Edge Model Based Super Resolution Video Generation

As applications involving the capture of digital images and video become more demanding in the recent years and at the same time more ambitious for critical real time applications like video surveillance. There is a driving need for digital images and videos of higher resolutions and visual quality. Since, the video is a sequence of still images representing scenes in motion, the problem of generation of high resolution video is narrowed down to the problem of high resolution image generation from low resolution video frame. High resolution image can be obtained using the conventional interpolation methods. Conventional linear interpolation schemes fail to capture the fast evolving statistics around edges and consequently produce interpolated images with blurred edges and annoying artifacts. Moreover, the relationship between highresolution pixels and low-resolution pixels plays the critical role in the performance of a learning based interpolation algorithm. Since edges are presumably very important features in natural images, retaining the geometric regularity of edges becomes the driving need to develop new approaches for image processing tasks. To generate sharp edges with no artifacts in the high resolution video is the main objective of this paper.