Manifold learning method can discover intrinsic low-dimensional submanifold embedded in the high-dimensional image space
it attracts more and more attention in the research area of biometrics and cognitive science. For the problem exits in locally linear embedding (LLE) algorithm
a self-organized LLE algorithm is proposed in this paper. This new method can not only determine the nearest neighbors of LLE automatically
save lots of operation
but also gain a perfect approximation of face manifold. This paper proposed theoretical analysis for details and applied to some dataset to verify the effeteness of this method. Experimental results on public face databases show that the proposed method can improves face classification performance effectively.