A fast approach for single image super resolution via dictionary learning

Abstract

In this paper, a fast method for single image super-resolution using dictionary learning is proposed. In this method, a local high resolution (HR) dictionary is constructed for every patch in the input image. To do this, the information from neighboring patches of the corresponding patch is used. Also, a low resolution (LR) dictionary consists of features obtained from patches of LR images in the corresponding place is obtained. Then, by learning relationship between features of a low resolution patch and LR dictionary, we construct high resolution patch using HR dictionary. The proposed local dictionary patch reconstruction is performed with small error. Also, high processing speed is reachable because of simplification in dictionary construction and patch extraction stages. The experimental results indicate the proposed method outperforms the existing methods in terms of PSNR and runtime.