Multimodal biometric systems by fusion of palmprint and Finger-Knuckle-Print using Hidden Markov Model

Biometrics technology has been attracting extensive attention due to the ever growing demand on access control, public security, forensics and e-banking. With the fast development of biometric data acquisition sensors and data processing algorithms, diverse biometric systems have been now widely used in various applications. Among these biometric technologies, the hand-based biometrics is most popular and has the largest shares in the biometrics market. The main objective of this paper consists of contributing in designing and developing efficiency hand based biometric algorithm for various applications. In this context, palmprint and Finger-Knuckle-Print (FKP) are used jointly for elaborating an efficient multimodal biometric recognition system. So, a fusion process is proposed for fusing these modalities. In this study, both modalities are characterized by the two dimensional Bloc based Discrete Cosine Transform (2D-BDCT) coefficients. Subsequently, we use the Hidden Markov Model (HMM) for modeling the observation vector. In addition, the two sub-systems are integrated in order to construct an efficient multimodal identification system based on matching score level fusion. The proposed scheme is tested and evaluated using a database of 165 users. The results show the effectiveness and reliability of the proposed method, which brings both high identification and accuracy rate.