Abstract: The main aim of Single Image Superresolution (SR) is to build a high resolution image from a sequences of single low resolution (LR) images. The SR is a classic and active image processing problem which aims to generate a high resolution (HR) image from low resolution input image. The SR is challenging because of the missing details in the given LR image. Thus it is very difficult to explore effective prior knowledge for boosting the reconstruction performance. To solve this problem it is necessary to have the prior knowledge of image to make the problem solvable and to improve the quality of generated image. The main goal of SR is to generate the HR image with good visual perception as similar as original image. These algorithms will evaluate the subjective visual effect, objective quality computational time and PSNR, etc. These algorithms will improve quality of image like better visual effect quality, lower reconstruction error and acceptable computation efficiency, etc.

Keywords: Super-resolution (SR), Visual Effect, PSNR, Computational Efficiency.