Elliptical object detection by a modified ransac with sampling constraint from boundary Curves' clustering

Yingdi Xie, Jun Ohya

    研究成果: Article

    5 引用 (Scopus)

    抄録

    This paper proposes a method for detecting ellipses from an image despite (1) multiple colors within the ellipses, (2) partially occluded ellipses' boundaries, (3) noisy, locally deformed boundaries of ellipses, (4) presence of multiple objects other than the ellipses in the image, and (5) combinations of (1) through (4). After boundary curves are obtained by edge detection, by utilizing the first-order difference curves of the edge orientation of each pixel in the boundary curves, a segment-reconnect method obtains boundary clusters. Then, a modified RANSAC detects ellipses by choosing five pixels randomly from the boundary clusters, where overlapped ellipses are merged. Experimental results using synthesized images and real images demonstrate the effectiveness of the proposed method together with comparison with the Randomized Hough Transform, a wellknown conventional method.

    元の言語English
    ページ(範囲)611-623
    ページ数13
    ジャーナルIEICE Transactions on Information and Systems
    E93-D
    発行部数3
    DOI
    出版物ステータスPublished - 2010

    Fingerprint

    Pixels
    Sampling
    Hough transforms
    Edge detection
    Color
    Object detection

    ASJC Scopus subject areas

    • Electrical and Electronic Engineering
    • Software
    • Artificial Intelligence
    • Hardware and Architecture
    • Computer Vision and Pattern Recognition

    これを引用

    @article{cf9ffd96b886453198a127951a4e81c0,
    title = "Elliptical object detection by a modified ransac with sampling constraint from boundary Curves' clustering",
    abstract = "This paper proposes a method for detecting ellipses from an image despite (1) multiple colors within the ellipses, (2) partially occluded ellipses' boundaries, (3) noisy, locally deformed boundaries of ellipses, (4) presence of multiple objects other than the ellipses in the image, and (5) combinations of (1) through (4). After boundary curves are obtained by edge detection, by utilizing the first-order difference curves of the edge orientation of each pixel in the boundary curves, a segment-reconnect method obtains boundary clusters. Then, a modified RANSAC detects ellipses by choosing five pixels randomly from the boundary clusters, where overlapped ellipses are merged. Experimental results using synthesized images and real images demonstrate the effectiveness of the proposed method together with comparison with the Randomized Hough Transform, a wellknown conventional method.",
    keywords = "A modified RANSAC, Ellipse detection",
    author = "Yingdi Xie and Jun Ohya",
    year = "2010",
    doi = "10.1587/transinf.E93.D.611",
    language = "English",
    volume = "E93-D",
    pages = "611--623",
    journal = "IEICE Transactions on Information and Systems",
    issn = "0916-8532",
    publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
    number = "3",

    }

    TY - JOUR

    T1 - Elliptical object detection by a modified ransac with sampling constraint from boundary Curves' clustering

    AU - Xie, Yingdi

    AU - Ohya, Jun

    PY - 2010

    Y1 - 2010

    N2 - This paper proposes a method for detecting ellipses from an image despite (1) multiple colors within the ellipses, (2) partially occluded ellipses' boundaries, (3) noisy, locally deformed boundaries of ellipses, (4) presence of multiple objects other than the ellipses in the image, and (5) combinations of (1) through (4). After boundary curves are obtained by edge detection, by utilizing the first-order difference curves of the edge orientation of each pixel in the boundary curves, a segment-reconnect method obtains boundary clusters. Then, a modified RANSAC detects ellipses by choosing five pixels randomly from the boundary clusters, where overlapped ellipses are merged. Experimental results using synthesized images and real images demonstrate the effectiveness of the proposed method together with comparison with the Randomized Hough Transform, a wellknown conventional method.

    AB - This paper proposes a method for detecting ellipses from an image despite (1) multiple colors within the ellipses, (2) partially occluded ellipses' boundaries, (3) noisy, locally deformed boundaries of ellipses, (4) presence of multiple objects other than the ellipses in the image, and (5) combinations of (1) through (4). After boundary curves are obtained by edge detection, by utilizing the first-order difference curves of the edge orientation of each pixel in the boundary curves, a segment-reconnect method obtains boundary clusters. Then, a modified RANSAC detects ellipses by choosing five pixels randomly from the boundary clusters, where overlapped ellipses are merged. Experimental results using synthesized images and real images demonstrate the effectiveness of the proposed method together with comparison with the Randomized Hough Transform, a wellknown conventional method.

    KW - A modified RANSAC

    KW - Ellipse detection

    UR - http://www.scopus.com/inward/record.url?scp=77950193677&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=77950193677&partnerID=8YFLogxK

    U2 - 10.1587/transinf.E93.D.611

    DO - 10.1587/transinf.E93.D.611

    M3 - Article

    AN - SCOPUS:77950193677

    VL - E93-D

    SP - 611

    EP - 623

    JO - IEICE Transactions on Information and Systems

    JF - IEICE Transactions on Information and Systems

    SN - 0916-8532

    IS - 3

    ER -