Face recognition based on Laplacian Eigenmaps

This paper put forward an image recognition method based on Laplacian Eigenmaps and Minimum Neighbor, the method makes use of incremental Laplacian Eigenmap reducing the dimension and extracting the feature to data points. In the dimension reduction process, it is maintained local information optimization in some extent. In the aspect of operation efficiency, it avoids the adjacent map to re-build in the entire dataset after the new point arriving, so greatly reducing the computational complexity. Simulations show that the method of image recognition rate is better than the PCA, LPP and other methods.