Gabor-based Discriminant Locality Preserving Projections for face recognition

In this paper, a novel face recognition method called gabor-based discriminant locality preserving projections (Gabor-DLPP) is proposed. Locality preserving projections (LPP) is a recently proposed method for linear dimensionality reduction. Gabor-DLPP first gets the high-order statistic information by calculating the Gabor wavelet representation of face images. Based on LPP, Gabor-DLPP takes into account the between-class information, changes the objective function, and extracts the discriminant feature of face for recognition. We performed comparative experiments of various face recognition schemes, including PCA, Gabor-DLPP, LPP and LDA. Experimental results on AR and Yale database show the superior of the proposed method.