1/12/2021 dblp: DICTAP 2016

What do you think of dblp? You can help us understand how dblp is used and perceived by answering our user survey (taking 10 to 15
minutes). Your help is highly appreciated!

home | browse | search | about

search dblp

[+]
[–] 6. DICTAP 2016: Konya, Turkey
> Home > Conferences and Workshops > DICTAP Dagstuhl

Sixth International Conference on Digital Information and Communication Technology and its
Applications, DICTAP 2016, Konya, Turkey, July 21-23, 2016. IEEE 2016, ISBN 978-1-4673-9609-7

Ibrahim Aydin, Ümit Aygölü:

Energy-efficient link adaptation for distributed space time block code in wireless relay networks. 1-5

Murat Yücel , Nail Ferhat Ozturk, Cemal Gemci:

Design of a Fiber Bragg Grating multiple temperature sensor. 6-11

Yunus Sarikaya, Hazer Inaltekin , Tansu Alpcan, Jamie S. Evans:

Interference scaling in multi-cell network with rate guarantees. 12-17

Takashi Takeuchi, Yuki Nishikawa, Ryosuke Fujiwara:

Fast channel switching technique for interference avoidance with 5 GHz dual channel wireless LAN. 18-
23

Kerem Kucuk:
Design and implementation of a real-time ERP-OFDM SDR receiver on the USRP2 platform. 24-29

Muhammad Fahim, Asad Masood Khattak, Thar Baker , Francis Chow, Babar Shah:
Micro-context recognition of sedentary behaviour using smartphone. 30-34

Muhammad Fahim, Halil Ibrahim Cebe, Jawad Rasheed , Farzad Kiani :

mHealth: Blood donation application using android smartphone. 35-38
https://dblp.org/db/conf/dictap/dictap2016.html 1/4
1/12/2021 dblp: DICTAP 2016

Afdallyna Fathiyah Harun, Steve Benford, Claire O'Malley, Nor Laila Md. Noor:
Glyphing it! Children using markers in school. 39-44

Mostafa Korashy, Khaled. F. Hussain, H. M. Ibrahim:

Teleoperation of dogs using controlled laser beam. 45-49

Haipeng Wang, Sun Weiwei, Xueyuan Lin, Ziling Wang:

A centralized multi-sensor particle filter algorithm of formation targets. 50-55

Haipeng Wang, Yu Cong, Xinlong Pan, Weiwei Sun:

Track anti-bias refined correlation algorithm of group targets based on double fussy topology. 56-62

Hassan Saad Alqahtani, Paul Sant:

A multi-cloud approach for secure data storage on smart device. 63-69

Daniel Krauss, Christoph Thomalla:

Ontology-based detection of cyber-attacks to SCADA-systems in critical infrastructures. 70-73

Tanjila Farah, Moniruzzaman Shojol, Maruf Hassan , Delwar Alam:

Assessment of vulnerabilities of web applications of Bangladesh: A case study of XSS & CSRF. 74-78

Shahid Alam , Ryan Riley, Ibrahim Sogukpinar , Necmeddin Carkaci:

DroidClone: Detecting android malware variants by exposing code clones. 79-84

Shahid Alam:
SUIS: An online graphical Signature-Based User Identification System. 85-89

Saiful Izwan Suliman, Graham Kendall, Ismail Musirin:

Implementing immune-based algorithm for bandwidth spectrum allocation in wireless communication.
90-96

Sarah Alnasser, Hachemi Bennaceur:

An efficient Genetic Algorithm for the global robot path planning problem. 97-102

Can Umut Ileri , Orhan Dagdeviren:

Performance evaluation of distributed maximum weighted matching algorithms. 103-108

Ozan Zorlu , Ozgur Koray Sahingoz :

Increasing the coverage of homogeneous wireless sensor network by genetic algorithm based
deployment. 109-114
Haitham Alsaif, Naz Islam:

https://dblp.org/db/conf/dictap/dictap2016.html 2/4
1/12/2021 dblp: DICTAP 2016

Truncated compact design of monopole antenna with modified ground plane for UWB applications. 115-
118

