计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (29): 170-172.DOI: 10.3778/j.issn.1002-8331.2010.29.049

• 图形、图像、模式识别 • 上一篇    下一篇

一种具有边界保持特性的图像修补方法

王 瑞1,刘红喜1,2,郝德芳1,孙俊喜1,杨华民1   

  1. 1.长春理工大学 电子信息工程学院,长春 130022
    2.长春工程学院 电气与信息工程学院,长春 130012
  • 收稿日期:2009-04-08 修回日期:2009-06-16 出版日期:2010-10-11 发布日期:2010-10-11
  • 通讯作者: 王 瑞

Image inpainting method with boundary keeping

WANG Rui1,LIU Hong-xi1,2,HAO De-fang1,SUN Jun-xi1,YANG Hua-min1   

  1. 1.School of Electronic & Information Engineering,Changchun University of Science and Technology,Changchun 130022,China
    2.College of Electronic & Information,Changchun Institute of Technology,Changchun 130012,China
  • Received:2009-04-08 Revised:2009-06-16 Online:2010-10-11 Published:2010-10-11
  • Contact: WANG Rui

摘要: 将边界优先级的图像修补方法引入GMRF修补模型中,通过对待修补区域边界像素点优先级的计算,确定像素的修补顺序,并和传统GMRF修补算法的修补结果做出比较。实验结果表明,基于边界优先级的图像修补算法对待修补区域包含边缘的图像有较好的修补效果,能够保持图像缺损区域的结构信息。

Abstract: An image inpainting method with boundary keeping is introduced based on GMRF model.The priority of pixels in the edge of unknown region is pre-selected so that the inpainting process is completed in the order of priority of pixel.Some experimental results are given and compared with the classical algorithm according to the PSNR rule,which show that the proposed algorithm can inpaint the unknown region that contains some edges and maintain the structural information to a certain extent.

中图分类号: