Scale-space approach to image thinning using the most prominent ridge-line in the image pyramid data structure

Mark E. Hoffman, Edward Wong

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

    Abstract

    Image thinning methods can be divided into two categories based on the type of image they are designed to thin: binary image thinning and grayscale image thinning. Typically, grayscale images are thresholded to allow binary image thinning methods to be applied. However, thresholding grayscale images may introduce uneven object contours that are a difficulty for binary methods. The scale-space approach to image thinning includes scale as an additional dimension where images at scale t are derived from the original image at scale zero by applying the Gaussian filter. As scale increases finer image structure is suppressed. By treating the image as a 3-D surface with intensity as the third dimension, the Most Prominent Ridge-Line (MPRL) is the union of topographical features: peak, ridge, and saddle point, such that each has greatest contrast with its surroundings. The MPRL is computed by minimizing its second spatial derivative over scale. The result forms a trajectory in scale-space. The thinned image is the projection of the MPRL on the base level. The MPRL has been implemented using the image pyramid data structure, and has been applied to binary and grayscale images of printed characters. Experimental results show that the method is less sensitive to contour unevenness. It also offers the option of choosing different levels of fine structure to include.

    Original languageEnglish (US)
    Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
    EditorsD.P. Lopresti, J. Zhou
    Pages242-252
    Number of pages11
    Volume3305
    DOIs
    StatePublished - 1998
    EventDocuments Recognition V - San Jose, CA, United States
    Duration: Jan 28 1998Jan 29 1998

    Other

    OtherDocuments Recognition V
    CountryUnited States
    CitySan Jose, CA
    Period1/28/981/29/98

    Fingerprint

    Image thinning
    data structures
    pyramids
    Data structures
    ridges
    Binary images
    Trajectories
    Derivatives

    Keywords

    • Contour perturbation
    • Image pyramid
    • Image thinning
    • Ridge-line
    • Scale-space
    • Sensitivity
    • Topography

    ASJC Scopus subject areas

    • Electrical and Electronic Engineering
    • Condensed Matter Physics

    Cite this

    Hoffman, M. E., & Wong, E. (1998). Scale-space approach to image thinning using the most prominent ridge-line in the image pyramid data structure. In D. P. Lopresti, & J. Zhou (Eds.), Proceedings of SPIE - The International Society for Optical Engineering (Vol. 3305, pp. 242-252) https://doi.org/10.1117/12.304636

    Scale-space approach to image thinning using the most prominent ridge-line in the image pyramid data structure. / Hoffman, Mark E.; Wong, Edward.

    Proceedings of SPIE - The International Society for Optical Engineering. ed. / D.P. Lopresti; J. Zhou. Vol. 3305 1998. p. 242-252.

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

    Hoffman, ME & Wong, E 1998, Scale-space approach to image thinning using the most prominent ridge-line in the image pyramid data structure. in DP Lopresti & J Zhou (eds), Proceedings of SPIE - The International Society for Optical Engineering. vol. 3305, pp. 242-252, Documents Recognition V, San Jose, CA, United States, 1/28/98. https://doi.org/10.1117/12.304636
    Hoffman ME, Wong E. Scale-space approach to image thinning using the most prominent ridge-line in the image pyramid data structure. In Lopresti DP, Zhou J, editors, Proceedings of SPIE - The International Society for Optical Engineering. Vol. 3305. 1998. p. 242-252 https://doi.org/10.1117/12.304636
    Hoffman, Mark E. ; Wong, Edward. / Scale-space approach to image thinning using the most prominent ridge-line in the image pyramid data structure. Proceedings of SPIE - The International Society for Optical Engineering. editor / D.P. Lopresti ; J. Zhou. Vol. 3305 1998. pp. 242-252
    @inproceedings{540ff3835ea34797822dc6199d7cbd9d,
    title = "Scale-space approach to image thinning using the most prominent ridge-line in the image pyramid data structure",
    abstract = "Image thinning methods can be divided into two categories based on the type of image they are designed to thin: binary image thinning and grayscale image thinning. Typically, grayscale images are thresholded to allow binary image thinning methods to be applied. However, thresholding grayscale images may introduce uneven object contours that are a difficulty for binary methods. The scale-space approach to image thinning includes scale as an additional dimension where images at scale t are derived from the original image at scale zero by applying the Gaussian filter. As scale increases finer image structure is suppressed. By treating the image as a 3-D surface with intensity as the third dimension, the Most Prominent Ridge-Line (MPRL) is the union of topographical features: peak, ridge, and saddle point, such that each has greatest contrast with its surroundings. The MPRL is computed by minimizing its second spatial derivative over scale. The result forms a trajectory in scale-space. The thinned image is the projection of the MPRL on the base level. The MPRL has been implemented using the image pyramid data structure, and has been applied to binary and grayscale images of printed characters. Experimental results show that the method is less sensitive to contour unevenness. It also offers the option of choosing different levels of fine structure to include.",
    keywords = "Contour perturbation, Image pyramid, Image thinning, Ridge-line, Scale-space, Sensitivity, Topography",
    author = "Hoffman, {Mark E.} and Edward Wong",
    year = "1998",
    doi = "10.1117/12.304636",
    language = "English (US)",
    volume = "3305",
    pages = "242--252",
    editor = "D.P. Lopresti and J. Zhou",
    booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