Kishor Krishnan Nair, Albert Helberg , Johannes Van Der Merwe:

An approach to Improve the Match-on-Card Fingerprint Authentication System security. 119-125

Hazwani Mohd Mohadis , Nazlena Mohamad Ali :

Designing persuasive application to encourage physical activity at workplace among older workers. 126-
130

Taner Cevik, Ali Mustafa Ali Alshaykha, Nazife Cevik:

Performance analysis of GLCM-based classification on Wavelet Transform-compressed fingerprint
images. 131-135

Syed Yasser Arafat, Muhammad Inzimam Saghir, Mubah Ishtiaq, Umer Bashir:
Comparison of techniques for leaf classification. 136-141

Baha Fergani, Mohamed-Khireddine Kholladi :

A discrete particle swarm optimisation algorithm for geographical map contour reconstruction. 142-144

Necip Gözüaçik, Hatice Murtaza:

Mobile Number Portability service regulation interacts with national VoIP switch. 145-147

Nazanin Mazlooman, Behrouz Sahebkar:

Evaluate the effect of teledentistry and information technology on dentistry: Case study Iran. 148-155

Serkan Turan, Ekram Hussien Mohammed, Sarp Ertürk:

Hardware CODEC based JPEG2000 video encoder design for combat system video networks. 156-158

Islam A. T. F. Taj-Eddin, Mahmoud Afifi , Mostafa Korashy, Doha Hamdy, Marwa Nasser, Shimaa Derbaz:
A new compression technique for surveillance videos: Evaluation using new dataset. 159-164

Farag Sallabi, Khaled Shuaib:

Internet of things network management system architecture for smart healthcare. 165-170

Yasin Tekin, Ozgur Koray Sahingoz :

A Publish/Subscribe messaging system for wireless sensor networks. 171-176

Muhammad Aamir Nadeem, Taimur Karamat:

A survey of cloud network overlay protocols. 177-182
Musa Aydin , Emin Kugu :
Finding smoothness area on the topographic maps for the unmanned aerial vehicle's landing site
https://dblp.org/db/conf/dictap/dictap2016.html 3/4
1/12/2021 dblp: DICTAP 2016

estimation. 183-186

 last updated on 2021-01-11 00:10 CET by the dblp team

 all metadata released as open data under CC0 1.0 license 
see also: Terms of Use | Privacy Policy | Imprint

the dblp computer science bibliography is funded by:

       

https://dblp.org/db/conf/dictap/dictap2016.html 4/4
 Performance Analysis ofGLCM-based
Classification on Wavelet Transform-Compressed
Fingerprint Images

Taner (:evik Ali Mustafa Ali Alshaykha Nazife (:evik

Department of Computer Department of Computer Department of Computer
Engineering Engineering Engineering
Fatih University Fatih University Istanbul Arel University
Istanbul, Turkey Istanbul, Turkey Istanbul, Turkey

