Face recognition with single training sample per person based on generalized slide window and weighted 2DLDA
Yongjun Liu · Computer Engineering and Applications Journal · 2008
For face recognition with single training sample per person,the conventional face recognition methods which work with many training samples don’t function well.Especially,a number of methods based on Fisher linear discriminant criterion can’t work because the within-class scatter matrix is a matrix with all elements being zero.To overcome the above problem,we propose a new sample augment method,called generalized slide window,in this paper.In order to effectively maintain and strengthen the within-class and between-class information,we obey the rulebig window,small step to produce a set of window images for each training image.Finally,weighted two-dimensional Fisher linear discriminant analysis is performed on the window images obtained.The experimental results on ORL face database show that the proposed method is effective and promising in face recognition with single training sample per person.