I.J. Mathematical Sciences and Computing, 2015, 3, 9-19
Published Online September 2015 in MECS (http://www.mecs-press.net)
DOI: 10.5815/ijmsc.2015.03.02
Available online at http://www.mecs-press.net/ijmsc
Analysis of Vascular Pattern Recognition Using Neural Network
Navjot Kaur
a
, Dr. Amardeep Singh
b
a
M.Tech Scholar, Punjabi University, Patiala (147002), India
b
Professor, Punjabi University, Patiala (147002), India
Abstract
Biometric identification using vein patterns is a recent technique. The vein patterns are unique to each
individual even in twins and they don‟t change over age except their size. As veins are beneath the skin it is
difficult to forge. BOSPHOROUS hand vein database is used in this work. Hand vein images are uploaded first
and key points using Scale Invariant Feature Transform (SIFT) are extracted. Then the neural network is used
for training these images. Finally neural network is used for testing these images to check whether the image
used for testing matches with the existing database or not. Results are computed like False Acceptation Rate
(FAR), False Rejection Rate (FRR), accuracy and error per bit stream.
Index Terms: Neural Network, Scale Invariant Feature Transform, False Acceptation Rate, False Rejection
Rate, Canny Edge Detector
© 2015 Published by MECS Publisher. Selection and/or peer review under responsibility of the Research
Association of Modern Education and Computer Science
1. Introduction
Biometric is the term used in computer science to refer to the field of mathematical analysis of unique human
features [2]. It refers to the identification of humans by their characteristics and traits. Traditionally
authentication is based on token based identification systems for example Passport, driving license etc. and
knowledge based system for identification such as a password, personal identification number etc. As threats
and attacks are increasing day by day, so there is a need of reliable security mechanism. Traditional methods
are not as efficient that can provide sufficient security, so biometric systems are created to overcome the
limitations of traditional systems. Biometrics is the science of identifying a person using its behavioral and
physiological features [2]. Biometrics systems are classified in 2 categories that are physical and behavioral.
Physical systems are related to the shape of the body such as fingerprints, face recognition, DNA, vascular
patterns, iris of the eye etc. Behavioral biometrics system are related to behavior of a person like voice, gait etc.
Advantages of these systems are that they are difficult to copy or forge; hence these systems are more secure
* Corresponding author. Tel.:
E-mail address: