SoGaussianGradientTensorProcessing2d engine computes the structure tensor. More...
#include <ImageViz/Engines/EdgeDetection/EdgeMarking/SoGaussianGradientTensorProcessing2d.h>
Public Member Functions | |
SoGaussianGradientTensorProcessing2d () | |
Public Attributes | |
SoSFImageDataAdapter | inImage |
SoSFVec2f | standardDeviation |
SoImageVizEngineOutput < SoSFImageDataAdapter, SoImageDataAdapter * > | outTensorImage |
The SoGradientOperatorProcessing2d engine computes the structure tensor of the input image by convolving the input image with the square first order derivatives of a Gaussian Kernel.
For instance, , .
This filter provides an output spectral image where each channel represents a product of two partial derivative set in the following order , , .
In order to extract the eigenvalues or vectors of the result image the SoEigenDecompositionProcessing2d command can be applied on the spectral image output.
SoEigenDecompositionProcessing2d, SoHessianMatrixProcessing2d, SoEigenvaluesToStructurenessProcessing2d.
inImage | NULL |
standardDeviation | 3.f 3.f |
SoGaussianGradientTensorProcessing2d::SoGaussianGradientTensorProcessing2d | ( | ) |
Constructor.
inImage Default value is NULL.
Supported types include: grayscale color image.
SoImageVizEngineOutput<SoSFImageDataAdapter,SoImageDataAdapter*> SoGaussianGradientTensorProcessing2d::outTensorImage |
outTensorImage Default value is NULL.
Supported types include: grayscale binary label color image.
The standard deviation for each direction (X, Y) of the Gaussian Kernel that is used for computing first order derivatives.
Each value must be greater than or equal to 0.1. Default value is SbVec2f( 3.f, 3.f ).