#include <itkShapeDetectionLevelSetImageFilter.h>

Inheritance diagram for itk::ShapeDetectionLevelSetImageFilter< TInputImage, TFeatureImage, TOutputPixelType >:

Collaboration diagram for itk::ShapeDetectionLevelSetImageFilter< TInputImage, TFeatureImage, TOutputPixelType >:

Public Types
typedef SmartPointer< const Self >	ConstPointer
typedef Superclass::FeatureImageType	FeatureImageType
typedef Superclass::OutputImageType	OutputImageType
typedef SmartPointer< Self >	Pointer
typedef ShapeDetectionLevelSetImageFilter	Self
typedef ShapeDetectionFunctionType::Pointer	ShapeDetectionFunctionPointer
typedef ShapeDetectionLevelSetFunction < OutputImageType, FeatureImageType >	ShapeDetectionFunctionType
typedef SegmentationLevelSetImageFilter < TInputImage, TFeatureImage, TOutputPixelType >	Superclass
typedef Superclass::ValueType	ValueType
Public Member Functions
virtual const char *	GetNameOfClass () const
Protected Member Functions
void	GenerateData ()
void	operator= (const Self &)
virtual void	PrintSelf (std::ostream &os, Indent indent) const
	ShapeDetectionLevelSetImageFilter (const Self &)
Private Attributes
ShapeDetectionFunctionPointer	m_ShapeDetectionFunction
virtual ::itk::LightObject::Pointer	CreateAnother (void) const
static Pointer	New ()
	~ShapeDetectionLevelSetImageFilter ()
	ShapeDetectionLevelSetImageFilter ()

Detailed Description

template<class TInputImage, class TFeatureImage, class TOutputPixelType = float>
class itk::ShapeDetectionLevelSetImageFilter< TInputImage, TFeatureImage, TOutputPixelType >

Segments structures in images based on a user supplied edge potential map.

IMPORTANT: The SegmentationLevelSetImageFilter class and the ShapeDetectionLevelSetFunction class contain additional information necessary to gain full understanding of how to use this filter.

OVERVIEW: This class is a level set method segmentation filter. An initial contour is propagated outwards (or inwards) until it ''sticks'' to the shape boundaries. This is done by using a level set speed function based on a user supplied edge potential map. This approach for segmentation follows that of Malladi et al (1995).

INPUTS: This filter requires two inputs. The first input is a initial level set. The initial level set is a real image which contains the initial contour/surface as the zero level set. For example, a signed distance function from the initial contour/surface is typically used. Note that for this algorithm the initial contour has to be wholly within (or wholly outside) the structure to be segmented.

: The second input is the feature image. For this filter, this is the edge potential map. General characteristics of an edge potential map is that it has values close to zero in regions near the edges and values close to one inside the shape itself. Typically, the edge potential map is compute from the image gradient, for example:

$g(I) = 1 / ( 1 + | (\nabla * G)(I)| )$

$g(I) = \exp^{-|(\nabla * G)(I)|}$

where $ I $ is image intensity and $(\nabla * G)$ is the derivative of Gaussian operator.

: See SegmentationLevelSetImageFilter and SparseFieldLevelSetImageFilter for more information on Inputs.

PARAMETERS: The PropagationScaling parameter can be used to switch from propagation outwards (POSITIVE scaling parameter) versus propagating inwards (NEGATIVE scaling parameter).

The smoothness of the resulting contour/surface can be adjusted using a combination of PropagationScaling and CurvatureScaling parameters. The larger the CurvatureScaling parameter, the smoother the resulting contour. The CurvatureScaling parameter should be non-negative for proper operation of this algorithm. To follow the implementation in Malladi et al paper, set the PropagtionScaling to $\pm 1.0$ and CurvatureScaling to $\epsilon$ .

Note that there is no advection term for this filter. Setting the advection scaling will have no effect.

OUTPUTS: The filter outputs a single, scalar, real-valued image. Negative values in the output image represent the inside of the segmentated region and positive values in the image represent the outside of the segmented region. The zero crossings of the image correspond to the position of the propagating front.

: See SparseFieldLevelSetImageFilter and SegmentationLevelSetImageFilter for more information.

REFERENCES

: "Shape Modeling with Front Propagation: A Level Set Approach", R. Malladi, J. A. Sethian and B. C. Vermuri. IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol 17, No. 2, pp 158-174, February 1995

See also:: SegmentationLevelSetImageFilter; ShapeDetectionLevelSetFunction; SparseFieldLevelSetImageFilter

Definition at line 108 of file itkShapeDetectionLevelSetImageFilter.h.