Feature extraction of face based on the sparse manifold configuration

In the field of recognition, it is a way to improve the rate of recognition by extracting the key feature of the target effectively. In this paper, we proposed an improved method of sparse manifold configuration to solve the problem of feature extraction in face recognition, which is based on manifold learning and the sparsity, and then we used this method to build the configuration and finish the tasks of subspace learning. After a large number of image experiments, we completed the categorization of these images.