#include <itkAntiAliasBinaryImageFilter.h>

Inheritance diagram for itk::AntiAliasBinaryImageFilter< TInputImage, TOutputImage >:

Collaboration diagram for itk::AntiAliasBinaryImageFilter< TInputImage, TOutputImage >:

Public Types
typedef TInputImage::ValueType	BinaryValueType
typedef SmartPointer< const Self >	ConstPointer
typedef CurvatureFlowFunction < OutputImageType >	CurvatureFunctionType
typedef Superclass::IndexType	IndexType
typedef Superclass::InputImageType	InputImageType
typedef Superclass::OutputImageType	OutputImageType
typedef SmartPointer< Self >	Pointer
typedef AntiAliasBinaryImageFilter	Self
typedef SparseFieldLevelSetImageFilter < TInputImage, TOutputImage >	Superclass
typedef Superclass::TimeStepType	TimeStepType
typedef Superclass::ValueType	ValueType
Public Member Functions
virtual ::itk::LightObject::Pointer	CreateAnother (void) const
unsigned int	GetMaximumIterations ()
virtual const char *	GetNameOfClass () const
	typedef (Concept::Convertible< double, typename TOutputImage::PixelType >) DoubleConvertibleToOutputCheck
	typedef (Concept::OStreamWritable< typename TInputImage::PixelType >) InputOStreamWritableCheck

virtual BinaryValueType	GetUpperBinaryValue () const
virtual BinaryValueType	GetLowerBinaryValue () const

void	SetMaximumIterations (unsigned int i)
Static Public Member Functions
static Pointer	New ()
Protected Member Functions
virtual ValueType	CalculateUpdateValue (const IndexType &idx, const TimeStepType &dt, const ValueType &value, const ValueType &change)
void	GenerateData ()

	AntiAliasBinaryImageFilter ()
	~AntiAliasBinaryImageFilter ()
virtual void	PrintSelf (std::ostream &os, Indent indent) const
Private Member Functions
	AntiAliasBinaryImageFilter (const Self &)
void	operator= (const Self &)
Private Attributes
CurvatureFunctionType::Pointer	m_CurvatureFunction
const TInputImage *	m_InputImage
BinaryValueType	m_LowerBinaryValue
BinaryValueType	m_UpperBinaryValue

Detailed Description

template<class TInputImage, class TOutputImage>
class itk::AntiAliasBinaryImageFilter< TInputImage, TOutputImage >

A method for estimation of a surface from a binary volume.

: This filter implements a surface-fitting method for estimation of a surface from a binary volume. This process can be used to reduce aliasing artifacts which result in visualization of binary partitioned surfaces.

: The binary volume (filter input) is used as a set of constraints in an iterative relaxation process of an estimated ND surface. The surface is described implicitly as the zero level set of a volume $\phi$ and allowed to deform under curvature flow. A set of contraints is imposed on this movement as follows:

: $u_{i,j,k}^{n+1} = \left\{ \begin{array}{ll} \mbox{max} (u_{i,j,k}^{n} + \Delta t H_{i,j,k}^{n}, 0) & \mbox{$B_{i,j,k} = 1$} \\ \mbox{min} (u_{i,j,k}^{n} + \Delta t H_{i,j,k}^{n}, 0) & \mbox{$B_{i,j,k} = -1$} \end{array}\right.$

: where $u_{i,j,k}^{n}$ is the value of $\phi$ at discrete index and iteration , is the gradient magnitude times mean curvature of $\phi$ , and is the binary input volume, with 1 denoting an inside pixel and -1 denoting an outside pixel.

NOTES: This implementation uses a sparse field level set solver instead of the narrow band implementation described in the reference below, which may introduce some differences in how fast and how accurately (in terms of RMS error) the solution converges.

REFERENCES: Whitaker, Ross. "Reducing Aliasing Artifacts In Iso-Surfaces of Binary Volumes" IEEE Volume Visualization and Graphics Symposium, October 2000, pp.23-32.

PARAMETERS: The MaximumRMSChange parameter is used to determine when the solution has converged. A lower value will result in a tighter-fitting solution, but will require more computations. Too low a value could put the solver into an infinite loop. Values should always be less than 1.0. A value of 0.07 is a good starting estimate.

: The MaximumIterations parameter can be used to halt the solution after a specified number of iterations.

INPUT: The input is an N-dimensional image of any type. It is assumed to be a binary image. The filter will use an isosurface value that is halfway between the min and max values in the image. A signed data type is *not* necessary for the input.

OUTPUT: The filter will output a level set image of real, signed values. The zero crossings of this (N-dimensional) image represent the position of the isosurface value of interest. Values outside the zero level set are negative and values inside the zero level set are positive values.

IMPORTANT!: The output image type you use to instantiate this filter should be a real valued scalar type. In other words: doubles or floats.

USING THIS FILTER: The filter is relatively straightforward to use. Tests and examples exist to illustrate. The important thing is to understand the input and output types so you can properly interperet your results.

: In the common case, the only parameter that will need to be set is the MaximumRMSChange parameter, which determines when the solver halts.

Wiki Examples:

All Examples

Anti alias a binary image

Definition at line 107 of file itkAntiAliasBinaryImageFilter.h.