Package boofcv.struct.gss

Interface GaussianScaleSpace<T extends ImageGray<T>,D extends ImageGray<D>>

All Known Implementing Classes:
NoCacheScaleSpace
public interface GaussianScaleSpace<T extends ImageGray<T>,D extends ImageGray<D>>

Interface for computing the scale space of an image and its derivatives. The scale space of an image is computed by convolving a Gaussian kernel across the image. The image's scale is determined by the Gaussian's standard deviation. See [1] for a summary of scale-space theory.

[1] Tony Lindeberg, "Scale-space: A framework for handling image structures at multiple scales," In. Proc. CERN School of Computing, Egmond aan Zee, The Netherlands, 8-21 September, 1996

  • Method Summary

    Modifier and Type
    Method
    Description
    BorderType
    getBorderType()
    Returns how image borders are processed.
    double
    getCurrentScale()
    Returns the value of the current active scale.
    D
    getDerivative(boolean... isX)
    Returns the partial derivative of the image.
    double
    getScale(int level)
    Returns the scale for the specified layer in the pyramid.
    T
    getScaledImage()
    Returns the scaled image at the active scale.
    int
    getTotalScales()
    Returns number of scaled images inside of this scale space.
    void
    setActiveScale(int index)
    Sets the active scale.
    void
    setBorderType(BorderType type)
    Change how image borders are handled.
    void
    setImage(T input)
    Specifies the original un-scaled image.
    void
    setScales(double... scales)
    Sets the scales/blur magnitudes for which the scale-space should be computed over.

  • Method Details

    • setScales

      void setScales(double... scales)
      Sets the scales/blur magnitudes for which the scale-space should be computed over.
      Parameters:
      scales - All the scales. These are absolute and not relative to the previous level.

    • getScale

      double getScale(int level)
      Returns the scale for the specified layer in the pyramid. This is equivalent to the standard deviation of the Gaussian convolved across the original input image.

    • setImage

      void setImage(T input)
      Specifies the original un-scaled image.
      Parameters:
      input - Original image.

    • setActiveScale

      void setActiveScale(int index)
      Sets the active scale. Must call setImage(ImageGray) before this function.
      Parameters:
      index - Index of active scale

    • getTotalScales

      int getTotalScales()
      Returns number of scaled images inside of this scale space.
      Returns:
      Number of scales.

    • getCurrentScale

      double getCurrentScale()
      Returns the value of the current active scale.
      Returns:
      active scale.

    • getScaledImage

      T getScaledImage()
      Returns the scaled image at the active scale.
      Returns:
      scaled image.

    • setBorderType

      void setBorderType(BorderType type)
      Change how image borders are handled.
      Parameters:
      type - The BorderType.

    • getBorderType

      BorderType getBorderType()
      Returns how image borders are processed.
      Returns:
      how image borders are processed.

    • getDerivative

      D getDerivative(boolean... isX)

      Returns the partial derivative of the image.

      Examples:
      derivative X = getDerivative(true)
      derivative Y = getDerivative(false)
      derivative XY = getDerivative(true,false)

      Parameters:
      isX - specifies which partial derivative is to be returned.
      Returns:
      The image's derivative.