A Restoration Method for Images Including Lost Pixels based on A Similarity between A Local Region and A Neighbor Region of A Lost Pixel

Abstract

In the field of image restoration of unnecessary regions, a method based on optical flow has been proposed. This method can accurately restore unnecessary regions in case of changing of edge direction or of plural edges in unnecessary regions. However, when continuity of intensity between a pixel in unnecessary regions and its neighboring pixels is not satisfied, the method can not accurately restore unnecessary regions. In this paper, to solve the problem, we propose a method based on similarity between a local region and a neighboring region of a lost in unnecessary regions.