Facial expression recognition based on enhanced supervised locally linear embedding and local binary patterns

Facial expression recognition is an interesting and challenging subject in artificial intelligence, signal processing, pattern recognition, computer vision, etc. In this paper, a new method of facial expression recognition based on enhanced supervised locally linear embedding (ESLLE) and local binary patterns (LBP) is presented. The LBP features are first extracted from the original facial expression images. Then ESLLE is used to produce the low-dimensional discriminative embedded data representations from the extracted LBP features with striking performance improvement on facial expression recognition tasks. Finally, the nearest neighbor classifier is used for classification. The performance of ESLLE is compared with principal component analysis (PCA), linear discriminant analysis (LDA), locally linear embedding (LLE) as well as the original supervised locally linear embedding (SLLE). Experimental results on the popular JAFFE facial expression database demonstrate that the presented method of facial expression recognition based on ESLLE and LBP gives the best recognition accuracy of 81.14% with 30 reduced features, outperforming the other used methods.