Face Recognition Techniques - An evaluation Study

Int. J. Advanced Networking and Applications
Volume: 6 Issue: 4 Pages: 2393-2397 (2015) ISSN: 0975-0290
2393
Face Recognition Techniques - An evaluation
Study
Dr.Asmahan M Altaher
Department of Management Information System, Applied Science University, 166-11391, Jordan
Email: a_altaher68@hotmail.com
----------------------------------------------------------------ABSTRACT------------------------------------------------------------------
Face recognition is the ability of categorize a set of images based on certain discriminatory features.
Classification of the recognition patterns can be difficult problem and it is still very active field of research. The
paper introduces conceptual framework for descriptive study on techniques of face recognition systems. It aims to
describe the previous researches have been study the face recognition system, in order scope on the algorithms,
usages, benefits , challenges and problems in this felids, the paper proposed the face recognition as sensitive
learning task experiments on a large face databases demonstrate of the new feature. The researcher recommends
that there's a needs to evaluate the previous studies and researches, especially on face recognition field and 3D,
hopeful for advanced techniques and methods in the near future.
Keywords - Face recognition Techniques, recognition algorithm, automated Facial
System, evolutionary Pursuit, and Kernel Methods.
--------------------------------------------------------------------------------------------------------------------------------------------------
Date of Submission: October 30, 2014 Date of Acceptance: December 29, 2014
-------------------------------------------------------------------------------------------------------------------------------------------------
I. INTRODUCTION
The Humans have recognition capabilities that are
unparalleled in the modern computing era. Humans
have always had the a ability to recognize and distinguish
between faces features . Recently the computers have been
shown the same ability. In the mid of 1960s, scientists
began work on using the computer to recognize human
faces, since that time the facial recognition software has
come a long way, the governments and private companies
start to used the face recognition system . A lot of changes
are being made to enhance scientists capabilities in this
field (Zhang al et., 2009). Many researchers belief that
face recognition need to be about face detection, the prior
step to face recognition, the accurate detection of human
faces in arbitrary scenes, is the most important process
involved. Human faces are remarkably similar in global
properties, including size, aspect ratio, and location of
main features, but can vary considerably in details across
individuals, gender, race, or due to facial expression, most
traditional face recognition systems attempt to achieve a
low recognition error rate, implicitly assuming that the
losses of all misclassifications are the same (Shlizerman,
2011). Facial recognition software in any given episode,
the security department at the fictional. Montecito Hotel
and Casino uses its video surveillance system to pull an
image of a card counter, thief or blacklisted individual,
then runs that image through the database to find a match
and identify the person. In 2001, the Tampa Police
Department installed police cameras equipped with facial
recognition technology in their City nightlife district in an
attempt to cut down on crime in the area, the system failed
to do the job, and it was scrapped in 2003 due to
ineffectiveness (Patra and Das. 2008).The remainder of
this paper is organized as follows; the first Section was,
introduced framework and discuss -phase of this
framework from reviewed the previous studies. In Second
Section, were about methodology, finally, the researcher
results and concludes.
Research Framework
The paper descript the conceptual framework of face
recognition systems through scope on the algorithms,
usages, benefits, challenges and problems in this felids,
Figure (1)
Figure (1) Face Recognition Framework
Resource: researcher adapted
Literature review
D. Wilkes et. Al., (1992) start with a study related face
recognition system exploited the mobility of the camera by
used low-level image data to drive the camera to a
standard viewpoint with respect to an unknown object,
from such a viewpoint, the object recognition task is
reduced to a two-dimensional pattern recognition problem,
the system was used an efficient tree-based, probabilistic
indexing scheme to find the model object that is likely to
face recognition
techniques
APPLICATION&
ALHORITHMS
FACE RECGNITION
BENFITE
FACE RECGNITION
CHALENGES