    }

    TY - GEN

    T1 - Scale-space approach to image thinning using the most prominent ridge-line in the image pyramid data structure

    AU - Hoffman, Mark E.

    AU - Wong, Edward

    PY - 1998

    Y1 - 1998

    N2 - Image thinning methods can be divided into two categories based on the type of image they are designed to thin: binary image thinning and grayscale image thinning. Typically, grayscale images are thresholded to allow binary image thinning methods to be applied. However, thresholding grayscale images may introduce uneven object contours that are a difficulty for binary methods. The scale-space approach to image thinning includes scale as an additional dimension where images at scale t are derived from the original image at scale zero by applying the Gaussian filter. As scale increases finer image structure is suppressed. By treating the image as a 3-D surface with intensity as the third dimension, the Most Prominent Ridge-Line (MPRL) is the union of topographical features: peak, ridge, and saddle point, such that each has greatest contrast with its surroundings. The MPRL is computed by minimizing its second spatial derivative over scale. The result forms a trajectory in scale-space. The thinned image is the projection of the MPRL on the base level. The MPRL has been implemented using the image pyramid data structure, and has been applied to binary and grayscale images of printed characters. Experimental results show that the method is less sensitive to contour unevenness. It also offers the option of choosing different levels of fine structure to include.

    AB - Image thinning methods can be divided into two categories based on the type of image they are designed to thin: binary image thinning and grayscale image thinning. Typically, grayscale images are thresholded to allow binary image thinning methods to be applied. However, thresholding grayscale images may introduce uneven object contours that are a difficulty for binary methods. The scale-space approach to image thinning includes scale as an additional dimension where images at scale t are derived from the original image at scale zero by applying the Gaussian filter. As scale increases finer image structure is suppressed. By treating the image as a 3-D surface with intensity as the third dimension, the Most Prominent Ridge-Line (MPRL) is the union of topographical features: peak, ridge, and saddle point, such that each has greatest contrast with its surroundings. The MPRL is computed by minimizing its second spatial derivative over scale. The result forms a trajectory in scale-space. The thinned image is the projection of the MPRL on the base level. The MPRL has been implemented using the image pyramid data structure, and has been applied to binary and grayscale images of printed characters. Experimental results show that the method is less sensitive to contour unevenness. It also offers the option of choosing different levels of fine structure to include.

    KW - Contour perturbation

    KW - Image pyramid

    KW - Image thinning

    KW - Ridge-line

    KW - Scale-space

    KW - Sensitivity

    KW - Topography

    UR - http://www.scopus.com/inward/record.url?scp=0032225264&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=0032225264&partnerID=8YFLogxK

    U2 - 10.1117/12.304636

    DO - 10.1117/12.304636

    M3 - Conference contribution

    VL - 3305

    SP - 242

    EP - 252

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

    A2 - Lopresti, D.P.

    A2 - Zhou, J.

    ER -