MULTIMODAL BIOMETRIC SYSTEM USING SCORE LEVEL FUSION OF PALMPRINT AND FINGER KNUCKLE PRINT

Biometric based recognition systems are now effectively used in industries, educational institutions and banks for reliable personal identification. Among various biometric characteristics hand based biometrics has received greater attention among researchers because of its stability, feature richness, reliability and high user acceptability. In this paper, the finger knuckle print which refers to the inherent skin patterns that are formed at the joints in the finger back surface is used to extract the features. Speeded Up Robust features (SURF) and Empirical mode decomposition (EMD) are used to extract features from finger knuckle print. Score level fusion is used to combine the matching scores using the sum rule.The performance of the proposed algorithm is evaluated on the PolyU database. The proposed system is combined with the previous work using palmprint for personal identification. A multimodal system is thus developed based on score level fusion of palmprint and finger knuckle print. It provides a low value of false acceptance rate, false rejection rate and a high genuine acceptance rate in comparison to the unimodal system using palmprint or finger knuckle print.