Alaa Eleyan, Hasan Demirel
Department of Electrical and Electronic Engineering, Eastern Mediterranean University
{alaa.eleyan, hasan.demirel}@emu.edu.tr
Abstract— this paper introduces a new face recognition method based on the graylevel cooccurrence matrix (GLCM). Both distributions of the intensities and information about relative position of neighbourhood pixels are carried by GLCM. Two methods have been used to extract feature vec tors from the GLCM for face classification. The first, method extracts the wellknown Haralick features to form the feature vector, where the second method directly uses GLCM by con verting the matrix into a vector which can be used as a feature vector for the classification process. The results demonstrate that using the GLCM directly as the feature vector in the rec ognition process outperforms the feature vector containing the statistical Haralick features. Additionally, the proposed GLCM based face recognition system outperforms the well known techniques such as principal component analysis and linear discriminant analysis.
Index Terms— gray level cooccurrence matrix, Haralick features, face recognition.
I.
INTRODUCTION
Face recognition has always been attracting the attention of the researchers as one of the most important techniques for human identification. One of the limitations of realtime recognition systems is the computational complexity of the existing approaches. Many systems and algorithms have been introduced in the last couple of decades with high recognition rates. The general problem of these systems is their computational cost in data prepreparation and trans formation to another domains such as eigenspace [3][4], fisherspace[5][6], wavelet transform [10][11] and cosine transform [12]. Many researchers have used the graylevel cooccurrence matrix [13] for the extraction of texture features to be used in texture classification. Gelzinis et al. [14] presented a new approach to exploiting information available from the cooccurrence matrices computed for different distance parameter values. Attempt to extend the approach was in troduced in [16], where a new matrix called motif co occurrence matrix was proposed. Walker et al. [15] have proposed to form cooccurrence matrixbased features by weighted summation of cooccurrence matrix elements from localized areas of high discrimination. The colour co occurrence histograms was used in [17] for object recogni tion, where they used false alarm rate to adjust some algo rithm’s parameters for optimizing the performance. The idea behind this paper is to use this cooccurrence ma trix and its extracted features in face recognition. To the
best of our knowledge, no one has attempted to implement this method for face recognition before. The idea is simple and it is straight forward. For each face image, a feature vector is formed directly by converting the generated GLCM to a vector and used for classification. Additionally, Haralick features [13] containing 14 statistical features can be extracted from the GLCM to form a new feature vector with 14 coefficients. The GLCM method is compared with two wellknown face recognition techniques. The first one is the principal com ponent analysis (PCA) [1][2], which is the standard tech nique used in statistical pattern recognition and signal processing for dimensionality reduction and feature extrac tion. As patterns often contain redundant information, mapping them to a feature vector helps to get rid of redun dancies and yet preserve most of the intrinsic information. These extracted features have great role in distinguishing input patterns. The proposed GLCM based method is also compared with the other wellknown technique, linear dis criminant analysis (LDA), which overcomes the limitations of PCA method by applying the Fisher’s linear discrimi nant criterion. This criterion tries to maximize the ratio of the determinant of the betweenclass scatter matrix of the projected samples to the determinant of the withinclass scatter matrix of the projected samples. Fisher discriminant analysis method groups images of the same class and sepa rates images of different classes. Images are projected from N dimensional space (N is the number of pixels) to C di mensional space (where C is the number of classes). Unlike the PCA method that extracts features to best repre sent face images; the LDA method tries to find the sub space that best discriminates different face classes and en codes discriminatory information in a linear separable space, where the bases are not necessarily orthogonal. The paper is organized as follows: second section discusses the gray level cooccurrence matrix. Third section explains 14 wellknow GLCM Haralick features. The databases are given in section four, while the experimental results, com parison with other algorithms and discussions are given in fifth section followed by the conclusions in the final sec tion.
II.
GRAYLEVEL COOCCURRENCE MATRIX
One of the simplest approaches for describing texture is to use statistical moments of the intensity histogram of an image or region [9]. Using only histograms in calculation, will result in measures of texture that only carry informa
tion about distribution of intensities, but not about the
relative position of pixels with respect to each other in that
texture. Using a statistical approach such as cooccurrence
matrix will help to provide valuable information about the
relative position of the neighbouring pixels in an image.
Given an image I, of size N×N, the cooccurrence, matrix
P can be defined as: 

