Facial expression recognition using embedded Hidden Markov Model

Embedded Hidden Markov Model (EHMM) has been applied to many areas due to its excellent features. In this paper, we present a novel method for Facial expression recognition by using the EHMM. We use five scales and eight orientations Gabor features to represent the expression image. Further, we use the EHMM to recognize the facial expression. In the EHMM structure, the super states are used to model the expression image along vertical direction while the inner states are used to model the expression image along horizontal direction. Our test results and analysis based on the JAFFE database demonstrate that the proposed method is effective and achieves higher average recognition accuracy (96.16%).