Super-resolution restoration of MMW image using sparse representation based on couple dictionaries

Abstract

This paper addresses the problem of the super-resolution restoration of a single millimeter wave image using sparse representation based on couple dictionary training. Utilizing the coefficients of the sparse representation of each low-resolution image patch, the high-resolution image patches can be generated, further, the low resolution image can be reconstructed well. The quality of a restoration MMW image was measured by the relative single noise ratio (RSNR. Compared with image restoration results of bicubic, experimental results prove that the sparse representation is effective in the super-resolution restoration of MMW. © 2012 Springer-Verlag.