Gabor Wavelet-based Face Expression Recognition

Human facial expression recognition technology is one of the most challenging subjects in the fields of machine vision and pattern recognition. It is a comprehensive topic which involves in the pattern recognition, the computer vision, image processing, artificial intelligent, affective computing, and so on. At present, facial expression recognition is a very active research. Computer facial expression recognition is an entry to analyze and comprehend humans' emotion, it extracts and analyzes facial expression information by humans' knowledge and thought, and then get the results of facial expression recognition by learning and inferring. This paper focuses on the methods based on Gabor Wavelets Transform. At last, the computer can recognize seven basic facial expression automatically : neutrality ,angry, disgust, fear, happy, sadness , surprise .In this paper, firstly, the research background, the origin and the development of facial expression recognition are introduced. The common methods for extracting feature, classifier and recognition of facial expression are given in details. Secondly, we discuss the methods for feature extraction based on Two-dimensional Gabor transform. Two-dimensional Gabor wavelets transform is realized by computing the convolutions of a bank of two-dimensional Gabor filters and the grey values of pixels in a given area in an image. Two-dimensional Gabor wavelets seem to be a good approximation to the receptive fields of the simple cells in the mammalians' visual cortex. In this paper, it is validated by selecting the parameters of Gabor filters. Recognition based on two-dimensional Gabor wavelets transform surpasses the one based on the grey values of the original image directly. Thirdly, elastic model matching algorithm is improved. The elastic model with 26 key points of facial expression is applied for matching. The result of classier and recognition is analyzed by K-Nearest Neighbor. Compared with the conventional method, this model has less computation and higher recognition accuracy. Finally, the system for facial expression recognition is designed with Visual C++ 6.0 Language .The system woks well with the function of static expression image preprocessing, feature extraction and expression classifier and recognition.