Research On Facial Expression Recognition And Discussion On Applying Into Game

Affective computing is a new research area, which tries to enable machine (computer) to have the ability of understanding and expressing affection, just like human beings. Affective computing plays an important role in intelligent human computer interface (HCI). Since human affection is expressed mainly by facial expression, researchers begin to pay more and more attention to facial expression analysis. Facial expression recognition is to analyze and detect the special expression state from given expression images or video frames and then to ascertain the subject's specific inborn emotion, achieving smarter and more natural interaction between human beings and computers. The whole system can be divided into five parts: expression image capture, pre-processing (face feature detecting and locating), region of interest selection, expression feature abstraction, and face expression recognition.Learned many interrelated literatures and research papers concerning facial expression recognition at home and abroad in recent years, some problems about feature extraction in facial expression recognition are analyzed. We do some researches deeply and discuss how to use facial expression recognition into game. The research work in this paper mainly includes the following several respects:(1) The significance and application future of our paper is generalized. The current domestic and overseas study about affective computing, facial expression recognition and game study are reviewed and some existing facial expression recognition methods are summarized.(2) The methods of facial expression image preprocessing and the theory of feature region segmentation are discussed. Firstly, the expression images are converted from color images to gray images (This step can be passed over because our training data comes from the Japanese JAFFE expression database and they are also gray images.) when at image preprocessing. Then the expression images are processing manually step by rotating adjustment and size normalization. Then the method of average face template matching is applied to face detection and location. Finally, the expression image histogram is equalized. After equalization the details of the image get clearer, and the distribution of gray levels of histogram gets evener.(3) A combined method of facial expression recognition is proposed based on Gray Level Co-occurrence Matrix (GLCM) and Chaos in Genetic Algorithms (CGA). Chaos in Genetic Algorithms is obtained by using randomness, ergodicity and regularity of chaos in order to solve the asymmetric of individual distributions in solution domain. Through the feature extraction by Gray Level Co-occurrence Matrix and Chaos in Genetic Algorithm, an approach is proposed to solve the two tasks, searching region of interest selection (ROI) and feature extraction, simultaneously using a single evolving process. At last, Support Vector Machine (SVM) is applied to image classification.(4)The face expression recognition is applied to game firstly. Some player's expression images are captured by PC Camera. According to the face expression recognition system, the expression images are divided into several classification. Then combining theories from psychology, the player's individuation log can be built and the player's state of psychology can be expressed by game. So the helpful reference for human-computer can be provided.