Research On Facial Feature Extraction Algorithm Based On Gabor Filter

Face recognition is a rich theoretical basis and the value of study. A typical face recognition system includes:face detection, face feature extraction and face recognition. At the same time, extracting the face feature includes pretreatment and feature extraction. Pretreatment and feature extraction occupies a very important role in the automatic face recognition system. Two-dimensional Gabor wavelets transform coefficients describe a small patch of grey values in an image around every given position, feature extraction and classification which are based on the two-dimensional Gabor wavelets transform coefficients of face image are called Gabor wavelets transform based face recognition. This dissertation researches into the theory and technology of face recognition through two-dimensional Gabor wavelets transform.In the image preprocessing, the traditional color elliptical model can detect the face in the color image, but it can not effectively detect the approximate location of the face organs. The paper describes a new thinking to realize the detection of face area and organs, which adopts subsection of color space, and the definition of color space elliptical model in each section of it. In aspect of feature extraction, classical method of Gabor texture extraction has strong robust in uneven illumination, position changes and changes in facial expressions, but it have to make Gabor wavelet transform in the whole image, which needs a complex process es, long time-consuming, and can not meet the requirements of real-time. The image processing mainly has three steps. First, it uses the Gabor filters to detect the eye area. Secondly, by generalized hough transform, it realizes the connection of image edge breakpoint. Finally, with the help of Gabor wavelet transform two-dimensional for the eye edge, it extracts the feature of the eye edge to displace the classical feature extraction in Gabor texture features, which is better for dimension reduction in feature space and reducing the image processing area and the complexity of computing. It reduces the dimension of face texture feature from 137 to 19, which shorten the time of face recognition efficiently.