Performance Improvement for Blurred Images Utilizing Learning-Based Image Restoration Method

Abstract

Blind image restoration, which restores a clear image from a single blurry image, is a difficult process of estimating two unknowns: a point-spread function (PSF) and an ideal image. In this paper, we use a learning-type restoration method that learns a sharp image without blur and an image that contains blur, and supposes the restoration of blurred images that occur especially when shooting fast-moving objects such as cars. We study the improvement of the image quality of the restored image by examining the training image used for the image.