International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 2, Issue 6, June 2013
ISSN 2319 - 4847
A Survey on Enhanced Human Identification
Using Gait Recognition based on Neural
Network And Support Vector Machine
Supreet Kaur, and Er.Ranjeet Kaur Sandhu
M.Tech Student (CSE), GIMET, Amritsar
Assistant Professor in Dept. of IT, GIMET, Amritsar
Abstract
Gait recognition is one kind of biometric technology that can be used to monitor people without their cooperation. Controlled
environments such as banks, military installations and even airports need to be able to quickly detect threats and provide
differing levels of access to different user groups. Gait shows a particular way or manner of moving on foot and gait
recognition is the process of identifying an individual by the manner in which they walk. Gait is less unobtrusive biometric,
which offers the possibility to identify people at a distance, without any interaction or co-operation from the subject; this is the
property which makes it so attractive. This paper proposed new method for gait recognition. In this thesis I will present the
review of gait recognition system, different approaches and classification categories of Gait recognition like Model Free
Approach, GMM for background substraction, Silhouette Segmentation. We combinedly used SVM(Support vector Machine)
with NN(Neural Network) then compare it with BPNN
Keywords: GAIT Recognition, BPNN, SVM and NN
1.INTRODUCTION
Recognition of an individual is an important task to identify people. Identification through biometric is a better way
because it associate with individual not with information passing from one place to another. Biometrics is a
physiological or behavioral characteristic, which can be used to identify and verify the identity of an individual. There
are numerous biometric measures which can be used to help derive an individual identity. They are physiological, like
fingerprints, face recognition, iris-scans and hand scans and behavioral, like keystroke-scan and speech patterns. Gait
recognition is relatively new biometric identification technology which aims to identify people at a distance by the way
they walk. It has the advantage of being unobtrusive, difficult to conceal, noninvasive and effective from a distance.
Human gait recognition as a new biometric aimed to recognize person via the style of people walking, which contain
the physiological or behavioral characteristics of human.
A) Gait recognition system
System will identify unauthorized individual and compare his gait with stored sequences and recognize him.
Background subtraction is the common approach of gait recognition. Background subtraction method is used to subtract
moving objects and to obtain binary. Using background subtraction, preprocessing is done to reduce noise. Background
subtraction techniques are also classified into two types: non- recursive methods and recursive methods. Non recursive
techniques use sliding window approach for background subtraction. Recursive methods use single Gaussian method
and Gaussian mixture model. Gait recognition method contains two parts 1) training part 2) testing part.
B) Silhouette Segmentation
In gait recognition, silhouette is defined as a region of pixels of the walking person. Silhouette extraction mainly
focuses on segmenting the human body. The silhouette extraction process is shown in “Figure”.
Fig 1.1: Silhouette Extraction
Volume 2, Issue 6, June 2013
Page 24
International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 2, Issue 6, June 2013
ISSN 2319 - 4847
Each of the frames in the image sequence is subtracted from a background model of the respective image sequence. If
the pixel value of each frame is not the same with the pixel value of the background To remove shadow from the
difference image, a threshold value is applied to the difference images. The difference image map is first analyzed by
generating the intensity histogram of the image so that the pixels distribution along the image can be represented
clearly and in an effective way according to an applied threshold value. The threshold must be suitable so that the
foreground image is neither under segmented nor over-segmented. Under-segmentation and over-segmentation purpose
is to produce first and second reliable silhouette respectively.
To remove noises produced during segmentation of silhouette, morphological filters are used. The main components of
morphological filters that are used in the system are morphological opening, morphological closing and area
thresholding through connected component labeling.
C) Model Based Approach
Model based approach is difficult to follow in low resolution images also they have high computational complexity.
Advantage of this approach is the ability to derive gait signature from model parameter and free from the effect of
different clothing . Features used in this approach are insensitive to background cluttering and noise . Model based gait
recognition system includes motion of thigh and lower leg rotation that describes both walking and running . Lee and
Grimson used seven ellipses to model human body. Yamet.al used double pendulum to describe the thigh and lower leg
movement. Model based method construct human model to recover features describing gait dynamics such as stride and
kinematics of joint angle . Parameters used in this approach are height, distance between head and pelvis.
D) BPNN(Back Propagation Neural Network)
Neural networks which have been widely used in image and signal processing1 are very effective for solving multipleclass classification problems. Many researchers have successfully applied neural networks to face/gait recognition Chau
notes that neural networks facilitate gait recognition because of their highly flexible, inductive, and non-linear
modeling ability. In this paper, we use one classical type of neural networks –BPNN. BPNN usually has input and
output layers, with some hidden layers in between. Actually, BPNN can be likened to a flexible mathematical function
which has many configurable internal parameters. In order to accurately represent the complicated relationships among
gait variables, these internal parameters need to be adjusted through training process.
