Research On Face Recognition Algorithm Based On Kernel Methods

Human face is our primary focus of attention in social intercourse, playing a major role in conveying identity and emotion. On the other hand, due to the face recognition technology has a broad range of potential plications in video surveillance, access control, personal card identification, multimedia database retrieval and security, etc; face recognition is one of the most active research subjects in the area of pattern recognition and computer vision in the past 20 years. Face representation (or facial feature extraction) and classifier are two key problems in face recognition. We applied the kernel methods—KCCA and SVDD, which proposed in pattern recognition recently to deal with above two problems.This paper mainly study face extraction and class method, which concept can be summarized as follows:1. The latest feature extraction method based on KCCA is introduced, and the framework of KCCA plus SVDD-based classifier used in face recognition is also proposed.2. In order to enhance the extraction efficiency, this paper proposes to utilize a feature vector selection (FVS) scheme based on geometrical consideration. The algorithm can select a subset of samples whose mappings in feature space are sufficient to represent all of the data in feature space as a linear combination of them. Hence, this will largely reduce the computational complexity of KCCA. Moreover, an improved algorithm inspired by principal component analysis (PCA) is developed. The algorithm can select the most contributive eigenvectors for training and classification instead of considering all the ones.3. A face recognition method based on Support Vector Data Description (SVDD) is presented. According to the results of SVDD and nearest neighbor classifier (NNC) at the different face database, we propose a face recognition method based on the combination of SVDD and NNC. The experiments results show the proposed methods achieved lower error rate than SVDD or NNC.