Abstract: The application of biometric authentication has no limit and the dimension of biometric feature can be extended to any level. Based on the biometric features, there are many authentication mechanisms discussed earlier but suffer with the problem of authentication efficiency where there will be micro differences between one to many people. The biometric authentication system uses the micro difference and unique feature of any person to perform authentication and the efficiency of the system is depending on how accurate the system is most methods suffer with false authentication and poor verification accuracy and to solve those issues a multi view edge model has been discussed in this study. This study presents a multi view edge model which computes sectional similarity on each edge model where there will be unique noticeable difference between any two persons. The method converts the sectional features of the multi view edge model into the sectional similarity matrix and the method extracts the features of nose, eye and mouth into a signature matrix. Using multi feature signature resemblance technique, the method computes the similarity between different features of the data set. The proposed method improves the performance of biometric authentication and reduces the false positive ratio in biometric authentication.
S. Kannan and V. Seenivasagam, 2016. Enhanced Biometric Authentication Using Multi Feature Signature Resemblance and Multi View Edge Sectional Similarity. Asian Journal of Information Technology, 15: 2542-2548.