Gabor feature-based complete fisher discriminant framework for facial feature extraction

Abstract

In this paper, we propose a novel feature extraction approach using Gabor feature based complete fisher discriminant algorithm (GCFD). Four main steps are involved in the proposed GCFD: (i) Gabor features of different scales and orientations are extracted by the convolution of Gabor filter bank and original gray images; (ii) Complete fisher discriminant algorithm (CFD) is used for feature dimensionality reduction and to extract all discrimination information; (iii) Feature fusion algorithm and Euclidean distance based nearest neighbor classifier are finally used for classification. (iv)Simulation results show the effectiveness of our proposed GCFD.