Novel back-projection framework for single image super-resolution

A novel back-projection framework for single image super-resolution is proposed in this paper. In our framework, the edge of the high-resolution image is firstly detected by the canny edge detection algorithm and the bilateral filtering algorithm, and the high-resolution image is divided into edge area and flat area. Then the edge area of the high resolution image is filtered by the adaptive kernel regression denoising algorithm in the iterative back-projection process. And the flat area of the high resolution image is reconstructed by the iterative back-projection method. Experimental results demonstrate that the proposed method can reconstruct high quality images in both the measure of objective quality and the subjective perception.