Content-based image retrieval by scale-space object boundary shape representation

Mark E. Hoffman, Edward Wong

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

    Abstract

    Shape is a popular feature used for content-based image retrieval. In this paper we propose a new method for image retrieval using a shape boundary represented in scale-space. The proposed method is suggested by the notion of `dynamic shape` where all 2-D boundary representations evolve from a single, primeval, featureless shape - a circle. Shape is represented by linearizing the boundary based on the polar coordinates of boundary points relative to the object's centroid. Points on the shape boundary are mapped to a primeval circle, and two functions are defined, the Radius Difference Function and the Angle Difference Function, and smoothed through scale-space to devolve the shape. Maxima and minima of the Radius Difference Function are extracted and used to calculate similarity between objects. Similarity is calculated using Euclidean distance. Other scale-space approaches to shape representation use various techniques to maintain constant boundary arc length, that may otherwise change in non-intuitive ways over scale. We introduce the contour stability over scale property stating that the perceived boundary length should not change significantly over scale. Experiments show that significant similarity computation may be saved by using coarser scales without effectively reducing retrieval performance.

    Original languageEnglish (US)
    Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
    PublisherSociety of Photo-Optical Instrumentation Engineers
    Pages86-97
    Number of pages12
    Volume3972
    StatePublished - 2000
    EventProceedings of the 2000 'Storage and Retrieval for Media Databases 2000' - San Jose, CA, USA
    Duration: Jan 26 2000Jan 28 2000

    Other

    OtherProceedings of the 2000 'Storage and Retrieval for Media Databases 2000'
    CitySan Jose, CA, USA
    Period1/26/001/28/00

    Fingerprint

    Image retrieval
    retrieval
    polar coordinates
    radii
    centroids
    arcs
    Experiments

    ASJC Scopus subject areas

    • Electrical and Electronic Engineering
    • Condensed Matter Physics

    Cite this

    Hoffman, M. E., & Wong, E. (2000). Content-based image retrieval by scale-space object boundary shape representation. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 3972, pp. 86-97). Society of Photo-Optical Instrumentation Engineers.

    Content-based image retrieval by scale-space object boundary shape representation. / Hoffman, Mark E.; Wong, Edward.

    Proceedings of SPIE - The International Society for Optical Engineering. Vol. 3972 Society of Photo-Optical Instrumentation Engineers, 2000. p. 86-97.

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

    Hoffman, ME & Wong, E 2000, Content-based image retrieval by scale-space object boundary shape representation. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 3972, Society of Photo-Optical Instrumentation Engineers, pp. 86-97, Proceedings of the 2000 'Storage and Retrieval for Media Databases 2000', San Jose, CA, USA, 1/26/00.
    Hoffman ME, Wong E. Content-based image retrieval by scale-space object boundary shape representation. In Proceedings of SPIE - The International Society for Optical Engineering. Vol. 3972. Society of Photo-Optical Instrumentation Engineers. 2000. p. 86-97
    Hoffman, Mark E. ; Wong, Edward. / Content-based image retrieval by scale-space object boundary shape representation. Proceedings of SPIE - The International Society for Optical Engineering. Vol. 3972 Society of Photo-Optical Instrumentation Engineers, 2000. pp. 86-97
    @inproceedings{7641b0cb529a4c84bfd220f48a9ff6bb,
    title = "Content-based image retrieval by scale-space object boundary shape representation",
    abstract = "Shape is a popular feature used for content-based image retrieval. In this paper we propose a new method for image retrieval using a shape boundary represented in scale-space. The proposed method is suggested by the notion of `dynamic shape` where all 2-D boundary representations evolve from a single, primeval, featureless shape - a circle. Shape is represented by linearizing the boundary based on the polar coordinates of boundary points relative to the object's centroid. Points on the shape boundary are mapped to a primeval circle, and two functions are defined, the Radius Difference Function and the Angle Difference Function, and smoothed through scale-space to devolve the shape. Maxima and minima of the Radius Difference Function are extracted and used to calculate similarity between objects. Similarity is calculated using Euclidean distance. Other scale-space approaches to shape representation use various techniques to maintain constant boundary arc length, that may otherwise change in non-intuitive ways over scale. We introduce the contour stability over scale property stating that the perceived boundary length should not change significantly over scale. Experiments show that significant similarity computation may be saved by using coarser scales without effectively reducing retrieval performance.",
    author = "Hoffman, {Mark E.} and Edward Wong",
    year = "2000",
    language = "English (US)",
    volume = "3972",
    pages = "86--97",
    booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",
    publisher = "Society of Photo-Optical Instrumentation Engineers",

    }

    TY - GEN

    T1 - Content-based image retrieval by scale-space object boundary shape representation

    AU - Hoffman, Mark E.

    AU - Wong, Edward

    PY - 2000

    Y1 - 2000

    N2 - Shape is a popular feature used for content-based image retrieval. In this paper we propose a new method for image retrieval using a shape boundary represented in scale-space. The proposed method is suggested by the notion of `dynamic shape` where all 2-D boundary representations evolve from a single, primeval, featureless shape - a circle. Shape is represented by linearizing the boundary based on the polar coordinates of boundary points relative to the object's centroid. Points on the shape boundary are mapped to a primeval circle, and two functions are defined, the Radius Difference Function and the Angle Difference Function, and smoothed through scale-space to devolve the shape. Maxima and minima of the Radius Difference Function are extracted and used to calculate similarity between objects. Similarity is calculated using Euclidean distance. Other scale-space approaches to shape representation use various techniques to maintain constant boundary arc length, that may otherwise change in non-intuitive ways over scale. We introduce the contour stability over scale property stating that the perceived boundary length should not change significantly over scale. Experiments show that significant similarity computation may be saved by using coarser scales without effectively reducing retrieval performance.

    AB - Shape is a popular feature used for content-based image retrieval. In this paper we propose a new method for image retrieval using a shape boundary represented in scale-space. The proposed method is suggested by the notion of `dynamic shape` where all 2-D boundary representations evolve from a single, primeval, featureless shape - a circle. Shape is represented by linearizing the boundary based on the polar coordinates of boundary points relative to the object's centroid. Points on the shape boundary are mapped to a primeval circle, and two functions are defined, the Radius Difference Function and the Angle Difference Function, and smoothed through scale-space to devolve the shape. Maxima and minima of the Radius Difference Function are extracted and used to calculate similarity between objects. Similarity is calculated using Euclidean distance. Other scale-space approaches to shape representation use various techniques to maintain constant boundary arc length, that may otherwise change in non-intuitive ways over scale. We introduce the contour stability over scale property stating that the perceived boundary length should not change significantly over scale. Experiments show that significant similarity computation may be saved by using coarser scales without effectively reducing retrieval performance.

    UR - http://www.scopus.com/inward/record.url?scp=0033878271&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=0033878271&partnerID=8YFLogxK

    M3 - Conference contribution

    VL - 3972

    SP - 86

    EP - 97

    BT - Proceedings of SPIE - The International Society for Optical Engineering

    PB - Society of Photo-Optical Instrumentation Engineers

    ER -