Image super-resolution reconstruction based on fusion of K-SVD and semi-coupled dictionary learning

Abstract

In order to solve the problem of time-consuming dictionary training process, we propose an image super-resolution reconstruction approach based on fusion of K-SVD algorithm and semi-coupled dictionary learning framework. In this paper, we use K-SVD algorithm to training the dictionary pair in the semi-coupled dictionary learning model. In comparison with the existing methods, experiment results on image reconstruction show that the proposed algorithm reduce dictionary training time nearly four times, while improve the quality of reconstructed images.