Artifact reduction of compressed images and video combining adaptive fuzzy filtering and directional anisotropic diffusion

  • Ehsan Nadernejad (Invited author)
  • , Søren Forchhammer (Invited author)
  • , Jari Korhonen (Invited author)

    Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

    1160 Downloads (Orbit)

    Abstract

    Fuzzy filtering is one of the recently developed methods for reducing distortion in compressed images and video. In this paper, we combine the powerful anisotropic diffusion equations with fuzzy filtering in order to reduce the impact of artifacts. Based on the directional nature of the blocking and ringing artifacts, we have applied directional anisotropic diffusion. Besides that, the selection of the adaptive threshold parameter for the diffusion coefficient has also improved the performance of the algorithm. Experimental results on JPEG compressed images as well as MJPEG and H.264 compressed videos show improvement in artifact reduction of the proposed algorithm over other directional and spatial fuzzy filters.
    Original languageEnglish
    Title of host publication2011 3rd European Workshop on Visual Information Processing (EUVIP)
    PublisherIEEE
    Publication date2011
    Pages24-29
    ISBN (Print)978-1-4577-0070-5
    DOIs
    Publication statusPublished - 2011
    Event3rd European Workshop on Visual Information Processing - Paris, France
    Duration: 4 Jul 20116 Jul 2011
    Conference number: 3
    https://www.eurasip.org/newsletter/newsletter-2011-22-1-html/newsletter-2011-22-1/The%203rd%20European%20Workshop%20on%20Visual%20Information%20Processing.pdf

    Conference

    Conference3rd European Workshop on Visual Information Processing
    Number3
    Country/TerritoryFrance
    CityParis
    Period04/07/201106/07/2011
    Internet address

    Keywords

    • H.264
    • Artifact Reduction
    • Anisotropic Diffusion
    • Fuzzy filter

    Fingerprint

    Dive into the research topics of 'Artifact reduction of compressed images and video combining adaptive fuzzy filtering and directional anisotropic diffusion'. Together they form a unique fingerprint.

    Cite this