimization for iris image processing”, Computers 1(2), 2006, pp. 137–142.
& Mathematics with Applications, v.57 n.11-12, p.1862- [18] Tolias Y.A., Panas S.M., On applying spatial constraints in
1868, 2009. fuzzy image clustering using a fuzzy rule-based system,
[4] Mohand Saïd Allili, Djemel Ziou, “Globally adaptive region IEEE Signal Processing Letters 5(10), 1998, pp. 245–247.
information for automatic color-texture image [19] Tolias Y.A., Panas S.M., Image segmentation by a fuzzy
segmentation”, Pattern Recognition Letters, v.28 n.15, clustering algorithm using adaptive spatially constrained
p.1946-1956, 2007. membership functions, IEEE Transactions on Systems,
[5] Nikos Paragios , Rachid Deriche, “Geodesic Active Man and Cybernetics, Part A 28(3), 1998, pp. 359–369.
Regions and Level Set Methods for Supervised Texture [20] Noordam J.C., Van Den Broek W.H.A.M., Buydens
Segmentation”, International Journal of Computer Vision, L.M.C., Geometrically guided fuzzy C-means clustering for
v.46 n.3, p.223-247, 2002. multivariate image segmentation, [In] Proceedings 15-th
[6] Sara Arasteh, Chih-Cheng Hung, “Color and texture image International Conference on Pattern Recognition, Vol. 1,
segmentation using uniform local binary patterns”, 2000, pp. 462–465.
Machine Graphics & Vision International Journal, v.15 n.3, [21] Ahmed M.N., Yamany S.M., Mohamed N., Farag A.A.,
p.265-274, 2006. Moriarty T., A modified fuzzy C-means algorithm for bias
[7] Shoudong Han, Wenbing Tao, Xianglin Wu, “Texture field estimation and segmentation of MRI data, IEEE
segmentation using independent-scale component-wise Transactions on Medical Imaging 21(3), 2002, pp. 193–
Riemannian-covariance Gaussian mixture model in KL 199.
measure based multi-scale nonlinear structure tensor [22] Zhang D.Q., Chen S.C., Pan Z.S., Tan K.R., Kernel-based
space”, Pattern Recognition, v.44 n.3, p.503-518, 2011. fuzzy clustering incorporating spatial constraints for image
[8] Soo Chang Kim, Tae Jin Kang, “Texture classification and segmentation, [In] Proceedings of International Conference
segmentation using wavelet packet frame and Gaussian on Machine Learning and Cybernetics, Vol. 4, 2003, pp.
mixture model”, Pattern Recognition, Volume 40 Issue 4, 2189–2192.
pp.1207-1221, 2007.
8. AUTHORS PROFILE
[9] Won W. K. g, C. W. M. Yuen D. D. Fan L. K. Chan E. H.
K. Fung, “Stitching defect detection and classification Shoba Rani received the MCA degree from Bharathidasan
using wavelet transform and BP neural network”, Expert University. She worked as a Lecturer in LWIASE, Chennai.
Systems with Applications: An International Journal, Also, she completed M.Phil. from Annamalai University. After
Volume 36 Issue 2, 2009. that she worked for two years as a Consultant in Computer
Science Dept. for TamilNadu Open University, Chennai.
[10] Yining Deng, b.s. Manjunath, “Unsupervised Segmentation
Presently she is working as a Assist.Prof in Dr. MGR
of Color-Texture Regions in Images and Video”, IEEE
University, Chennai. She is also a Ph.D. research scholar in the
Transactions on Pattern Analysis and Machine Intelligence,
department of Computer Science at Mother Teresa Women’s
v.23 n.8, p.800-810, 2001.
University, Kodaikanal.
[11] Workgroup on"Texture Analysis of DFG, “TILDA Textile
Texture Database” ,http://lmb.informatik.uni-freiburg.de/ Dr. S. Purushothaman, is presently working as Dean, Udaya
research/dfg-texture/tilde School of Engineering, Vellamodi, Ammandivilai, Nagercoil.
He is obtained his Ph.D. degree from IIT(M), Chennai. He has
[12] P.Brodatz, Textures: A Photographic Album for Artists and so far successfully guided nearly Ten Ph.D. Research Scholars.
Designers , Dover, New York, (1966)
[13] Weina Wang, Yunjie Zhang, Yi Li, Xiaona Zhang, The
global fuzzy c-means clustering algorithm,[In] Proceedings
of the World Congress on Intelligent Control and
Automation, Vol. 1, 2006, pp. 3604–3607.
[14] Zadeh L.A., Fuzzy sets, Information and Control, Vol. 8,
1965, pp. 338–353.

50