Abstract- Fingerprint detection is one of the primary
methods for identifying individuals. Gray Level
occurrence Matrix (GLCM) is the oldest and prominent
statistical textual feature extraction method applied in many
fields for texture analysis. GLCM holds the distribution of
co-occurring intensity patterns at a given offset over a given
image. However, images occupy excessive space in storage by
its original sizes. Thus, Discrete Wavelet Transform (DWT)
based compression has become popular especially
reducing the size of the fingerprint images. It is important to
investigate whether GLCM-based classification can
utilized efficiently on DWT-compressed fingerprint images.
In this paper, we analyze the performance of GLCM-based
classification on DWT-compressed fingerprint images. We
performed satisfying simulations for different levels ofDWT­
compressed images. Simulation results identify
classification performance sharply decreases by the increase
of DWT-compression level. Besides, instead of utilizing all
Haralick features, it is recognized that 8 of them are the most
prominent ones that affect the accuracy performance of the
classification.
Index Terms- Fingerprint Detection; GLCM; DWT;
Classification.
1. INTRODUCTION
Biometrics refers to the human specific behavioral and
physiological characteristics belonging to a person which
are distinctive of the other individuals and used for the
personal identification [1-2]. A biometric system includes
many distinctive personal biological characteristics such
as iris, fingerprint, voice, signature or face recognition that
are utilized for personal identification. Since biological
features are unique and thus more reliable than traditional
methods during individual identification, such systems are
more useful and suitable than conventional procedures.
Biometrics refers to the human specific behavioral and
physiological characteristics belonging to a person which
are distinctive of the other individuals and used for the
personal identification [1-2]. A biometric system includes
many distinctive personal biological characteristics such
Co­ as iris, fingerprint, voice, signature or face recognition that
are utilized for personal identification. Since biological
features are unique and thus more reliable than traditional
methods during individual identification, such systems are
more useful and suitable than conventional procedures.
Biometric systems are now commonly used in different
for parts of everyday life such as building access and computer
login. Fingerprint detection is one of the most prevalent
be methods for personal identification within all biometric
systems [3]. Throughout the world, fingerprint recognition
is accepted by a large part of the population because of its
fast, secure, and easy way of personal identification.
Fingerprints are the inherent parts of humans and unique
that
for each individual. Even twins have different fingerprints.
Using the patterns found on the finger tips, fingerprint
biometric data is captured. As with every technology,
using fmgerprint for biometrics has its own advantages and
disadvantages. Age of the user, dryness and wetness of the
finger, temporary or permanent cuts on the fmger, or dirt
on the fingerprint scanner all cause the fmgerprints to have
weak patterns of ridge and valleys that can degrade the
performance of a fingerprint biometric system. The
mentioned issues lead to poor recognition rates in
fingerprint identification system and users would
experience diffi cuIties with the system's authentication
process. Furthermore, as discussed before there can be
artificial fmgers which threaten system's security. For
creating artificial fingers a copy of the real fmgerprint is
required which can be obtained when the victim is
touching objects like glasses and can be replicated in
gelatin. For making the systems secure against artificial
fingers, liveliness detection can be used. Liveliness
detection searches for certain attributes which can only be
found on the live fingers such as the pulse in the fingertip.
However, individuals have different heart rate which can
be different from time to time. Besides, users need to keep
their fingers on the sensor without moving for few seconds

ISBN: 978-1-4673-9609-7 ©2016 IEEE 131

Authorized licensed use limited to: ULAKBIM UASL - Istanbul Arel University. Downloaded on January 11,2021 at 20:35:12 UTC from IEEE Xplore. Restrictions apply.
 to detect the fingertip pulse. However, there are artificial defmes a wavelet family If/.,,((jJ(s, r) E R+* xR) from
fingers called wafer thin artificial fmgers that can fool such . ,

a system by producing pulse. the mother wavelet:

Typically, fingerprints consist of ridge and valley 1 t-"C'
VtE R, If/,·,r(t) , If/( ) (1)
patterns on the tips of human fmgers. Thanks to their
= -

'\I s s
uniqueness and continuity, the use of fmgerprints is
considered to be one of the most reliable methods of
By extension, the families of functions on sub manifolds
personal verification. Due to the reasons such as
maintaining the public order as is proper, law enforcement of Rn invariant under a group of locally isomorphic to
requirements and increasing interest of civilian application the group affine transformation can also be described as
developers, automated fingerprint verification systems wavelet families. The Daubechies wavelet transfonnation
have become globally prevalent and been extensively is shown in Fig. 1. Tn this figure the red curve shows the
examined by the researchers and communities [4]. As a wavelet coefficient and the blue color shows the scaling
result, fmgerprint recognition is today being increasingly function. This curve is used for 4-tap Daubechies wavelet
used in diverse number of applications such as access which is used for implementation in this paper.
control or online identification [5]. Structure-based
Daubechies 4 tap wavelet
classification of fmgerprint images grooves local images
1.5
[6, 8] are used to classify. [6] In the fifth class to separate
classification accuracy report rate (with 18% rejection) is 1
90% and the source [7] to classify the four classes is
91.5%. A change in the classification of syntactic grammar 0. 5
fingerprint and hence their classification [9, 11] are used.
Good accuracy report [10] was in the classification of five o
classes (with 20% rejection) of 95.6%.
-0.5
In this paper, we perform a comprehensive performance
analysis of GLCM-based classification over DWT -1
compressed images. During DWT, Daubechies wavelets
are applied. First-level of wavelet transformation is -1.5 scaling function -­

