Package gov.nih.mipav.model.algorithms.levelset

Class LseGradientAnisotropic3

  • public class LseGradientAnisotropic3
extends LsePdeFilter3

    This class implements the gradient anistropic curvature flow filter for 3D images using a finite-difference-based solver for the partial differential equation:

        du/dt = divergence(exp(-0.5*|grad(u)|^2/(k^2*average(|grad(u)|^2))*grad(u))

    where u(x,y,z,t) is the evolved image at time t, du/dt is the time derivative of u, and grad(u) is the gradient of u with respect to the spatial variables. The value k is a used-selected parameter and may be thought of as a local curvature parameter. The notation divergence(Q) = Q0_x + Q1_y + Q2_z, the sum of first-order derivatives of the vector-valued function Q = (Q0,Q1).

    Version:
    0.1 November 7, 2006
    Author:
    David Eberly

    • Field Detail

      • m_fK

        private float m_fK
        The local curvature parameter, call it k.

      • m_fParam

        private float m_fParam
        1/(k^2*average(|grad(u)|^2))

      • m_fMHalfParam

        private float m_fMHalfParam
        -0.5/(k^2*average(|grad(u)|^2))

    • Constructor Detail

      • LseGradientAnisotropic3

        public LseGradientAnisotropic3​(int iXBound,
                               int iYBound,
                               int iZBound,
                               float fXSpacing,
                               float fYSpacing,
                               float fZSpacing,
                               float[] afData,
                               boolean[] abMask,
                               float fBorderValue,
                               int eScaleType,
                               float fK)
        Creates a new LseCurvatureFlow3 object.
        Parameters:
        iXBound - The x-bound of the image.
        iYBound - The y-bound of the image.
        iZBound - The z-bound of the image.
        fXSpacing - The x-spacing of the image.
        fYSpacing - The y-spacing of the image.
        fZSpacing - The z-spacing of the image.
        afData - The image elements, stored in lexicographical order.
        abMask - The image mask, stored in lexicographical order. A pixel value is processed by the PDE solver only when the mask value is false.
        fBorderValue - Specifies how to handle the image value. When set to Float.MAX_VALUE, Neumann conditions are in use, in which case zero-valued derivatives are assumed on the image border. Otherwise, Dirichlet conditions are used, in which case the image is assumed to be constant on the border with value specified by fBorderValue.
        eScaleType - The type of scaling to apply to the input image. The choices are NONE, UNIT, SYMMETRIC, or PRESERVE_ZERO.
        fK - The local curvature parameter in the exponential term of the PDE.

    • Method Detail

      • computeParam

        protected void computeParam()
        Compute the average gradient magnitude squared for the current image u(x,y,t). The terms p = 1/(k*k*average(|grad(u)|^2) and -0.5*p are computed once and used in the per-pixel update in the PDE solver.

      • onPreUpdate

        protected void onPreUpdate()
        The function onPreUpdate was abstractly defined in the LsePdeFilter base class. It is called before any per-pixel updates. In this class, the parameters in the exponential term of the PDE are precomputed for use by the onUpdate() function.
        Overrides:
        onPreUpdate in class LsePdeFilter3

      • onUpdate

        protected void onUpdate​(int iX,
                        int iY,
                        int iZ)
        The per-voxel update function of the PDE solver.
        Specified by:
        onUpdate in class LsePdeFilter3
        Parameters:
        iX - The x-value of the voxel to be updated.
        iY - The y-value of the voxel to be updated.
        iZ - The z-value of the voxel to be updated.