A new LDA-based method for face recognition

Linear discriminant analysis (LDA) is a feature extraction technique for classification. In this paper, we propose a new LDA-based method that can overcome the drawback existed in the traditional LDA methods. It redefines the between-class scatter by adding a weight function according to the between-class distance, which helps to separate the classes as much as possible. At the same time, it projects the between-class scatter into the null space of the within-class scatter that contains the most discriminant information. Hence, the transformation matrix composed with the eigenvectors corresponding to the largest eigenvalues of the transferred between-class scatter can maximize the Fisher criteria. Experimental results show our method achieves better performance in comparison with the traditional LDA methods.