Skip to main navigation Skip to search Skip to main content

Image restoration involving connectedness

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

Abstract

Two local adaptative nonlinear filters for image denoising are described and compared to other methods. The first one is a median filter computed over a connected neighborhood that fits the local conformation of level lines. The second one deals with the size of grains in image. Both filters depend on a single parameter, the area A, are morphological and associated with smooth as well as unsmooth fixed points. They make the total variation decrease and are able more than other classic morphological filters to denoise while preserving image structures. They are particularly designed for automation.

Original languageEnglish
Title of host publicationProceedings of the Sixth International Workshop on Digital Image Processing and Computer Graphics
EditorsEmanuel WENGER, Leonid I. DIMITROV
PublisherSPIE
Pages84-95
Number of pages12
Volume3346
DOIs
Publication statusPublished - 1 Feb 1998
Externally publishedYes
Event6th International Workshop on Digital Image Processing and Computer Graphics - Vienna, Austria
Duration: 20 Oct 199722 Oct 1997

Workshop

Workshop6th International Workshop on Digital Image Processing and Computer Graphics
Abbreviated titleDIP-97
Country/TerritoryAustria
CityVienna
Period20/10/9722/10/97

Bibliographical note

Acknowledgments:
The first author thanks Leonid Yaroslavsky for useful discussions.

Keywords

  • Additive or impulse noise
  • Automation
  • Blind restoration
  • Connectedness
  • Local adaptative morphological filtering
  • Smooth and unsmooth fixed point

Fingerprint

Dive into the research topics of 'Image restoration involving connectedness'. Together they form a unique fingerprint.

Cite this