N 
N 
1, 
if I 
( 
x 
, 
y 
) 


P (, i 
j 
) 


x 

1 
y 

1 

0, 
otherwise 
i
&
I
(
x
,
y
)
x
j
(1)
where the offset ( _{x} , _{y} ), is specifying the distance be
tween the pixelofinterest and its neighbour. Note that the
offset ( _{x} , _{y} ) parameterization makes the cooccurrence
matrix sensitive to rotation. Choosing the offset vector, so
a rotation of the image not equal to 180 degrees, will
result in a different cooccurrence matrix for the same
(rotated) image. This can be avoided by forming the co
occurrence matrix using a set of offsets sweeping through
180 degrees at the same distance parameter to achieve a
degree of rotational invariance (i.e. [0 ] for 0 ^{o} : P hori
zontal, [ ] for 45 ^{o} : P right diagonal, [ 0] for 90 ^{o} : P
vertical, and [  ] for 135 ^{o} : P left diagonal).
Figure 1 shows how to generate the four cooccurrence
matrices using N _{g} =5 levels and offsets {[0 1], [1 1], [1
0], [1 1]} defined as one neighboring pixel in the
possible four directions. We can see that two neighboring
pixels (2,1) of the input image is reflected in P _{H}
concurrence matrix as 3, because there are 3 occurrences
of the pixel intensity value 2 and pixel intensity value 1
adjacent to each other in the input image. The neighboring
pixels (1, 2) will occur again 3 times in P _{H} , which makes
these matrices symmetric. In the same manner, the other
three matrices P _{V} , P _{L}_{D} , P _{R}_{D} are calculated.
represents the number of neighboring pixel pairs used in
calculating GLCM, to form a new normalized matrix p
which will be used later on in the classification process or
in generation of Haralick features.
(a)
Figure 2 – (a) Example from ORL Database (112×92),
(b) corresponding CLGM with 64 gray levels (64×64) and
(c) corresponding CLGM with 256 gray levels (256×256).
III.
HARALICK FEATURES
In 1973, Haralick [13] introduced 14 statistical features.
These features are generated by calculating the features for
each one of the cooccurrence matrices obtained by using
the directions 0 ^{o} , 45 ^{o} , 90 ^{o} , and 135 ^{o} , then averaging these
four values. The symbol representing the offset parame
ter can be selected as 1 or higher value. In general value
is set to 1 as the offset parameter. A vector of these 14 sta
tistical features is used for characterizing the cooccurrence
matrix contents. Calculation of these features can be done
using the following equations.
0
1
2
3
Input image
P _{H} (0 ^{o} )
P _{V} (90 ^{o} )
Intensity
P _{R}_{D} (45 ^{o} )
P _{L}_{D} (135 ^{o} )
Figure 1 – Cooccurrence matrix generation for N _{g} =5 levels and four different offsets: P _{H} (0 ^{o} ), P _{V} (90 ^{o} ), P _{R}_{D} (45 ^{o} ), and P _{L}_{D} (135 ^{o} ).
Notations:
p(i, j) is the (i, j)’th entry in normalized GLCM.
N _{g} , dimension of GLCM (number of gray levels)
p _{x} (i) and p _{y} (j) are the marginal probabilities:
p
x ( )
i
N
g
( ,
pi
j
),
p
(
j
)
j
1
N
g
( ,
pi
j
).
i
1
is the mean of _{x} and _{y} ;
x
N
g
xy
ip
( ),
i
i
1
N
g
ip
( ).
i
i
1
_{x} and _{y} are the standard deviations of p _{x} and p _{y} , respec
tively.
N
g
N
g
1/2
x
2
( )(
pii
xx
)
i
1
2
( )(
pii
yy
)
i
1
,
y
1/2
The averaged GLCM matrix can be calculated by:
p
x
y
(
k
)
N
g
N
g
i
1 j
1
( ,
pi
j
);
i
jk
k
2, 3,
, 2
N
P
P
H
PP
V
RD
P
LD
4
For normalization, we
GLCM
matrix
P
by
then divide
normalization
(
k
)
N
g
N
g
i
1 j
1
( ,
pi
j
);
k
,
0,1,
N
1
N 
N 

