计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (19): 164-165.DOI: 10.3778/j.issn.1002-8331.2010.19.048

• 图形、图像、模式识别 • 上一篇    下一篇

一范数在图像修复中的应用

贺文熙1,2,丁兴号2   

  1. 1.江西理工大学理学院,江西赣州341000
    2.厦门大学信息科学与技术学院,福建厦门361000
  • 收稿日期:2008-12-23 修回日期:2009-03-12 出版日期:2010-07-01 发布日期:2010-07-01
  • 通讯作者: 贺文熙

Image inpainting based on 1 Norm

HE Wen-xi1,2,DING Xing-hao2   

  1. 1.College of Sciences,Jiangxi Technology University,Ganzhou,Jiangxi 341000,China
    2.School of Information Science and Technology,Xiamen Universty,Xiamen,Fujian 361000,China
  • Received:2008-12-23 Revised:2009-03-12 Online:2010-07-01 Published:2010-07-01
  • Contact: HE Wen-xi

摘要: 考虑到图像相邻像素具有相关性,修复后的图像能量应为最低状态。定义了复合向量的范数。利用改进的热传导模型进行图像修复。提出的算法既可用来有效地修复划痕,也可用于去除文字,还可以填充较大破损区域。

Abstract: By studying correlated relation of neighborhood pixel,the energy of digital image that is inpainted is the lowest. This paper defines the norm of complex vector.The digital image may be inpainted by the algorithm based on a renovated
heat conduction model.This paper proposes an efficient inpainting algorithm for inpainting scratches and removing words and
filling the large missing portions.

中图分类号: