Recognizing facial images using Gabor Wavelets, DCT-Neural Network, Hybrid Spatial Feature Interdependence Matrix

Recognizing faces from images acquired from distant cameras are still a challenging task because these images are usually corrupted by various noises and blurring effects. In this paper we have developed and analyzed Gabor Wavelets, Discrete Cosine Transform (DCT)-Neural Network and Hybrid Spatial Feature Interdependence Matrix (HSFIM) for face recognition in the presence of various noises and blurring effects. We simulate the real world scenario by adding noises: Gaussian noise, Salt and pepper noise and also adding blurring effects: Motion blur and Gaussian blur. To compare the performance of Gabor Wavelets, DCT-Neural Network, and HSFIM we have considered six standard public face databases: IITK, ATT, JAFEE, CALTECH, GRIMACE, and SHEFFIELD.