HXY 1 
g g ( , p i j ) log{ p x ( ) i p y 
( j 

i 
1 j
1 

N N 

g g 

HXY 2 
p ( ) i p xy ( j ) log{ p 
( ) i 

i 
1 j
1 

Q (, i j ) 
N g k 1 N 
p ik (, ) p ( jk , ) p x () ip y ( j ) 

g 

HX _{} i 1 
( ) log{ pi x ( )} pi x 

N 

g 

HY _{} j 1 
( pj y ) log{ ( pj y )} 

Haralick Features: 

N 
N 

f 1 g g 
(, pi j ) 2 . 

i
1 
j
1 

N 1 
NN 

f 2
g gg n
0 2 n
11 ij
pij ,   ijn
. N g N g
i
x
i
y (, pi j ) 

f 3 
. 

i
1 
j
1 


x y 

N 
N 

f 4
g g 2( i
)
(, pi j ) 

i
1 
j
1 

N 
N 

f 5
g g 
1 2 (, pi j ) 

i
1 
j
1 
1
i
j


2 N 

f 6 g 

kp x 
y ( k ) 

k 2 

2 N 

g 

f 7
2
kf
6
k 2 p x
y ( k ) 

2 
N 

g 

f 8 _{} k 2 
pk
xy ( ) log{ pk
xy ( )} 

N N 

f 9 g 
g ( , pi j ) log{ ( , pi j )} 

i 
1 j

1 

N 
1 
N
1 

f 10
g g [ k
lp xy
( )] l 2 p xy
( k ) 

k

0 
l
0 

N 
1 

g 

f 11 _{} pk xy
( k 0 ) log{ pk xy
( )} 

f 9

HXY 1 

f


12 max( 
HX HY , ) 

f 13 1 exp[ 2( HXY 2 f )] 9
1/2 f 14
second largest eignvalue of Q 
1/ 2 
)}
p
xy
(
j
)}
HX and HY are the entropies of p _{x} and p _{y} , respectively.
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
IV.
PROPOSED SYSTEM
We adopted two scenarios for the implementation of
GLCM based face recognition. In the first scenario, we
used the cooccurrence matrix directly by converting it into
a column vector for each image in the recognition process.
In the second scenario, we used the vector of Haralick fea
tures extracted from the cooccurrence matrix for recogni
tion. Figure 3 shows these proposed scenarios.
Figure 3 – Proposed scenarios for using cooccurrence matrix and
its features for face recognition.
In the first scenario, we are using the GLCM directly after
converting it to a column vector. However, for 8bit gray
level representation the size of this vector is 256×256,
which corresponds to a 65536 dimensional vector. The di
mension of this vector can be reduced by reducing the
number of quantization/gray levels, N _{g} , of the image. For
example, if N _{g} is reduced to 16, the GLCM will be 256
(16×16) dimensional vector. The effects of dimensionality
reduction on the recognition performance are studied in the
following section.
In the second scenario, a vector of 14 Haralick features is
used to represent the given image. The size of this vector,
which is 14, is independent of the size of the GLCM.
_{V}_{.} _{S}_{I}_{M}_{U}_{L}_{A}_{T}_{I}_{O}_{N} RESULTS AND DISCUSSIONS
The simulations have been conducted on the two data sets:
FERET and ORL face databases. The standard test bed
adopted in similar studies for the FERET database [7], has
been used to test our algorithm for identification. The re
sults are reported in terms of the standard recognition
rates (%). 600 frontal face images out of 200 subjects are
selected, where all the subjects are in the upright and fron
tal position. Each face image is cropped to the size of
128×128 to extract the facial region and normalized to
zero mean and unit variance. In order to test the algo
rithms, two images of each subject are randomly chosen
for training, while the remaining one is used for testing
(i.e. 400 training and 200 test images). In Figure 4 the first
two rows are the sample training images while the third
row shows samples from test images. It can be observed
from this figure that the test images contain variations in
illumination and facial expressions.
The second database used, was the face database from Oli
vetti Research Lab [8]. This database is widely used for
evaluating face recognition algorithms. The database con
sists of 400 images acquired from 40 persons. All images
were taken under a dark background. The images are grey
scale with a 92×112 pixels resolution.
Out of the 10 images per subject of the ORL face data
base, first n were selected for training and the remaining
10n were used for testing. Hence, no overlap exists be
tween the training and test face images. Figures 4 and 5
show examples of both databases.
Figure 4 – Example FERET images used in our experiments. The top two rows the examples of training images used and the bottom row shows the examples of test images.
Figure 5 – Example ORL images used in our experiments.
Table 1 shows the result of using GLCM column vector
directly (direct GLCM) on ORL database using L _{1} norm
distance, _{L}_{1} . Reduced number of gray levels has been
used for GLCM dimensionality reduction. The training set
has been formed by using n=5 samples of each individual,
where the test set has been formed by using the remaining
5 samples. The results indicate that the direct GLCM ap
proach outperforms the Haralick features based approach.
It is also, important to note that, the dimensionality reduc
tion based on the reduction of the number of gray levels
does not change the recognition performance significantly
as we reduce the number of gray levels from 256 down to
16 levels. Highest performance is achieved by using 64
gray levels.
Table 2 shows the results on FERET database using L _{1}
norm distance. Different number of gray levels has been
used for GLCM dimensionality reduction. The training set
has been formed by using 2 samples of each subject,
where the test set has been formed by using the remaining
1 sample per subject. Similarly, the results indicate that
the direct GLCM approach outperforms the Haralick fea
tures based approach. Highest performance is achieved
by using 32 gray levels.
Table 1 – Face recognition performance (%) for different number of levels using ORL database and similarity measures: _{L}_{1} , with 5 training images / 5 testing images
Number of Gray Levels 

