Extracting printed designs and woven patterns from textile images

Wei Jia, Stephen J. McKenna, Annette A. Ward

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    3 Citations (Scopus)

    Abstract

    The extraction of printed designs and woven patterns from textiles is formulated as a pixel labelling problem. Algorithms based on Markov random field (MRF) optimisation and reestimation are described and evaluated on images from an historical fabric archive. A method for quantitative evaluation is presented and used to compare the performance of MRF models optimised using alpha-expansion and iterated conditional modes, both with and without parameter reestimation. Results are promising for potential application to content-based indexing and browsing.

    Original languageEnglish
    Title of host publicationVISAPP 2009
    Subtitle of host publicationproceedings of the Fourth International Conference on Computer Vision Theory and Applications, Lisboa, Portugal, February 5-8, 2009
    EditorsAlpesh Ranchordas, Helder Araujo
    Place of PublicationSetubal
    PublisherInstitute for Systems and Technologies of Information, Control and Communication
    Pages201-208
    Number of pages8
    Volume1
    ISBN (Print)9789898111692
    Publication statusPublished - 2009
    Event4th International Conference on Computer Vision Theory and Applications - Lisboa, Lisbon, Portugal
    Duration: 5 Feb 20098 Feb 2009
    http://www.visapp.visigrapp.org/VISAPP2009/

    Conference

    Conference4th International Conference on Computer Vision Theory and Applications
    Abbreviated titleVISAPP 2009
    Country/TerritoryPortugal
    CityLisbon
    Period5/02/098/02/09
    Internet address

    Keywords

    • Textile segmentation
    • Pixel labelling
    • Markov random field
    • Quantitative evaluation
    • Energy minimization
    • Segmentation

    Fingerprint

    Dive into the research topics of 'Extracting printed designs and woven patterns from textile images'. Together they form a unique fingerprint.

    Cite this