Skip to Main Content
Different facial expressions are related to a small set of muscles and limited ranges of motions. In this paper we propose an automatic facial expression recognition system, different from other automatic methods in both face detection and feature extraction. In system the facial expressions identify itself in video sequences. First, the differences between neutral and emotional states are detected. So as automatically locate faces and the facial organs which changes. Region-based method to extract LBP features is applied and AdaBoost is used to find the most important features for each expression on essential facial parts. At last, SVM with polynomial kernel is used to classify expressions. The method is evaluated on JAFFE database and obtains better recognition rate than other automatic or manual annotated systems.