Minimum spanning tree filters for edge-preserving impulse noise removal

Kohei Inoue, Kenji Hara, Kiichi Urahama

    Research output: Contribution to journalArticlepeer-review

    1 Citation (Scopus)


    We propose edge-preserving smoothing (EPS) filters based on minimum spanning trees (MSTs) on image planes. The shape of the filter window is adaptively determined for each pixel by the MST, which avoids stepping over edges in images. Experimental results showed that the proposed MST filters can remove impulse noise better than conventional EPS filters.

    Original languageEnglish
    Pages (from-to)J287-J289
    JournalKyokai Joho Imeji Zasshi/Journal of the Institute of Image Information and Television Engineers
    Issue number8
    Publication statusPublished - 2012

    All Science Journal Classification (ASJC) codes

    • Media Technology
    • Computer Science Applications
    • Electrical and Electronic Engineering


    Dive into the research topics of 'Minimum spanning tree filters for edge-preserving impulse noise removal'. Together they form a unique fingerprint.

    Cite this