Image segmentation based on a multiresolution Bayesian framework

Chang Tsun Li, Roland Wilson

    Research output: Book chapter/Published conference paperConference paperpeer-review

    5 Citations (Scopus)

    Abstract

    A Multiresolution Markov Random Field (MMRF) approach to texture segmentation which makes use of both region and boundary information is proposed in this work. The algorithm is designed to improve the efficiency of that reported in [1] and solve two problems: over-segmentation and under-segmentation, which the algorithms developed in [2] and [3] are not able to cope with. At each resolution level, regional information and boundary information between textured regions is extracted to describe each block using the Multiresolution Fourier Transform (MFT). The image is then modelled as a sequence of Markov random fields (MRF's) of varying resolution and a Gibbs sampling scheme is applied to label the constituent sites at each scale by seeking the configuration with minimal interaction energy. The interaction energy is defined as a function of the regional and boundary information. The constraints of region context, and boundary smoothness and connectivity are also encoded in the interaction energy. Once the algorithm converges at a given scale, the segmentation result is propagated down to the next resolution for further refinement.

    Original languageEnglish
    Title of host publicationIEEE International Conference on Image Processing
    PublisherIEEE Comp Soc
    Pages761-765
    Number of pages5
    Volume3
    Publication statusPublished - 1998
    EventProceedings of the 1998 International Conference on Image Processing, ICIP. Part 2 (of 3) - Chicago, IL, USA
    Duration: 04 Oct 199807 Oct 1998

    Conference

    ConferenceProceedings of the 1998 International Conference on Image Processing, ICIP. Part 2 (of 3)
    CityChicago, IL, USA
    Period04/10/9807/10/98

    Fingerprint

    Dive into the research topics of 'Image segmentation based on a multiresolution Bayesian framework'. Together they form a unique fingerprint.

    Cite this