Overview on image super resolution reconstruction

Abstract

Image super resolution (SR) reconstruction technique is receiving increasing attention from the image processing community, and it has been widely used in many applications such as remote sensing image, medical image, video surveillance and high definition television. The essential of image SR reconstruction technique is how to produce a clearly high resolution (HR) image from the information of one or several low resolution (LR) images. Firstly, the fundamental idea of representative methods, the history and state of art of super resolution reconstruction are stated briefly according to the classification between the reconstruction based method and the learning based method. Secondly, advantages and defects of each methods are analyzed and summarized systematically, as well as the limitation exist in general ability. Finally, the further research directions of image super resolution reconstruction technique are proposed.