Approach 
256 
128 
64 
32 
16 
8 
4 
Haralick features 
71.90 
70.20 
70.50 
74.90 
79.50 
78.30 
69.60 
Direct GLCM 
95.10 
96.00 
96.30 
95.90 
95.20 
91.20 
70.40 
Table 2 – Face recognition performance (%) for different number of levels using FERET database and similarity measures: _{L}_{1} , with 2 training Images / 1 testing Images
Number of Gray Levels 

Approach 
256 
128 
64 
32 
16 
8 
4 
Haralick features 
29.33 
29.67 
26.17 
26.67 
34.50 
36.83 
13.16 
Direct GLCM 
88.67 
90.67 
91.67 
92.33 
88.67 
74.83 
14.33 
The results in both tables indicate that, the proposed direct
GLCM method is much more suitable method than the
method using wellknown Haralick features. The GLCM
scalability, provided by the scalable number of gray levels
N _{g} , is an important flexibility.
An important result, which can be observed in Tables 1 and
2, is that in some cases when the cooccurrence matrix be
comes smaller due to the use of reduced number of gray
levels, N _{g} , the recognition performance of direct GLCM
approach slightly increases. This can be due to the quantiza
tion process which suppresses the noise in the input image
that helps to increase the recognition performance.
Number of training images
Figure 6 – Face recognition performance of the Direct GLCM for ORL database, using different number of training images and different number of levels for generation of GLCM matrix, with the _{L}_{1} distance measure.
Figures 6 shows the performance of the direct GLCM
method for changing number of training samples in the
training set for different number of gray levels. The results
show that the recognition performance of the proposed sys
tem changes slightly as we reduce the number of gray levels
from 256 to 16. Although, this observation is specific to
these two datasets and cannot be generalized, the results
indicate the scalability of the proposed direct GLCM ap
proach.
Table 3 compares the performance of the proposed direct
GLCM based recognition system with the state of the art
techniques PCA and LDA by using the ORL face data
base. The results demonstrate the superiority of the pro
posed direct GLCM based face recognition system over
the well know face recognition techniques PCA and LDA.
Table 3 – Face recognition performance (%) of PCA, LDA, direct GLCM and Haralick features for different number of training images using ORL database with _{L}_{1} and N _{g} =256.
# Train 
Haralick Direct 