2394
have generated the observed data, and for line tracking
uses a modification of the token-based tracking scheme of
The system has been successfully tested on a set of
origami objects. Given sufficiently accurate low-level data
(CHEN AL ET. 2005) (Jianke, 2009). Recognition time is
expected to grow only logarithmically with the number of
objects stored. (Xiong, 2010).
Thus, Dyar. C.R (1994) argued that there are two basic
behaviors that allow reconstruction of a patch around any
point in a reconstruction surface region. These behaviors
rely only on information extracted directly from face
images, and are simple enough to be executed in real time.
Global surface reconstruction can be provably achieved by
integrating these behaviors to iteratively “grow” the
reconstructed regions integrating these behaviors to
iteratively “grow” the reconstructed regions (Dyar.
C.R.,1994) (PIter, alet, 2008).
In the other hand Jianke Zhu,(2009) was present a fusion
approach to solve the nonrigid shape recovery problem,
which takes advantage of both the appearance information
and the local features, they have two major contributions.
First, they propose a novel progressive finite Newton
optimization scheme for the feature-based nonrigid surface
detection problem, which is reduced to only solving a set
of linear equations. The key is to formulate the non rigid
surface detection as an unconstrained quadratic
optimization problem that has a closed-form solution for a
given set of observations. Second, they proposed a
deformable Lucas-Kanade algorithm that triangulates the
template image into small patches and constrains,
algorithm for both efficient and effective issues.
However, YUN FU (2010) introduced paper about Age
estimation from facial images has promising applications
in human-computer interaction, biometrics, visual
surveillance, and electronic customer relationship
management, etc. Most existing techniques and systems
can only handle frontal or near frontal view age estimation
due to the difficulties of (1) differentiating diverse
variations from uncontrollable and personalized aging
patterns on faces and (2) collecting a fairly large database
covering the chronometrical image series for each
individual in different views, the researcher proposed a
robust framework to deal with multi view age estimation
problem, also a large face database, with significant age,
pose, gender, and identity variations, which should be
followed by classification or regression algorithms (YIN.,
2010).
Also Zheng, (2009) was written in the same topic and
focused on the human action recognition problem and
propose a new Curve-Distance approach based on the
geometry modeling of video appearance manifold and the
human action time series statistics on the geometry
information, he present experimental results on the KTH
database demonstrate the solution to be effective and
promising (Zhang, 2009).
Finally Gao.,et al.,(2008) was proposed a novel high-
order local pattern descriptor, local derivative pattern
(LDP), for face recognition. LDP is a general framework
to encode directional pattern features based on local
derivative variations. The nth-order LDP is proposed to
encode the (n-1)th -order local derivative direction
variations, which can capture more detailed information,
they discover that the high-order LDP consistently
performs much better than LBP for both face
identification and face verification under various
conditions. Next section will be on face recognition
applications
.
Applications of Face Recognition
Allot of application design are available to help in face
recognition, her bellow some of the applications are:
1- Application number one is related to find a face within
a large database of faces. In this approach the system
returns a possible list of faces from the database. The most
useful applications contain crowd surveillance, video
content indexing, and personal identification: driver’s
license, mug shots matching are examples of this type of
application (Wang. 2009) (Whaite et al., 1991).
2- Application number tow is about the real time face
recognition: Here, face recognition is used to identify a
person on the spot and grant access to a building or a
compound, thus avoiding security hassles. In this case the
face is compared will be against a multiple samples of a
person (Bartoli, 2006) (Wright, 2009).
Existing Algorithms
Face Recognition Based on Principal Component
Analysis.
Principal Component Analysis (PCA) is known as
algorithm that used in face recognition. The basic idea in
PCA is to determine a vector of much lower dimension
that best approximates in some sense a given data vector,
thus, in face recognition, it takes an s-dimensional vector
representation of each face in a training set of images
as input, and determines a t-dimensional subspace
whose basis vector is maximum corresponding to the
original image, the dimension of this new subspace is
lower than the original one (t <<s). If the original image
elements are considered as random variables, then the
principal components are along the eigenvectors
corresponding to the larger eigen values of the correlation
matrix, and error minimization is done in a least-squares
sense(Qing Chen, Xiaoli Yang, Jiying Zhao, 2006).
Face Recognition Based on Independent
Component Analysis
Like the PCA based technique, this is also a technique
to extract the statistics of the random variables. However,
in this case, the second-order and higher-order
dependencies of the input data are minimized by the
technique and the basis along which the data is statically
independent are found(Xu.,et al 2007)
Evolutionary Pursuit
This is an eigenspace-based approach that searches for
the best set of projection axes in order to maximize a
fitness function, measuring at the same time the
classification accuracy and generalization ability of the