utilized and subsequently the GLCMs are generated from wavelet function --

the resulting DWT images. We use 8 global features for -2

feature extraction. Seven diverse images of each individual
are used for training process.
The organization of the paper is as follows. Section IT
introduces some preliminary infonnation about GLCM­
bases feature extraction as well as the methodology and
calculations. Section III presents the results and
discussions among them. Finally, Section IV concludes the
article.
II. PRELIMINARIES AND METHODOLOGY
In this section, fundamental knowledge about the
methods applied in the study is introduced. As previously
mentioned, we analyze the classification performance of
GLCM-based classification over DWT compressed
images. Transfonnation is performed using Daubechies
wavelets. First-level of wavelet transfonnation is utilized
and subsequently the gray level co-occurrence matrix
(GLCM) is generated from the resulting DWT images.
Eight Haralick features are extracted from the GLCMs and
used for feature extraction. Seven images of each
individual are used for training process.
A. Discrete Wavelet Transform
Tn mathematics, a wavelet If/. is a square integral
function of the Hilbert space L2 (R) , usually oscillating
and zero mean, chosen as an analytical tool and
reconstruction multi- ladder.
Wavelets are usually found in families consisting of a
mother wavelet and all of this images by the elements of a
subgroup A group of affine transformations of Rn . This
o 0.5 1 1.5 2 2.5 3
Fig. 1. Example of Daubechies wavelet transform.
B. Gray-Level Co-occurrence matrix
GLCM which is introduced firstly by Haralick [12] is
one of the oldest and prominent statistical textual feature
extraction method [13] applied in many fields [14-17] for
texture analysis. GLCM is the matrix that holds the
distribution of co-occurring intensity patterns at a given
offset over a given image. Second-order statistical
(Haralick) features are extracted to analyze the texture of
the image which can subsequently be used for
classification tasks [18].
GLCM incorporates the spatial relationships of
intensity values with each other as well as their occurrence
quantities. Letfis an image whose intensity values vary in
the interval [0, L-Ij. Each element of GLCM indicates the
number of times that the pixel pair (Zi, Zj) occurred infwith
orientation Q. The orientation represented with Q
eventually represents a displacement vector d=(dx,dy I
dx=dy=dgJ where dg is the number of gaps between the
pixels of interest. For the situation of adjacency dg=O.
Orientation can also be represented with two parameters as
the distance d that the intensities Zi, Zj apart from each other
with angle a [19-20]. d can take values between 0 and L-2
theoretically. Orientation of the pixel pattern can be at four
different directions as 0°, 45°, 90° and 135° that a can take.
That is, each image can have four different GLCMs for
each angle (0°, 45°, 90° and 135°) for a specific d. The size
of a GLCM depends on the discrete intensity values in the

ISBN: 978-1-4673-9609-7 ©2016 IEEE 132

Authorized licensed use limited to: ULAKBIM UASL - Istanbul Arel University. Downloaded on January 11,2021 at 20:35:12 UTC from IEEE Xplore. Restrictions apply.
 image. If the intensity values of the image vary in the
interval [0, L-I} then the size of the GLCM becomes (L-
1) x(L-I). An example of configuration of the four GLCM
matrixes (GLCMoo, GLCM45o, GLCM90o, GLCM135o) of an (2)
image/for d=O is demonstrated in Fig. 2. k=l
The pOSItIOn of a point in an Euclidean n-space is
an Euclidean vector. Euclidean norm, Euclidean length,
or magnitude of a vector measures the length of the vector:

where the last equation involves the dot product.

