Multimodal Biometric System

Abstract

Single biometric system has certain inherent problems such as noisy sensor data, non-universality of biometric trait, restricted degrees of freedom and unaccepted error rates. Multibiometric system tries to overcome these problems by providing multiple evidences of the same identity. One other advantage is spoofing of these multibiometrics simultaneously become more difficult for an intruder. In this paper, iris and palmprint are the two biometrics used to build multibiometric system. Information presented by these two biometrics is fused to make the system efficient. Here fusion of two modalities is done at the matching score level by using a technique called sum rule. This paper presented experimental results of an Iris recognition, palmprint authentication and multibiometric system (fusion of iris and palmprint). An experimental result shows that the Multimodal biometric system is efficient compared to unimodal biometric system.