New Algorithm of Image Inpainting Curvature-Driven Diffusion Model
CSTR:
Author:
Affiliation:

Clc Number:

Fund Project:

  • Article
  • |
  • Figures
  • |
  • Metrics
  • |
  • Reference
  • |
  • Related
  • |
  • Cited by
  • |
  • Materials
  • |
  • Comments
    Abstract:

    CDD inpainting model is based on TV model local spread and it introduces curvature term, then realizes the connectivity and holistic principle. For the different size of the gradient value of each reference point and diffusion efficiency defects, an improved CDD inpainting algorithm is proposed by degenerate diffusion model. Firstly, it calculates the gradient information of pixel and curvature of isolux line to control the contribution of the reference point on the repair area, and then they are weighted. Finally, the discrete model of partial differential equations is given. Experimental results show that this algorithm realizes the connectivity principle partially. In addition, it can repair the damaged image better than CDD algorithm with same iterative times.

    Reference
    Related
    Cited by
Get Citation

王军锋,裴艳侠,王涛.图像修复的CDD模型新算法.计算机系统应用,2016,25(8):135-138

Copy
Share
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
History
  • Received:December 07,2015
  • Revised:January 14,2016
  • Adopted:
  • Online: August 16,2016
  • Published:
Article QR Code
You are the firstVisitors
Copyright: Institute of Software, Chinese Academy of Sciences Beijing ICP No. 05046678-3
Address:4# South Fourth Street, Zhongguancun,Haidian, Beijing,Postal Code:100190
Phone:010-62661041 Fax: Email:csa (a) iscas.ac.cn
Technical Support:Beijing Qinyun Technology Development Co., Ltd.

Beijing Public Network Security No. 11040202500063