作者: Christine Guillemot , Gagan Rath
DOI:
关键词:
摘要: This paper introduces the concept of a complementary matching pursuit for sparse approximation. The algorithm is analogous to classical but done in row-space dictionary matrix. A deeper analysis shows that residual error at any iteration may not be orthogonal immediately selected atom, however, this brings about possibility increasing convergence speed and improving sparsity solution vector. validated through simulations with random created using K-SVD algorithm.