点击次数：

论文类型：期刊论文

发表时间：2015-07-20

发表刊物：Journal of Information and Computational Science

收录刊物：EI、Scopus

卷号：12

期号：11

页面范围：4349-4356

ISSN号：15487741

摘要：Image inpainting is the process of filling in missing parts of damaged images based on information gleaned from surrounding areas. It is a difficult problem of solving the length term with Mumford-Shah model. Therefore, an image inpainting method is proposed based on a convex variant of the Mumford-Shah model and Split-Bregman algorithm, which increases the diffusion ability of the model in the texture and the smooth regions of an image area. In addition to fill the whole area of information loss, the algorithm can also removal defect on the external information noise area. Experimental results showed that the proposed method is a more Effective repair than the traditional classical method for the image text removal, scratch repair, etc. ?, 2015, Binary Information Press. All right reserved.