Laplacianfaces incorporated inside nonnegative matrix factorization for face recognition

In this paper, we propose a face recognition method called the Laplacian Nonnegative Matrix Factorization. By incorporating Laplacianfaces inside the Nonnegative Matrix Factorization (NMF) decomposition, the goal is to extend the NMF algorithm in order to extract discriminant information by preserving locality information in face subspac. With Laplacian NMF decomposition, it is expected to own Laplacianfaces characteristic in the face subspace. Thus Laplacian NMF have more discriminant power than NMF. The proposed method has been applied to face recognition on Yale database. Experimental results show that our proposed method achieves better face recognition performance than Eigenfaces, Fisherfaces, Laplacianfaces and NMF.