DWT Feature Extraction Based Face Recognition using Intensity Mapped Unsharp Masking and Laplacian of Gaussian Filtering with Scalar Multiplier

Face recognition under varying illumination and poses at certain angles is challenging, and hence improved edge prominence and contrast enhancement techniques are an effective approach to solve this problem. This paper proposes two novel techniques, namely, Intensity Mapped Unsharp Masking (IMUM) which provides a much finer outline of the face image by reducing the background intensity, and Laplacian of Gaussian based filtering with Scalar Multiplier (LOGSM) which provides an improved edge detection. Individual stages of the FR System are examined and an attempt is made to improve each stage. 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, obtained by applying the proposed algorithm on ORL, UMIST, Extended YaleB, ColorFERET face databases, show that the proposed system outperforms other FR systems. A significant increase in the overall recognition rate and a substantial reduction in the selected features are observed.