The Method of Human Facial Expression Recognition Based on Wavelet Transformation Reducing the Dimension and Improved Fisher Discrimination

The segmentation of skin color region is carried on through the mathematics morphology processing, using Hough transform to locate the eyes, and then the expression image is proceeded by geometry standardization. After using wavelet transform to reduce the dimension of images, the feature extraction of facial expression can be realized with the discrimination of improved Fishier. It can solve actual problems on dispersion matrix singular within the class. Finally, the method of minimum Mahalanobis distance classifier is carried on facial expression recognition. It is proved in the CMU facial expression database that the method can reduce computation and enhance the recognition rate of facial expression recognition.