Face recognition using entropy based feature enhancement and Parallel Dual Pose testing

One of the major challenges encountered by current Face Recognition (FR) techniques lies in the difficulties of handling varying poses, i.e., recognition of faces in arbitrary in-depth rotations. This paper proposes an Entropy-based Feature Enhancement technique as a preprocess to a FR system and a Parallel Dual Pose testing method to improve its recognition rate. The proposed feature enhancement technique makes use of the local entropy map of the intensity image in order to enhance regions of the image with maximum information. The Discrete Wavelet Transform along with the Discrete Cosine Transform have been used for feature extraction, supplemented by a Binary Particle Swarm Optimization algorithm for feature selection. Experimental results on four face databases (ColorFERET, HP, FEI and CMUPIE) show that the proposed FR system is effective and robust to variations in pose angle.