A novel multibiometric template security scheme for the fusion of dual iris, visible and thermal face images

Multibiometric fusion is a very promising approach to improve the overall system's accuracy and the recognition performance. Due to the multi templates stored in the system, the protection of the multibiometric template is much more important than unimodal. In recent years, there are several approaches toward studying multibiometric template protection. However, most of them haven't considered the forgery, large difference of intra-class and the security of unimodal biometrics leakage. A novel multibiometric template protection scheme based on fuzzy commitment is proposed in this paper. Initially, the thermal and visible face images are fused to overcome the forgery problem. Second, the score level fusion based on Aczél-Alsina triangular-norm for left and iris are concentrated after the first stage to acquire the final recognition result. Eventually, the entropy of mutlibiometrics and unimodal information leakage is analyzed to show the security of the proposed approach. The experimental tests are conducted on a virtual multibiometrics database, which merges the challenging CASIA-Iris-Thousand database and the NVIE face database. The comparative experiments show that the proposed multibiometric template protection approach outperforms the unimodal biometric systems in terms of recognition and security.