Int. J. Advanced Networking and Applicatio
Volume: 6 Issue: 4 Pages: 2393-2397 (2
system. Because the dimension of the solu
too big, it is solved using a specific k
algorithm called; Evolutionary Pursuit
2010)
Figure (2) Face Recognition- eigensp
Recourse: Moghaddam et. al.,
Elastic Bunch Graph Matching
All human faces share a similar topolog
faces are represented as graphs, with node
fiducially points (eyes, nose, etc.) and edg
2-D distance vectors. Each node contain
complex Gabor wavelet coefficients at d
and orientations (phase, amplitude), th
"jets". Recognition is based on labeled gra
graph is a set of nodes connected by edge
labeled as jets, edges are labeled as distanc
2006).
Kernel Methods
The face manifold in subspace need n
Kernel methods are a generalization of li
Direct non-linear manifold schemes are
learn this non-linear manifold (Wiskott, al e
Trace Transform
The Trace transform, a generalization
transform, is a new tool for image process
be used for recognizing objects under tr
e.g. rotation, translation and scaling; to
Trace transform one computes a fun
tracing lines of an image. Different Trace
be produced from an image using d
functional (Hu L and Wei Z, 2009).
ons
015) ISSN: 0975-0290
ution space is
ind of genetic
(Yong, et al.,
pace-based
, 1998
gical structure;
es positioned at
es labeled with
ns a set of 40
different scales
hey are called
aphs, a labeled
s; each node is
ces (Zhao et al,
not be linear.
inear methods.
e explored to
et, 1995).
of the Radon
sing which can
ransformations,
o produce the
nctional along
transforms can
different trace
Figure (3) Tesco’s facial rec
Resource: Does Tesco’s facial recog
cross “the creepy lin
(Does, 2013).
Research Methodology
The researcher use descriptive st
framework which formulates the face
as a multiclass cost-sensitive learnin
algorithms was proposed to face recog
Benefits of face recognition systems
Understanding of face scanning bio
important tools for works controlling
biometric clocks can benefit compa
employees.
Benefit 1 - No More Time Fraud:
One of the big benefits of using facial b
in companies is that they weren't have
fraud, it will be impossible for buddy p
since everyone has to have go thru face
biometrics devices to clock (Yun. al., e
Benefit 2 - Better Security :
The companies also enjoy better se
biometrics system, not only can track t
biometrics time attendance tracking, b
be added to the system and tracked
too, anyone that is not in the system
access(Gao , B. Cao , S. Shan , X. C
Zhang and D. Z
Benefit 3 - Automated Facial System
Many companies like the fact that
systems are automated, they weren't h
having someone there to monitor the
day (Xuet al,. 2007).
2395
cognition
gnition ad targeting
ne
tudy to propose a
recognition problem
ng task, and related
nition.
ometrics security is
and monitoring, the
anies in monitoring
biometric systems
to worry about time
punching to occur,
e scanning
et 2010)
ecurity with a face
their employees thru
but any visitors can
throughout the area
m will not be given
Chen , D. Zhou , X.
Zhao ., 2008).
m:
biometric imaging
have to worry about
e system 24 hours a