m. TMPLEMENTATlON RESULTS AND DISCUSSIONS

The performance analysis has been tested on the lllT-D

SLF fingerprint database [22]. An example of a fmgerprint
Fig. 2. GLCM construction based on a (a) test image along four image for an individual is shown in Fig. 3.
possible directions (b) 0° (c) 45° (d) 90° and (e) 135° with a distance
d =0. The utilized database consists of 256x256 RGB colored
C. Haralick Features images. All RGB images are converted to gray scale
format to be compatible with the co-occurrence matrix in
Haralick [21] proposed 14 features that can be extracted the pre-processing stage.
from the abovementioned GLCMao matrices. Haralick
claimed that each of these 14 features gives hints about some
textual characteristic of an image. The Haralick features are
given below:
Angular Second f1� I'N=g1Ij=
Ng .. 2
1 per,})
Moment
Ng 1 2 Ng Ng .. }
Contrast
f2 - 'Ln�O n {'Li�l 'Lj�l p(t,J) li jl�n
_

Correlation - 'LNg 'LNg (i I'x)(i I'y)(p(i,j)

f3 - i�l j�l uxay

Sum of Squares f INg Ng 2 ..

4= i=l Ij=l Cl-IJ.) pel,])
Fig. 3. Sample fingerprint data.
f, Ng Ng
= 'Li�l 'Lj�11+(i_j)2 P Ct,J')
Inverse Difference 1
5 The flowchart of the analysis is shown in Fig. 4. As
Moment
shown in the corresponding flowchart, DWT of the images
Sum Average
f6 = 'L::� iPx+y(i) is performed for feature extraction which is subsequently
2
followed by Euclidean Distance-based classification.
Sum Variance
f7 = 'L::�(i - fa) Px+y(i)
Sum Entropy :)111"1.
fa= - 'L::� Px+y(i)log{Px+y(i)}
F.I'··'.i.I"
Entropy f9 = - 'Li 'Lj p(i,j)log{p(i,j)}
Difference Variance flO = variance of Px-y Fig. 4. Flowchart of the first analysis method.
Difference Entropy -1
fll = - 'L�:2 Px-y(i) log{Px_y(i)}
Fig. 5 demonstrates the classification performance for
HXY - HXY1 diverse number of training images without applying
f12= max HX HY GLCM. As clarified in this figure, the highest
Information { , }
Measures of classification accuracy is achieved for 5 training image
1/2
Correlation f13 = (1 - exp[ -2.0(HXY2 - HXY)J) data.
HXY= -'Li 'Lj p(i,j)logp(i,j)
. p(i,k)p(j,k)
Maximal Q (i,J)= 'Lk PxCi)py(k)
Correlation
Coefficient f14 = (Second Largest Eigenvalue of Q)"/2

D. Euclidean norm
For i = (h iJ, ..., in) andj = Ol,h, ...,jn) are two points
in Euclidean n-space, the Euclidean distance d from i toj,
or fromj to i is given by the Pythagorean formula:

ISBN: 978-1-4673-9609-7 ©2016 IEEE 133

Authorized licensed use limited to: ULAKBIM UASL - Istanbul Arel University. Downloaded on January 11,2021 at 20:35:12 UTC from IEEE Xplore. Restrictions apply.
 100
---a--- 1 image train
-- 2 image train
90 ---a--- 3 image train
------*-- 4 image train
'�"'j Li D ;. � 5 image train
�.

80
-----+--- 6 image trBin
-- 7 image train
70
C
'"
" 80
Q; Fig. 8. Flowchart of the GLCM feature extraction with DWT.
c..

