Image Super-Resolution based on Efficient Novel Two-Stage Super-Resolution Technique

Super resolution increases the resolution of the image. This paper proposes the two stage novel SR method.. In the firststage, of the proposed method first jointly train two dictionaries for the high and low resolution image patches. Then apply a sparse representation for each low-resolution image patch, and correspondingly generate a high-resolution intermediate image by exploiting the high-resolution dictionary and low-resolution dictionary. In the second stage,of the proposed method a higher resolution image is obtained by fusing the intermediate high-resolution image sequence based on projection onto convex sets (POCS) method, increase image magnification while keeping goodeffectiveness. Experiment results show theeffectiveness of the proposed method and improved performance over other SR algorithms