Facial expression recognition under partial occlusion based on Weber Local Descriptor histogram and decision fusion

In order to solve the problem which the key information of facial expression missed under partial occlusion, this paper proposes a facial expression recognition method based on Weber Local Descriptor (WLD) histogram feature and decision fusion. Firstly, the image is divided into several non-overlapping rectangle regions with equal size and the WLD is used to extract the features of each region. The purpose of that is to extract face image spatial information and to be ready for the subsequent decision fusion. Secondly, each region is further divided into blocks and the histogram of each sub-block is computed and combined as the region features so as to extract the discriminative features accurately as many as possible from each region. Finally, in order to reduce the influence of partial occlusion for facial expression, we design a classifier for the histogram features of each region and the outputs of all classifiers are fused by decision rule to determine the class of test facial expression image. The proposed method achieves better performance in JAFFE database with eyes occlusion and mouth occlusion. Experimental results show that the method is robust to facial expression recognition under partial occlusion.