50
In Fig. 9, the results for one-level DWT with 8 Haralick
40
features is illustrated. As shown in this figure, the highest
accuracy is obtained with 7 training image data.
30

20
10 15 20 25 30
person

Fig. 5. Classification performance with DWT feature extraction

without GLCM.

The flowchart of the operations utilized during second

analysis is presented in Fig. 6. As depicted in the
flowchart, GLCM is used for feature extraction without
DWT.

5 10 15 20 25 30
person

Fig. 9. Classification performance for one-level DWT with 8 Haralick

features.

Fig. 6. Flowchart of the second analysis method. IV. CONCLUSION

In Fig. 7 the classification performance for the GLCM Tn this study, we make a comprehensive performance
feature extraction without DWT is shown. As shown in analysis of three different methods for fingerprint
this figure the highest accuracy is acquired for 7 training identification. In the first method, feature extraction and
image data. classification is applied to the raw DWT utilized images.
Second method considers only GLCM based feature
100 ����----�------.-----�------,-----, extraction without DWT. Finally, GLCM-based feature
---e--- 1 image train
-- 2 image train
extraction and classification is applied to the DWT
90
---a--- 3 image train compressed images. The third method gives promising
-----+--- 4 image train
5 image train
classification performance regarding other methods. The
80
-----+--- 6 image train best identification accuracy around 93% is achieved with
-- 7 image train
7 training images for each individual. Results of the
C 70 simulations identifY that Increase In the DWT­
'"
f:!
t1J compressIOn level significantly decreases the
c..
80
performance.
50

40

30
5 10 15 20 25 30
person

Fig. 7. Classification performance with GLCM feature extraction

without DWT.

The flowchart of the third analysis method is shown in

Fig. 8. As shown in this figure GLCM-DWT are used in
combination for feature extraction.

ISBN: 978-1-4673-9609-7 ©2016 IEEE 134

Authorized licensed use limited to: ULAKBIM UASL - Istanbul Arel University. Downloaded on January 11,2021 at 20:35:12 UTC from IEEE Xplore. Restrictions apply.
 TABLE I. Classification Performances
N urn ber of training images for each oerson
Scenario Method
1 2 3 4 5 6 7
wavelet feature extraction
1 44.5751 63.5165 69.0334 62.4762 80.4706 61.0157 75.1084
(without GLCM)
GLCM feature extraction
2 67.5972 68.1317 71.5104 77.7631 69.9185 70.5839 81.0558
without DWT

3 one-level DWT with GLCM 56.6369 50.8868 44.8066 54.9318 75.7699 48.5057 93.2915

Metastasis Risk," Microsc. Microanal., vol. FirstView, pp.

