Improved Locality Preserving Projections for Multimodal Biometrics

Because of its higher reliability, wider applicability and stronger security, multimodal biometrics has become a polar research direction of biometric recognition and attracts more and more research groups focusing on this area. Along with other fusion level of multimodal biometrics, feature level can reduce the redundant information to avoid calculation consumption, and simultaneously acquire the discriminative information to improve the system performance. This paper proposed improved locality preserving projection for multimodal biometrics that orthogonalized the projection vectors and took two distinct feature vectors as the real and imaginary part of complex vectors. Face and palm are selected as the experimental objects to evaluate the proposed algorithm. Experimental results shows the performance of our algorithm outperforms two unimodal biometrics and two traditional feature level multimodal biometrics.