Abstract: We analyzed the mathematic essence of sparse representation, sparse regularized signal decomposition. Studied a sparse representation algorithm of orthogonal matching pursuit Using the matrix Cholesky decomposition,we realined the OMP algorithm a fast version. We cast the recognition problem as one of classifying among multiple linear regression models and developed a new framework from sparse signal representation. We viewed a test sample as the linear combination of training samples. We conducted experiments on face recognition to verify the efficacy of the proposed algorithm.

Key words: Sparse representation, Sparse coding, Compressed sensing, Orthogonal matching pursuit, Feature extraction