Segmentation01

Embed Size (px)

Citation preview

  • 7/30/2019 Segmentation01

    1/8

    Image Segmentation

    Content

    Definition and methods classification

    Detection of discontinuities

    Point detection

    Line detection

    Edge detection

    Edge linking and boundary detection

    Thresholding

    Region-based segmentation

  • 7/30/2019 Segmentation01

    2/8

    Definition: Segmentation subdivides an image into its constituent

    regions or objects.

    Basic formulation: LetR represent the entire spatial region occupied

    by an image. Image segmentation is a process that partitionsR into

    n sub-regions,R1,R2,,Rn, such that

    a)

    b) Ri is a connected set. i = 1, 2,, n

    c) RiRj= for all i andj, i j

    d) Q (Ri) = TRUE for i = 1, 2,, n

    e) Q = FALSE for any adjacent regionsRi andRj

    where Q(Rk) is a logical predicate defined over the points in set Rk

    Fundamentals

  • 7/30/2019 Segmentation01

    3/8

    Classification: based on one of two basic categories dealing withproperties of intensity values

    discontinuity and similarity

    discontinuity-based: the partition of an image is based on abrupt

    changes in intensity, such as point, line and edge.

    similarity-based: partitions an image into regions that are similaraccording to a set of predefined criteria. Thresholding, regiongrowing, and region splitting and merging are examples.

    Methods Classification

  • 7/30/2019 Segmentation01

    4/8

  • 7/30/2019 Segmentation01

    5/8

    Edge pixels: pixels at which the intensity of an image functionchanges abruptly

    Edges/Edge segments: sets of connected edge pixels

    Edge detectors: local image processing methods designed to detectedge pixels

    Line: an edge segment in which the intensity of the background on

    either side of the line is either much higher or much lower than the

    intensity of the line pixels

    Isolated point: a line whose length and width are equal to one pixel

    Detection of discontinuities: background

  • 7/30/2019 Segmentation01

    6/8

  • 7/30/2019 Segmentation01

    7/8

  • 7/30/2019 Segmentation01

    8/8

    Conclusions

    First-order derivatives generally produce thicker edges in an image

    Second-order derivatives have a stronger response to fine detail,

    such as thin lines, isolated points, and noise

    Second-order derivatives produce a double-edge response at ramp

    and step transitions in intensity

    The sign of the second derivative can be used to determine

    whether a transition into an edge is from light to dark or dark tolight

    Detection of discontinuities: background (contd)