Facial expression recognition based on completed local binary pattern and SRC

In this paper, we propose an effective algorithm for facial expression recognition (FER), which is based on completed local binary pattern (CLBP) and sparse representation. The new method solves sparse representations on both gray facial expression images and completed local binary pattern (CLBP) of these images. Afterwards, we obtain the both expression recognition results on both of expression features by sparse representation classification (SRC) method. Finally, the final expression recognition is obtained by fusion of the both results via comparing the residue ratios of sparse representations. The proposed method is experimented on Japanese Female Facial Expression (JAFFE) database. The experiment results show that the performance improves obviously by fusion approach. The proposed fusion algorithm is also assessed in comparison with the well known algorithms such as KPCA+SVM, LDA+SVM etc. The results illustrate that the proposed method has better performance than those traditional algorithms.