Recognizing faces under facial expression variations and partial occlusions

Abstract

Recognizing faces under facial expression variations and partial occlusions is presented in this paper. PCA- and LDA-based approaches with the combination of the preprocessing techniques of histogram equalization and mean-and-variance normalization are used in order to reduce the effect of partial occlusions, facial expressions and illumination variations. Various distance measures are applied for classification under different facial expression variations. To be consistent with the research of others, our work has been tested on the JAFFE database and its performance has been compared with traditional PCA and LDA methods.