A Structure-Synthesis Image Inpainting Algorithm Based on Morphological Erosion Operation

Abstract

Image inpainting provides a means to restore damaged region of an image, such that the image looks complete and natural after the inpainting process. This paper presents a new inpainting algorithm, which implements the filling of damaged region based on morphological erosion and structure feature replication. The method, which is based on a structure/texture feature matching algorithm, can retain isophote continuity by propagating both texture and structure characteristics information of the known patch into the damaged region, and output a complete, natural-looking and non-blurred image. By implementing region-filling through morphological erosion, several pixels instead of one can be restored at every inpainting step, making the method faster than many traditional texture synthesis inpainting algorithms. We demonstrate through several examples the effectiveness of the algorithm in removing large occluding objects as well as thin scratches.