Feature Accentuation Using Uniform Morphological Correction as Pre-processing Technique for DWT based Face Recognition

This paper proposes a novel pre-processing technique to enhance the performance of a Face Recognition (FR) system employing a unique combination of Uniform Morphological Correction (UMC) and Pose Invariant Flipping (PIF). Discrete Wavelet Transform (DWT) is used for efficient feature extraction and a Binary Particle Swarm Optimization (BPSO) - based feature selection algorithm is used to search the feature vector space for the optimal feature subset. Experimental results show the promising performance of the proposed technique on four benchmark face databases: Color FERET, Extended Yale B, Pointing Head Pose and CMU PIE.