Fusion of Mutiple Biometric Traits for Person Verification

In this paper we present an approach for person verification using multimodal biometric verification systems. One of the major applications of biometric verification systems is in the field of ATM, military and government applications, time & attendance records etc. A uni-modal biometric identification system is often not able to meet the system stringent performance requirements since it suffers from many limitations such as noise in sensed data, inter & intra class variations, distinctiveness, non-universality, spoof attacks etc. So, there is a need to combine multiple traits of individual for efficient performance requirements. We have used palm print, hand geometry and face as the input traits and fusion is applied at score level. The approach is divided into two parts: (i) template generation and (ii) verification. The proposed system is tested on a database of 100 persons and the experimental results indicate the efficient possibility of the combination.