DWT based Feature Extraction using Edge Tracked Scale Normalization for Enhanced Face Recognition

This paper proposes a novel preprocessing technique in order to achieve an increased recognition rate in face recognition (FR) systems. The proposed Edge Tracked Scale Normalization (ETSN) process involves the use of scale normalization along with edge detection as a preprocessing technique in order to eliminate unwanted background details. Feature extraction is performed on the preprocessed image using Discrete Wavelet Transform (DWT) and optimization in feature selection is achieved by Binary Particle Swarm Optimization (BPSO) technique. Computationally efficient FR system is obtained for Color FERET, Cambridge ORL, Extended YALE B and CMU PIE face databases using the proposed ETSN technique.