Facial Expression Recognition under Partial Occlusion Based on Gabor Filter and Gray-Level Cooccurrence Matrix

This paper presents a novel facial expression recognition approach in the presence of partial occlusion using Gabor filters and gray-level co-occurrence matrix (GLCM). At first, we design an algorithm to extract the block Gabor feature statistics according to the spatial distribution of the face organ. Then, GLCM is firstly introduced into expression recognition field to make up for the deficiency of block Gabor feature, in which the association between pixels is absent. Finally, the block Gabor feature statistics is linear superimposed with the texture feature extracted by GLCM, after Gaussian normalization there generates a set of low-dimensional feature vectors for expression feature representation. The experimental results on JAFFE and RaFD show the high robustness and better recognition rates of the proposed novel approach under different types of occlusion.