2396
Benefit 4 - Easy Integration:
Integrated Biometric facial systems are also easy to
program into the companies computer system, usually they
will work with existing software that they have in place
(Wiskott, ,al et, 1995).
Benefit 5 - High Success Rate:
Facial biometrics technology today has a high success
rate, especially with the emergence of 3d face recognition
technologies, it is extremely difficult to fool the system,
and so you can feel secure knowing that your biometrics
computer security system will be successful at tracking
time and attendance while providing better security
(Jianke Zhu, 2009). Companies with a biometric security
device today can purchase a variety of security devices,
including safes, locks, time clocks, and more (Wilkes et
al., 1992).
Face recognition techniques challenging
One of the most important challenging of face recognition
is the fact that capturing technology requires cooperation
from a subject, for example lens or laser based scanners
require the subject to be at a certain distance from the
sensor, furthermore, a laser scanner requires a few seconds
of complete immobility. In addition, there are currently
very few high-quality face databases available for testing
and evaluation purposes, those databases that are available
should compare to databases used for ( Yun et al., 2010).
Furthermore, the comparison of the different face
recognition techniques is challenging for a number of
reasons:
Firstly; there are a few standardized face databases which
are used for benchmarking purposes, thus, the size and
type of 3D face datasets varies significantly across
different publications.
Secondly; there are differences in the experimental setup
and in the metrics which are used to evaluate the
performance of face recognition techniques, in terms of
the data and algorithms used and the reported recognition
performance. (Yong., et al. 2010).
Even though 3D face recognition is still a new emerging
area, there is a need to compare the strength of each
technique in a controlled setting where they would be
subjected to the same evaluation protocol on a large
dataset; this need for objective evaluation prompted
the design of the evaluation studies as well as the
upcoming. (Moghaddam, & Pentland., 1998) (Bartoli,
2006)
The principles of biometric evaluation laid down in the
evaluation strategy, so far, these evaluation studies are
limited to 2D face recognition techniques but will
hopefully include 3D face recognition techniques in the
near future
II. CONCLUSION
Face images are the inputs of the face
recognition system, there are a deferent algorithms, and
methods used for faces recognizing, even though, the
fields still have a challenges and limitations; it is still now
need more accurate techniques, especially for the 3D, also
the capturing technology requires of cooperation subject
still face a problems and, challenges, furthermore the
e x t e n s i o n of current face recognition has a bigger
problem of face detection, the network need parameters
leading to reduced poor generalization capabilities, and
face detection tasks are needed for given highs
images detection. Finally researchers find that the
change of recognition rate is not appreciable when the
number of hidden neurons is more than 40, the researcher
recommends that there's a needs for evaluation studies,
and researches especially on face recognition field and
3D, in order reduces problems and challenges in the
felids, will hopefully for new advanced techniques and
methods in the near future.
ACKNOWLEDGEMENTS
The author asmahan altaher is grateful to the
Applied science Private University , Amman , Jordan , for
the full financial support granted to this research project
(DRGS-2014-2015-37)
References
[1] A. Patra and S. Das. (2008) Enhancing decision
combination of face and fingerprint by exploitation of
individual classifier space: An approach to multi-modal
biometry. Pattern Recognition, 41(7):2298-2308.
[2] D. Zhang and Y. Wang. (2009)Gender Recognition
Based on Fusion of Face and Multi-view Gait.(2009) In
LNCS, volume 5558, pages 1010-1018. Springer, 2009.
[3]Hu L and Wei Z,(2009) "An Algorithm of Glass-Image
Recognition Based on Wavelet Packet Decomposition,
"International Conference on Computational Intelligence
and Natural Computing, vol.1, 2009, pp.206-209.
[4] Yun Fu,(2010) Age Synthesis and Estimation via
Faces: IEEE CONFERENCE PUBLICATIONS A Survey,
2010 , Page(s): 2590 - 2597
Cited by: Papers (3).
[5] Zhang. D and Y. Wang.(2009)" Gender Recognition
Based on Fusion of Face and Multi-view Gait". In LNCS,
volume 5558, pages 1010-1018. Springer,
[6] Whaite., p.and Ferrie,.P (1991) "From uncertainty to
visual exploration", IEEE T-PAMI, vol. 13, no. 10,
pp.1038 -1049 1991.
[7] Wilkes,. D. and Tsotsos,. J. K. (1992) "Active object
recognition", Proc. CVPR, pp.136 -141 1992
[8] Jianke Zhu,(2009) A Fast 2D Shape Recovery
Approach by Fusing Features and Appearance Pattern

