Face Recognition Method Research And Implementation

With the development of our social and network technology, the authentication's methods based on biological characters have been searched extensively, such as fingerprint identification, iris recognition and face recognition. In all of the biological methods, the face recognition have been the researchers'favors because of its features include directly, friendly and conveniently.The human face recognition system is a kind of pattern recognition system based on information processing. It can be divided into two parts: feature extraction and pattern classification. The first part is to find out a set of features that can represent the images from different persons; the second part classifies the features got from the first part. The performance of the system depends on both of the parts. In this paper, we studied based on the two parts, and main tasks are as follows:1. Through extensive investigation and research, provided a thorough survey of the AFR history and the state-of-the-art; and mainly researches on some key questions about the automatic face recognition system, including studying the key technologies that has existed, analyzing and summarizing the problems and difficults in the present studies; in the last, we survey the main public face databases at home and abroad.2. Facial image preprocessing and face detection method were studied. In order to improve the accuracy of the face recognition, the face image preprocessing work is necessary before face recognition. Face image preprocessing work includes a lot of contents, such as filtering, gray-scale conversion, gray-scale normalization, geometry normalized and so on. Good pre-treatment is very important before a series of face recognition methods and work. In this paper, pre-treatment includes: geometric normalization and illumination processing. We also introduce some basic face detection methods, at last, we use SMQT and SnoW unified algorithm to experiment, the results show that this model has a strong effectiveness for face detection3. Face feature extraction and recognition. Here we simply introduce some basic methods of feature extraction, and further analyze the principle of PCA, LDA, NMF recognition algorithms, and verify results of the algorithms. A large number of experiments show that the method is effective and practicable.4. Facial feature alignment was studied. Accurate facial feature alignment is the prerequisite of a face recognition system. Currently, the Active Shape Model(ASM) is one of the main model for this problem. The Gabor wavelet has good characteristics of the spatial location and orientation selection. Based on these observations, a Gabor ASM feature extraction algorithm is introduced, which attempts to fuse the Gabor phase and magnitude information. The proposed algorithm blend the merits of Gabor wavelet-based and ASM algorithm. Encouragingly, experiment results have illustrated the better performance and illumination robustness of the proposed model.