A GSA-based method in human identification using finger vein patterns

Abstract

Biometric and multibiometric science play an important role in human authentication systems nowadays. Finger vein pattern is one of the most reliable and secure biometrics due to its invariability and safety from stealth. In this paper, a heuristic method is proposed for score level fusion of three different finger vein's patterns. In the proposed multibiometric system, Gravitational search Algorithm is used to tune the weights of sum fusion strategy. The performance of the method is evaluated using FAR, FRR and EER criteria. Experimental results confirm the superiority of the proposed method over classic fusion strategy in human identification.