SoCurvatureDrivenDiffusionProcessing Class Reference
[Smoothing And Denoising]

SoCurvatureDrivenDiffusionProcessing engine More...

#include <ImageViz/Engines/ImageFiltering/SmoothingAndDenoising/SoCurvatureDrivenDiffusionProcessing.h>

Inheritance diagram for SoCurvatureDrivenDiffusionProcessing:

List of all members.

Public Member Functions
	SoCurvatureDrivenDiffusionProcessing ()
Public Attributes
SoSFEnum	computeMode
SoSFImageDataAdapter	inImage
SoSFImageDataAdapter	inMaskImage
SoSFInt32	iterationNumber
SoSFFloat	sharpnessFactor
SoSFFloat	anisotropyFactor
SoImageVizEngineOutput < SoSFImageDataAdapter, SoImageDataAdapter * >	outImage

---

Detailed Description

SoCurvatureDrivenDiffusionProcessing engine

The SoCurvatureDrivenDiffusionProcessing engine smooths an image using an advanced local edge analysis technique.

For an introduction to image filters, see section Images Filters.

This engine smooths an image using an advanced local edge analysis technique. This algorithm allows smoothing an image with respect to its edges.

Figure 1: Curvature-Driven diffusion on a noisy image with default parameters (10 iterations, sharpness = 0.9 and anisotropy = 0.6)

The algorithm is ruled by three main parameters: number of iterations, sharpness and anisotropy.

• User must specify the requested number of iterations which should be suitable to reach convergence. In practice, a default value of 10 iterations allows obtaining good results in almost all cases.

• The sharpness parameter control amplitude of smoothing along edge direction. A small sharpness value results in a global image smoothing.

Figure 2: Influence of sharpness parameter on curvature-driven diffusion process (on the left, sharpness = 0.1; on the right, sharpness = 2.0)

• The anisotropy parameter control local curvature-driven diffusion along gradient direction (direction orthogonal to edge). A great value for anisotropy parameter result in a stronger local smoothing of edge regions.

Figure 3: Influence of anisotropy parameter on curvature-driven diffusion process (on the left, anisotropy = 0.3; on the right, anisotropy = 0.9)

SEE ALSO

SoRecursiveExponentialFilterProcessing.

FILE FORMAT/DEFAULT

CurvatureDrivenDiffusionProcessing {

computeMode	MODE_AUTO
inImage	NULL
inMaskImage	NULL
iterationNumber	10
sharpnessFactor	0.9f
anisotropyFactor	0.6f

}

Library references: anisotropicdiffusion

---

Constructor & Destructor Documentation

SoCurvatureDrivenDiffusionProcessing::SoCurvatureDrivenDiffusionProcessing

(

)

Constructor.

---

Member Data Documentation

SoSFFloat SoCurvatureDrivenDiffusionProcessing::anisotropyFactor

The anisotropy factor.

Default value is 0.6f.

SoSFEnum SoCurvatureDrivenDiffusionProcessing::computeMode

Select the compute Mode (2D or 3D or AUTO) Use enum ComputeMode.

Default is MODE_AUTO

SoSFImageDataAdapter SoCurvatureDrivenDiffusionProcessing::inImage

The input image.

Default value is NULL. Supported types include: grayscale binary label color image.

SoSFImageDataAdapter SoCurvatureDrivenDiffusionProcessing::inMaskImage

The binary image for the mask or empty (command looks for only inside).

This parameter is optional. Default value is NULL. Supported types include: binary color image.

SoSFInt32 SoCurvatureDrivenDiffusionProcessing::iterationNumber

The number of iterations (strictly positive integer).

Default value is 10.

SoImageVizEngineOutput<SoSFImageDataAdapter,SoImageDataAdapter*> SoCurvatureDrivenDiffusionProcessing::outImage

The output image.

Default value is NULL. Supported types include: grayscale binary label color image.

SoSFFloat SoCurvatureDrivenDiffusionProcessing::sharpnessFactor

The sharpness factor.

Default value is 0.9f.

---

The documentation for this class was generated from the following file:

• ImageViz/Engines/ImageFiltering/SmoothingAndDenoising/SoCurvatureDrivenDiffusionProcessing.h