Abstract
Biometric identification has been rapidly becoming a critical research category of image processing techniques. Several kinds of features have been proposed for recognizing people which are based on physiological or behavioral characteristics such as face, fingerprint, handwriting, iris and etc. Face recognition system is a computer-based system that is able to automatically recognize people by face images. For face recognition systems, the most import and difficult task is to find out a method for extracting the most separable features. In this thesis we propose a new feature extraction method based on singular value decomposition (SVD) and principal component analysis (PCA) for classifying facial images. Furthermore, it is compared with the other four famous feature extraction methods (Eigenfaces, Fisherfaces, 2DPCA and 2DLDA). From the experimental results, our approach obtains a good recognition rate and the memory cost is much smaller.