Images 
Features _{G}_{L}_{C}_{M} 
PCA 
LDA 
42.33 
61.44 
35.83 

56.50 
74.63 
76.88 

62.86 
82.86 
87.57 

68.33 
88.42 
93.25 

71.90 
90.00 
95.10 

74.50 
91.13 
96.35 

74.67 
94.00 
96.67 

76.00 
94.50 
97.25 

76.00 
97.50 
98.50 

VI. CONCLUSIONS 
In this paper, we introduced a new face recognition method,
direct GLCM, which performs recognition by using gray
level cooccurrence matrix. The direct GLCM method is
very competitive and outperforms the state of the art face
recognition techniques such as PCA and LDA. Using
smaller number of gray levels made the algorithm faster
and, at the same time, preserved the high recognition rate.
This can be due to the process of quantization which helps
in suppressing the noise of the images existing in higher
graylevels. Recognition in Table 1 for ORL database gen
erates the highest performance at N _{g} =64 which corresponds
to a moderate number of gray levels in image representa
tion. Same argument holds for FERET database as the rec
ognition was the highest at N _{g} =32 in Table 2. In general, it
is obvious from the results that the direct GLCM is a robust
method for face recognition with performance higher than
PCA and LDA techniques.
REFERENCES
[1] M. Kirby, & L. Sirovich, Application of the KarhunenLoeve Proce dure for the Characterization of Human Faces, IEEE PAMI, vol. 12, no. 1, pp. 103108, 01628828, January 1990. [2] L. Sirovich, & M. Kirby, LowDimensional Procedure for the Char acterization of Human Faces, JOSA, vol. 4, no. 3, pp. 519524, 1084 7529, March 1987. [3] M. Turk, & A. Pentland, Eigenfaces for Recognition, Journal of Cognitive Neuroscience, vol. 3, pp. 7186. 0898929X, 1991. [4] A. Pentland, B. Moghaddam, & T. Starner, View based and modular eigenspaces for face recognition, In Proceedings of Computer Vision and Pattern Recognition, pp. 84–91, 0818658258, IEEE Computer Society, Seattle, USA, June 1994. [5] P. Belhumeur, J. Hespanha, & D. Kriegman, Eigenfaces vs. Fisher faces: Recognition Using Class Specific Linear Projection. IEEE PAMI, vol. 19, no. 7, pp. 771720, 01628828, July 1997. [6] W. Zhao, R. Chellappa, & N. Nandhakumarm, Empirical perform ance analysis of linear discriminant classifiers. In Proceedings of Com
puter Vision and Pattern Recognition, pp. 164169, 0818658258, IEEE Computer Society, Santa Barbara, Canada, June 1998. [7] P. J. Philipps, H. Moon, S. Rivzi, & P. Ross, The Feret evaluation methodology fro facerecognition algorithms, IEEE PAMI, vol 22, pp. 1090–1104, 2000.
[8]
The Olivetti Database; http://www.camorl.co.uk/facedatabase.html.
[9] R. C. Gonzalez, & R. E. Woods, Digital Image Processing, 3rd Ed. Prentice Hall, 2008. [10] L. Wiskott, J.M. Fellous, N. Kruger, and C. von der Malsburg, “Face recognition by elastic bunch graph matching,” IEEE PAMI., vol. 19, no. 7, pp. 775–779, July 1997.
[11] C. Liu and H. Wechsler, “Independent component analysis of Gabor features for face recognition,” IEEE Trans. Neural Networks, vol. 14, no. 4, pp. 919–928, July 2003. [12] B. Manjunath, R. Chellappa, C. von der Malsburg, A feature based approach to face recognition. In: Proc. IEEE Conf. on Computer Vision and Pattern Recognition, pp. 373–378, 1992. [13] R.M. Haralick, K. Shanmugam, I. Dinstein, Textural features for
image classification, IEEE Trans. Syst. Man Cybern. vol. 3, no. 6, pp.
610–621, 1973. [14] A. Gelzinisa,A. Verikasa,b,, M. Bacauskienea, Increasing the dis
crimination power of the cooccurrence matrixbased features, Pattern
Recognition, vol. 40, pp. 23672372, 2004.
[15] R.F. Walker, P.T. Jackway, D. Longstaff, Genetic algorithm opti
mization of adaptive multiscale GLCM features, Int. J. Pattern Recogni
tion Artif. Intell. vol. 17, no. 1, pp.1739, 2003.
[16] N. Jhanwar, S. Chaudhuri, G. Seetharaman, B. Zavidovique, Con
tent based image retrieval using motif cooccurrence matrix, Image Vision
Comput. vol. 22, pp. 12111220, 2004
[17] P. Chang, J. Krumm. Object recognition with color cooccurrence
histograms. IEEE conference on computer vision and pattern recognition, Fort Collins, co, June 1999.