Hybrid Approach for Template Protection in Face Recognition System

Biometrics deals with identifying individuals with the help of their biological (physiological and behavioral) data. The security of biometric systems has however been questioned and previous studies have shown that they can be fooled with artificial artifacts. Also biometric recognition systems face challenges arising from intra-class variations and attacks upon template databases. To tackle such problems, a hybrid approach for liveness detection and protecting templates in face recognition system is proposed. Here, the system captures input face image in three different poses (left, front, right) based upon the order chosen by the random select module. This approach will perform live face detection based upon complete body movement of the person to be recognized and template protection by randomly shuffling and adding the components of feature set resulting after fusion of three poses of input face image. It overcomes the limitations imposed by intra-class variations and spoof attacks in face recognition system. The resulting hybrid template will be more secure as original biometric template will not be stored in the database rather it will be stored after applying some changes (shuffling and addition) in its components. Thus the proposed approach has higher security and better recognition performance as compared to the case when no measures are used for live face check and template protection in database. Keywords-Liveness detection, template protection, face recognition, multiple sample fusion, eigen-coefficients