1-9.2015.
ACKNOWLEDGEMENT
[16] Wang Wenbo , Wen Yusong, Dong Xue, Jin Xiaotong,
The authors wish to thank editors. This work was Kong Vida, Sea Ice Classification of SAR Image Based on
Wavelet Transform and Gray Level Co-occurrence Matrix,
supported in Fatih University.
Fifth International Conference on Instrumentation and
Measurement, Computer, Communication and Control
REFERENCES 2015,pp.1 04- 107
[17] Pratiwi. Alexander. J. Harefa. and S. Nanda.
[1] H. Vallabh, "Authentication using tinger-vein recognition,"
"Mammograms Classification Using Gray-level Co­
p. 198,2012.
occurrence Matrix and Radial Basis Function Neural
[2] A. Jain, L. Hong, and S. Pankanti, "Biometrics: Promising
Network," in Procedia Computer Science, 2015,vol. 59,pp.
frontiers for emerging identification market," Commun.
83-91.
ACM, vol. 43,no. 2,pp. 90-98,2000.
[18] X. Ou, W. Pan, and P. Xiao, "Tn vivo skin capacitive
[3] International Biometric Group, www.biometricgroup.com.
imaging analysis by using grey level co-occurrence matrix
[4] W. Sheng, G. Howells, M. Fairhurst, and F. Deravi, "A
(GLCM)," Int. J Pharm., vol. 460, no. 1-2, pp. 28-32,
memetic fingerprint matching algorithm," Ieee Trans. In[
2014.
Forensics Secur., vol. 2,no. 3,pp. 402-412,2007.
[19] E. Miyamoto and T. M. Jr., "Fast Calculation of Haralick
[5] A. K. Jain, A. Ross, and S. Pankanti, "Biometrics: A tool for
Texture Features, Human Computer Interaction Institute
information security," IEEE Trans. In! Forensics Secur.,
Department of Electrical and Computer Engineering
vol. 1,no. 2,pp. 125-143,2006.
Carnegie Mellon University," Hum. Comput. Interact. Inst.
[6] Polis akademisi ders notlan.
Dep. Electr. Comput. Eng. Carnegie Mellon Univ.
http://geocities.com/mdonmez1lfinger2.doc.
Pittsburgh PA, vol. 15213,pp. 1--6,2011.
[7] Wegstein, J. H., Computer Science and Technology An
[20] U. Ellis, "McGraw-Hill eBook Library.," Choice Curro Rev.
Automated Fingerprint, 1982.
Acad. Ubr., vol. 49,no. 12,p. 2242,2012.
[8] S. Wang, W. W. Zhang, and Y. S. Wang, "Fingerprint
[21] R. M. Haralick, K. Shanmugam, and 1. Dinstein, 'Textural
classification by directional tie Ids," in Proceedings - 4th
Features for Image Classification," IEEE Trans. Syst. Man.
IEEE International Conference on Multimodal Interfaces,
Cybern., vol. 3,no. 6,pp. 610-621,1973.
ICMI2002, 2002,pp. 395-399.
[22] A. Sankaran, T.1. Dhamecha, M. Vatsa, and R. Singh, On
[9] S. Shah and P. S. Sastry, "Fingerprint Classitication Using
Matching Latent to Latent Fingerprints, In Proceedings of
a Feedback-Based Line Detector," IEEE Trans. Syst. Man,
International Joint Conference on Biometrics, 20II.
Cybern. Part B Cybern., vol. 34,no. 1,pp. 85-94,2004.
[10] J.Cao and H.M.Dong, "Research on Syntactic Fingerprint
Classification and Matching ", Proc. of the second Int. Con.
on Machine Learning and Cybernetics, vol, 3, pp. 3066-
3071,2003.
[11] K. A. Nagaty, "Fingerprints classification using artificial
neural networks: a combined structural and statistical
approach," Neural Networks, vol. 14,no. 9,pp. 1293-1305,
2001.
[12] R. M. Haralick, "Statistical and structural approaches to
texture," Proc. IEEE, vol. 67,no. 5,pp. 786-804, 1979.
[13] A. Ali, X. Jing, and N. Saleem, "GLCM-based fingerprint
recognition algorithm," 2011 4th IEEE Int. Con[
Broadband Netw. Multimed. Technol., pp. 207-211,2011.
[14] T. Champion, C. Germain, J. P. Da Costa, A. Alborini, and
P. Dubois-Fernandez, "Retrieval of forest stand age from
SAR image texture for varying distance and orientation
values of the gray level Co-Occurrence matrix," IEEE
Geosci. Remote Sens. Lett., vol. 11,no. 1,pp. 5-9,2014.
[15] T. Vujasinovic, J. Pribic, K. Kanjer, N. T. Milosevic, Z.
Tomasevic, Z. Milovanovic, D. Nikolic-Vukosavljevic, and
M. Radulovic, "Gray-Level Co-Occurrence Matrix Texture
Analysis of Breast Tumor Tmages in Prognosis of Distant

ISBN: 978-1-4673-9609-7 ©2016 IEEE 135

Authorized licensed use limited to: ULAKBIM UASL - Istanbul Arel University. Downloaded on January 11,2021 at 20:35:12 UTC from IEEE Xplore. Restrictions apply.