Face verification using local binary pattern–unconstrained minimum average correlation energy correlation filters

Abstract

In this paper, we present a novel approach for face verification using local binary pattern (LBP) operators and optical correlation filters. Due to selected filter type and LBP method, different performances would have been expected. We let LBP operate on training images to form local binary pattern-unconstrained minimum average correlation energy (LBP-UMACE) filters as an optical correlation filter to enhance recognition rates and reduce error rates simultaneously. As a result, we demonstrate that the designed filters have better performance compared with UMACE filters. Moreover, the proposed filters are easy to implement and can be used in fast and reliable face verification systems.