A Novel Nonuniform Division Strategy for Wearing Eyeglasses Face Recognition

In this paper, we propose a novel nonuniform division strategy for wearing-glasses face recognition based on Gabor filters and Local Binary Patterns (LBP) operator. The proposed method, which looks forward to weaking the effect of eyeglasses variation, divides a facial image into nonuniform regions, followed by using Gabor filters and LBP. Our experimental results on FERET and Yale database reveal that the proposed approach is validity and has better recognition performance than that obtained by using other traditional methods.