Scribd Logo
Importer

Bienvenue sur Scribd !

  • Sobel Erosion Dilation Examples

    Transféré par

    Vu Van Thien
    288 vues8 pages
    The document describes examples of image processing techniques using morphological operations and filters: - It shows how to compute gradients Gx and Gy using Sobel filters and zero-padding for boundary handling. - An example of Laplacian filtering is shown using a Laplacian kernel and border replication. - Morphological erosion and dilation are demonstrated using structuring elements and constant padding. - A dilation followed by erosion is applied to another image using a structuring element. In all cases border replication or constant values are used to handle image boundaries before applying the linear or morphological operations.

    Description originale:

    Copyright

    © Attribution Non-Commercial (BY-NC)

    Formats disponibles

    PPT, PDF, TXT ou lisez en ligne sur Scribd

    Avez-vous trouvé ce document utile ?

    Ce contenu est-il inapproprié ?

    Signaler ce document
    The document describes examples of image processing techniques using morphological operations and filters: - It shows how to compute gradients Gx and Gy using Sobel filters and zero-padding for boundary handling. - An example of Laplacian filtering is shown using a Laplacian kernel and border replication. - Morphological erosion and dilation are demonstrated using structuring elements and constant padding. - A dilation followed by erosion is applied to another image using a structuring element. In all cases border replication or constant values are used to handle image boundaries before applying the linear or morphological operations.

    Droits d'auteur :

    Attribution Non-Commercial (BY-NC)
    Téléchargez comme PPT, PDF, TXT ou lisez en ligne sur Scribd
    Signaler comme contenu inapproprié
    288 vues8 pages

    Sobel Erosion Dilation Examples

    Transféré par

    Vu Van Thien
    The document describes examples of image processing techniques using morphological operations and filters: - It shows how to compute gradients Gx and Gy using Sobel filters and zero-padding for boundary handling. - An example of Laplacian filtering is shown using a Laplacian kernel and border replication. - Morphological erosion and dilation are demonstrated using structuring elements and constant padding. - A dilation followed by erosion is applied to another image using a structuring element. In all cases border replication or constant values are used to handle image boundaries before applying the linear or morphological operations.

    Droits d'auteur :

    Attribution Non-Commercial (BY-NC)
    Téléchargez comme PPT, PDF, TXT ou lisez en ligne sur Scribd
    Signaler comme contenu inapproprié
    sur 8

    Sobel filter example

    Compute Gx and Gy, gradients of the image performing the


    convolution of Sobel kernels with the image
    Use zero-padding to extend the image
    0
    30 30 0
    -30

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    Gx

    Gy

    x
    1

    -1

    -1

    -2

    -1

    -2

    -1

    hx

    hy

    40

    40

    -40

    40

    40

    -40

    40

    40

    -40

    30

    30

    -10 -30 -40 -30

    -30

    10

    30

    40

    30

    Sobel filter example


    Compute Gx and Gy, gradients of the image performing the
    convolution of Sobel kernels with the image
    Use border values to extend the image 0 40 40 0 0

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    Gx

    Gy

    x
    1

    40

    40

    40

    40

    40

    40

    40

    40

    0
    0

    -1

    -1

    -2

    -1

    -2

    -1

    hx

    hy

    Gy
    arctan

    G
    x

    Sobel filter example


    Compute Gx and Gy, gradients of the image performing the
    convolution of Sobel kernels with the image
    Use border values to extend the image 0 0 0 0 0

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    Gx

    -40 -40 -40 -40 -40

    Gy

    -40 -40 -40 -40 -40


    0

    x
    1

    0
    0

    -1

    -1

    -2

    -1

    -2

    -1

    hx

    hy

    Gy
    arctan

    G
    x

    -/2

    -/2

    -/2

    -/2

    -/2
    -/2

    -/2

    -/2

    -/2

    -/2

    Sobel filter example


    Compute Gx and Gy, gradients of the image performing the
    convolution of Sobel kernels with the image
    Use border values to extend the image 0 0 10 40 30

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    Gx

    10

    30

    30

    10

    10

    30

    30

    10

    20

    30

    10

    10

    10

    -10 -20 -10

    -10 -30 -30 -10

    -40 -30 -10


    -30 -10 0
    0

    -1

    -1

    -2

    -1

    -2

    -1

    hx

    hy

    -/4

    Gy
    arctan

    G
    x

    -10 -30 -30 -10 0

    Gy

    x
    1

    0
    0
    -0.15

    -/4

    -/4

    -/4

    -/4

    -/4

    -/4

    -/4

    -0.35

    -/4

    -/4

    -0.40

    -/4

    0
    0
    -0.10
    -/4

    Laplacian example
    Compute the convolution of image I with the Laplacian
    kernel
    Use border values to extend the image

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    10

    I
    1

    -8

    Laplacian

    10

    40

    -20

    10

    30

    -30 -10

    10

    30

    -30 -10

    40

    -30

    10

    -20 -10

    x
    Laplacian*I

    Erosion example
    Erode the image below using kernels S1 and S2
    Use border values to extend the image
    fS1

    x
    1

    S1

    S2

    fS2

    Dilation example
    Erode the image below using kernels S1 and S2
    Use border values to extend the image
    fS1

    x
    1

    S1

    S2

    fS2

    Dilation-Erosion example
    Dilate then erode the image below using kernels S2
    Use border values to extend the image
    fS2

    x
    0

    S2

    (fS2)S2

    Vous aimerez peut-être aussi