A New Facial Expression Recognition Method Based on Geometric Alignment and LBP Features

Automatic facial expression recognition has been drawn many attentions in both computer vision and artificial intelligence (AI) for the past decades. Although much progress has been made, facial expression recognition (FER) is still a challenging and interesting problem. In this paper, we propose a new FER system, which uses the active shape mode (ASM) algorithm to align the faces, then extracts local binary patterns (LBP) features and uses support vector machine (SVM) classifier to predict the facial emotion. Experiments on the Jaffe database show that the proposed method has a promising performance and increases the recognition rate by 5.2% compared to the method using Gabor features.