Certain studies on improving the performance metrics in multimodal biometric system using optimization techniques

Abstract

In recent years, the identification and verification of human beings based on their physiological or behavioral characteristics is a trend in places with high security needs. The biometric techniques have been used to a great extent to provide protection against theft of identity. Biometrics is the science and technology of the measurement and due analysis of the human bodyand#8223;s biological data. Biometric System is the extraction of a set of features from the data that is acquired and the compared with the set of data in the template. The physiological biometrics is based on the data derived from a part of the human body which includes fingerprints, ear patterns, eye patterns, facial features, DNA, palm vein and hand vein. Behavioral identifier is based on the behavior of a person that includes voice, signature, gait, typing patterns and others. Unimodal Biometric System (UBS) is not widely newlineacceptable to a certain user group or a particular situation where security requirement is high priority in applications like financial, banking, border security, immigration purposes and forensics. In the real time applications, UBS suffers from various factors like lack of uniqueness, non-universality, restricted degree of freedom, noisy data and unacceptable error rates. To alleviate the above problems, a Multimodal Biometric System (MBS) is needed to improve the system accuracy. Yet, the accuracy can still be improved by choosing unique biometric traits and efficient algorithms. MBS has become a major area of study in human recognition system. newline newline