Face Recognition Technique Using Gabor Wavelets And Singular Value Decomposition

Gabor Wavelets (GWs) (also known as Gabor filter) and Singular Value Decomposition (SVD) have been studied extensively in the area of face recognition. In this project, face recognition system is developed using combination of GWs and SVD. Both techniques are used to extract facial features from the human facial image and presented in the form of feature vector. For GWs, only 12 out of 40 GWs are selected to extract facial features from the facial images. This offers the advantage of reducing computational time of feature extraction. As for SVD, only the first five singular values are selected and its associated right singular vectors are used as the facial feature vectors. The use of SVD in addition to the GWs increases the reliability of the face recognition system. In the face verification and matching stage, the similarity level between facial images is determined by computing the distance between the resulting facial feature vectors obtained from GWs and SVD respectively. Overall, the Gabor-SVD based face recognition technique showed constructive and promising result in recognizing the valid user and rejecting invalid users on the JAFFE database.