Texture enhancement for improving single-image super-resolution performance

Abstract

Although various single-image super-resolution algorithms have been developed to increase image resolution, they still do not provide adequate performance in the texture region due to the lack of fine textures in the processed image. In this paper, we present a novel texture enhancement strategy in order to improve the super-resolution performance in the texture region. For texture enhancement, we extract a low-resolution texture layer from an input image and generate a high-resolution texture layer by applying the proposed texture synthesis algorithm. A texture enhanced high-resolution image is then obtained by properly combining the generated high-resolution texture layer with an image obtained by using an existing single-image super-resolution algorithm. Experimental results show that the proposed texture enhancement strategy provides sharper and more natural looking textures compared with the existing super-resolution algorithms. (C) 2016 Elsevier B.V. All rights reserved.