The Research Of Facial Features Extraction Based On Active Shape Model

Automatic Face Recognition (AFR) aims at endowing computers with the ability to identify different human beings according to face images. Such a research has both significant theoretic values and wide potential applications. After more than 30 years' development, AFR has made great progress especially in the past ten years. The state-of-the-art AFR system can perform identification successfully under well-controlled environment. However, evaluation results and practical experience have shown that AFR technologies are currently far from mature. A great number of challenges are to be solved before one can implement a robust practical AFR application, especially the accurate facial feature location problem, which is the prerequisite for sequent feature exaction and classification. Because the geometry shape information method has the high accuracy rate and the big robustness among main techniques of human facial features localization, this paper studies geometry information based algorithms of active shape model (ASM) and its application in human facial features localization.As the earlier period of the recognition, face detection is very important. This paper first describes some main current methods of face detection. Images preprocessing is an essential foundation for the whole system. Face recognition would not go on wheels without a reliable preprocessing. Several methods of images preprocessing are given in the paper. Face alignment should be carried out after face detection, and this paper discusses the face alignment algorithm using Active Shape Model as emphasis. Point Distribution Model is described firstly, and three aspects of ASM are discussed: aligning the training set, modeling shape variation and local grey model.Improvements are made to the standard ASM for facial feature detection. Local area constraint method and edge constraint method are proposed. Also, a points labeling strategy which increase the relativity of the labeled points is put forward, and a semi-automatic feature points labeling tool is designed which greatly improve both the accuracy and the efficiency of labeling work. Experiment shows the improved active shape model is successful in facial feature extraction and outperforms significantly better than standard active shape model in the localization of facial features.