A proximal method for the K-SVD dictionary learning

In this paper, we propose a dictionary updating method and show numerically that it can converge to a dictionary that outperforms the dictionary derived by the K-SVD method. The proposed method is based on the proximal point approach used in the convex optimization algorithm. We incorporate the approach into the well-known MOD and combine the result with the K-SVD method to obtain the proposed method. We analyze the complexity of the proposed method and compare it with that of the K-SVD method. The results of experiments demonstrate that our method outperforms K-SVD with only a slight increase in the execution time.