Sparse Representation Based Discriminant Locality Preserving Dictionary Learning via Volterrafaces for Face Recognition

In this paper, we present a new face classification system that combines Volterrafaces, discriminant locality preserving criterion, and dictionary learning. This method first divides the original image into sub-blocks of the same size, and then the discriminant analysis model based on the twodimensional Volterra kernel is used to find the nonlinear function mapping for each sub-block. Next, we use the proposed sparse representation based discriminant locality preserving dictionary learning method to individually optimize and classify the subfeatures of each projection axis. Finally, the final classification result of the image is determined by a majority vote.