A multimodal biometrie system for personal identification based on deep learning approaches

Multimodal biometrie systems seek to alleviate some of the limitations of unimodal biometrie systems by combining multiple pieces of evidence of the same person in the deeision-making process. In this paper, a novel multimodal biometric identification system is proposed based on fusing the results obtained from both the face and the left and right irises using deep learning approaches. Firstly, the facial features are extracted using a Deep Belief Network (DBN) architecture consisting of 3-layers. The first two RBMs are used as features detectors, while the last one is used as a discriminative model associated with softmax units for the multi-class classification purpose. Secondly, an efficient deep learning system is employed for iris recognition, whose architecture is based on a combination of Convolutional Neural Network (CNN) and Softmax classifier to extract discriminative features from an iris image. Extensive experiments on large-scale challenging databases, including FERET, CASIA V1.0 and MMU1, and SDUMLA-HMT have demonstrated the superiority of the proposed approaches by achieving new state-of-the-art Rank-1 identification rates on all the employed databases.