Research Of Facial Expression Recognition Based On Gabor Wavelet Transformation

Affective computing is a new research area, which tries to enable machine (computer) 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 on facial expression analysis. Facial expression analysis has wide application in many areas such as emotion and paralinguistic communication, clinical psychology, psychiatry, neurology, pain assessment, lie detection, intelligent environments, and mufti-modal human computer interface (HCI), and which has a well application foreground. The research of it has a great promotion to artificial intelligence.Facial expression recognition is a keystone and a difficulty of affective computing, just because facial expression recognition is cross-subjects, there still a lot of correlative problems need to solve.The general approach to facial expression analysis consists of three steps: face acquisition, facial data extraction and representation, and facial expression recognition. Now, there are many researches in the three aspects, but the problems on facial data extraction and facial expression recognition still haven't been resolved. In this paper have a deep research on this two aspects. At present, at the field of mode-recognition, two-dimensional wavelet transformation has an abroad application. This thesis has a research on the facial expression recognition application of two-dimensional wavelet transformation .as well using the methods of image process achieve the facial expression analysis recognition .First in the image pretreatment aspect, we make all the images have the same position, angle and size by using a lot of methods of image process, the aim is to make preparations to the second stage. Secondly in the facial feature extraction aspect, we made a one by one pixel two-dimensional wavelet transformation to the image, and then expression feature vectors of the expression sub-regions are extracted by wavelet transformation, at last in the facial expression recognition stage, we account the degree of similarity by Euclidean distance. In the research of model-matching algorithm, we proposed a simulation model-matching algorithm, it allow the template have a tiny move. Finally we use the K-nearest neighbor classifier to achieve the 6 kinds basic facial expression recognition.At last we discuss emphases and difficulties of above-all facial expression recognition methods, and put forward the future work development and direction.