Home>Results

  • Complex
  • Title
  • Keyword
  • Abstract
  • Scholars
  • Journal
  • ISSN
  • Conference
成果搜索

[期刊论文]

Edge fitting based image inpainting

Share
Edit Delete 报错

author:

Guo, Shimin (Guo, Shimin.) [1] | Wang, Meiqing (Wang, Meiqing.) [2] (Scholars:王美清) | Huang, Chensi (Huang, Chensi.) [3]

Indexed by:

EI Scopus

Abstract:

Image inpainting based on Partial Differential Equation (PDE) has been the state-of-art technology nowadays. BSCB model is the first digital image inpainting model and can get good inpainting result when the target region is small like scratches. But it is not quite suitable for large target regions. In this paper, to improve the inpainting result of the damaged image with big missing area, the edge fitting based BSCB (EFBSCB) image inpainting method is proposed. Firstly some pairs of disconnected edges broken by missing areas are matched and fitted as connected curves which partition the missing areas into small areas, then the intensities of the pixel points in the curves are approximated; finally, BSCB model is used to smoothly propagate the surrounding information into these small missing areas. Experimental results show that this model can get better restoration than the original BSCB model when used for large missing areas. © 2010 Binary Information Press.

Keyword:

Edge detection Image processing Partial differential equations

Community:

  • [ 1 ] [Guo, Shimin]College of Mathematics and Computer Science, Fuzhou University, Fujian 350108, China
  • [ 2 ] [Wang, Meiqing]College of Mathematics and Computer Science, Fuzhou University, Fujian 350108, China
  • [ 3 ] [Huang, Chensi]College of Mathematics and Computer Science, Fuzhou University, Fujian 350108, China

Reprint 's Address:

Show more details

Version:

Related Article:

Source :

Journal of Information and Computational Science

ISSN: 1548-7741

Year: 2010

Issue: 11

Volume: 7

Page: 2296-2301

Cited Count:

WoS CC Cited Count:

30 Days PV: 1

Online/Total:836/10208839
Address:FZU Library(No.2 Xuyuan Road, Fuzhou, Fujian, PRC Post Code:350116) Contact Us:0591-22865326
Copyright:FZU Library Technical Support:Beijing Aegean Software Co., Ltd. 闽ICP备05005463号-1