E) Neural Network:
Neural networks are composed of simple elements operating in parallel. These elements are inspired by biological
nervous systems. As in nature, the connections between elements largely determine the network function. You can train
a neural network to perform a particular function by adjusting the values of the connections (weights) between
elements. Typically, neural networks are adjusted, or trained, so that a particular input leads to a specific target output.
There, the network is adjusted, based on a comparison of the output and the target, until the network output matches
the
tar.
Typically,
many
such
input/target
pairs
are
needed
to
train
a
network.
Neural networks have been trained to perform complex functions in various fields, including pattern recognition,
identification, classification, speech, vision, and control systems. Neural networks can also be trained to solve problems
that are difficult for conventional computers or human beings. The toolbox emphasizes the use of neural network
paradigms that build up to—or are themselves used in— engineering, financial, and other practical applications.
F) Support Vector Machine (SVM):
The theory of SVM is based on the idea of structural risk minimization. In many applications, SVM has been
introduced as a powerful tool for solving classification problems. Consequently, many researchers have used SVM on
gait recognition. However, it is to be noted that SVM is fundamentally a two-class classifier. First class maps the
training samples into a high dimensions space and then finds a separating hyper plane that maximizes the margin
between two classes in this high dimension space.
2. WORK PLAN
Work plan is the master plan specifying the methods and procedures that are used to collect and analyze the needed
information. It is a blue print of describing number of activities we are performing in our research work. Following are
the steps for this task:
i. Research process starts with gathering the requirements and then we define the phase where requirements are
gathered for the research work that includes all the related material for the research, primary and secondary source of
data. After gathering the information and requirements for our research, next we will proceed to feasibility phase that
means whatever we decide to doing should be valid and should have not done before.
ii. The next step in research design is analysis phase that analyses all the data required and the information that we
gathered for our research.
iii. Next step is designing all the activities to be performed by us during our research process
Volume 2, Issue 6, June 2013
Page 25
International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 2, Issue 6, June 2013
ISSN 2319 - 4847
Figure : Flowchart
3.CONCLUSIONS AND FUTURE WORK
i. We will do simulation of proposed Algorithm in MATLAB (Image Processing Tool).
ii. We compare the proposed algorithm with the exiting approaches like BPNN
iii. We will use Model free Approach, GMM for background subtraction, Silhouette Segmentation considering the
following parameters:
 Distance between legs
 Distance between hand
 Distance between head and feet
REFERENCES
[1.] H. F. Alan J. Lipton and R. S. Patil, “Moving target classification and tracking from real-time video,” Proceedings
of Fourth IEEE Workshop on Applications of Computer Vision (WACV’98).
Volume 2, Issue 6, June 2013
Page 26
International Journal of Application or Innovation in Engineering & Management (IJAIEM)
Web Site: www.ijaiem.org Email: editor@ijaiem.org, editorijaiem@gmail.com
Volume 2, Issue 6, June 2013
ISSN 2319 - 4847
[2.] A.K,A.S ,Amit Kale, Aravind Sundaresan and R. Chellappa, “Identification of humans using gait,” IEEE
Transactions on Image Processing, vol. 13, no. 9.
[3.] A. K. R.-C. Aravind Sundaresan and R. Chellappa, “A hidden markov model based framework for recognition of
humans from gait sequences,” Proceeding of ICIP.
[4.] M. Bober, “Mpeg-7 visual shape descriptors,” IEEE Transactions on Circuit and Systems for Video Technology,
vol. 11, no. 6.
[5.] H. Z. Changhong Chen, Jimin Liang and H. HU, “Gait recognition using hidden markov model,” Lecture Notes on
Computer Science.
[6.] P. S. Huang, “Automatic gait recognition via statistical approaches for extended template features,” IEEE
Transaction on Systems, Man, and Cybernetics-Part B: Cybernetics, vol. 31, no. 5, October 2001.
[7.] L. Lee and W. E. L. Grimson, “Gait analysis for recognition and classification,” Proceedings of Fifth IEEE
International Conference on Automatic Face and Gesture Recognition., 2002.
[8.] W. H. Liang Wang, Tieniu Tan and H. Ning, “Automatic gait recognition based on statistical shape analysis,”
IEEE Transaction on Image processing, vol. 12, no. 9, September 2003.
[9.] “Silhouette analysis-based gait recognition for human identification,” IEEE Transaction on Image processing, vol.
25, no. 12, December 2003.
[10.] D. H. Nikolaos V. Boulgouris and K. N. Plataniotis, “Gait recognition: A challenging signal processing
technology for biometric identification,” IEEE sinal processing magazine, November 2005.
[11.] M. Piccardi, “Background subtraction techniques: A review,” IEEE International Conference on Systems, Man
and Cybernetics.
[12.] P. W. Power and J. A. Schoonees, “Understanding background mixture models for foreground segmentation,”
Proceedings of IEEE International Conference on Image and Vision Computing.
[13.] C. Stauffer and W. E. L. Grimson, “Adaptive background mixture models for real-time video tracking,”
Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 82, pp. 246–
252.
Volume 2, Issue 6, June 2013
Page 27