Multimodal Biometric Identification Approach Based on Face and Palmprint

Multimodal biometric identification technique utilizes two or more individual modalities to improve the identification accuracy and overcome some problems existing in conventional unimodal methods. This paper presents a multimodal biometric identification approach based on the features of face and palmprint. Two feature extraction methods are employed, one is based on the statistics properties (SP) of the biometric image and the other is the classical two-dimensional principal component analysis (2DPCA). The minimal distance rule (MDR) is adopted for fusion at the matching score level. We compare the results of the multimodality identification with the results of the unimodal face and palmprint identification. The experimental results show that the performance of multimodality outperforms the unimodal identification and the accuracy can reach 100% based on ORL face database and PolyU palmprint database using the fusion rule at the matching score level.