Image inpainting based on image segmentation

Abstract

Removing a target object and Ailing the missing regions of an image is the key technology generally applied to image restoration. In this paper, we propose a novel algorithm for exemplar-based image inpainting. By using segmentation-based searching strategy and redefined priority function, this algorithm can solve the problems that would occur in large area's filling. Experimental results show that this improved method can efficiently handle different cases, especially the large regions in complex background.