Abstract
The recognition of the human's emotion based on the facial image is accomplished by extracting the facial emotional features and classifying the emotional states based on extracted features. To realize this recognition procedure, we have proposed well-defined feature extraction and recognition methods based on reliable preceding research. The Active Shape Model (ASM) is a well-known method in which it can represent a non-rigid object, such as face, facial expression and facial image feature processing is faster than other method such as AAM. Also, the probabilistic reasoning of emotion states is accomplished by using Bayesian approach that can represent the causal relationship between a set of facial features for emotion recognition. In this paper, our approach to facial feature extraction lies in the proposed feature extraction method based on combining ASM with Facial Action Unit (FAU) for automatically modeling and extracting the facial emotional features. To recognize the facial expression, we use the Dynamic Bayesian Networks (DBNs) for modeling and understanding the temporal phase of facial emotion state in image sequences.