2397
Analysis and Machine Intelligence, IEEE Transactions on,
Volume:31 , Issue: 7
[9]. Pilet, V. Lepetit and P. Fua,(2008) "Fast Non-Rigid
Surface Detection, Registration and Realistic
Augmentation," Int',l J. Computer Vision, vol. 76, no. 2,
pp. 109-122, 2008.
[10] Zhao W., R. Chellappa , P. J. Phillips and A.
Rosenfeld,(2006) "Face recognition: A literature survey",
ACM Comput. Surv., vol. 35, no. 4, pp.399 -459 2006.
[11] Bartoli, A (2006)"Groupwise Geometric and
Photometric Direct Image Registration," Proc. 17th British
Machine Vision Conf., Sept. 2006.
[12] Moghaddam.B, Wahid W. & Pentland A., (1998) "
International conference on face & Gesture
Recognition,Nara, Japan.
[13]http://www.troyhunt.com/,Does (2013) Tesco’s facial
recognition ad targeting cross “the creepy line”?,
November 6, 2013.
[14] Zheng, Haomian (2009) "Efficient human action
recognition by luminance field trajectory and geometry
information, Multimedia and Expo", 2009. ICME 2009.
IEEE International Conference on, Pages, 842 – 845
[15] Qing Chen, Xiaoli Yang, Jiying Zhao,(2005)
"Robust image watermarking with Zernike moments,"
IEEE 1-4 May 2005 Page(s):1340 – 1343.
[16] Wright, J.; Yang, A.Y.; Ganesh, A.; Sastry, S.S.; Yi
Ma,(2009) "Robust Face Recognition via Sparse
Representation," Pattern Analysis and Machine
Intelligence, IEEE Transactions on , vol.31, no.2,
pp.210-227, Feb. 2009
[17] Shlizerman, Kemelmacher- (2011)'3D Face
Reconstruction from a Single Image Using a Single
Reference Face Shape", Biometrics Compendium, IEEE,
Volume:33 , Issue: 2). 2011 , Page(s): 394 – 405.
[18]Yong sheng Gao ; Sanqiang Zhao ; Jianzhuang
Liu.,(2010)Local Derivative Pattern Versus Local Binary
Pattern: Face Recognition With High-Order Local Pattern
Descriptor, Image Processing,Biometrics Compendium,
IEEE Transactions on (Volume:19 , Issue: 2 ).
[19] Gao , B. Cao , S. Shan , X. Chen , D. Zhou , X. Zhang
and D. Zhao (2008) "The CAS-PEAL large-scale Chinese
face database and baseline evaluations", IEEE Trans.
Syst., Man., Cybern. A, Syst. Humans, vol. 38, no. 1,
pp.149 -161 2008.
[20]Xiong Li ; Xu Zhao ; Yun Fu ; Yuncai Liu , ( 2010)
Bimodal gender recognition from face and fingerprint. :
IEEE CONFERENCE PUBLICATIONS A Survey.
[21]X. Xu and T. S. Huang. (2007)SODA-Boosting and its
application to gender recognition. In IEEE International
Workshop on Analysis and Modeling of Faces and
Gestures, volume 4778, pages 193-204, 2007.
[22]Yin Zhang,(2010) " Cost-Sensitive Face Recognition",
Biometrics Compendium, 32 , Issue: 10 , Biometrics
Compendium, IEEE , 2010 , Page(s): 1758 – 1769
[23] Wiskott,L, Fellous, J. Kruger,N and C. von der
Malsburg.(1995)" Face recognition and gender
determination. In IEEE International Workshop on
Automatic Face and Gesture Recognition, 1995.
[24] Dyar. C.R (1994) "Global surface reconstruction by
purposive control of observer motion", Computer Vision
and Pattern Recognition, 1994. Proceedings CVPR '94.,
1994 IEEE Computer Society Conference .
Authors Biography
Dr.Asmahan M Altaher, Associate
professor Faculty of Business
Administration. Management Information
system Department –Applied Science
Private University, Amman- Jordan Sep –
2009 – till now.

Face Recognition Techniques - An evaluation Study

More Related Content

What's hot (19)

Viewers also liked (19)

Similar to Face Recognition Techniques - An evaluation Study (20)

More from Eswar Publications (20)

Recently uploaded (20)

Face Recognition Techniques - An evaluation Study