ITK  4.0.0
Insight Segmentation and Registration Toolkit
Classes | Public Types | Public Member Functions | Static Public Attributes | Protected Member Functions | Static Protected Member Functions | Private Member Functions | Private Attributes
itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType > Class Template Reference

This class implements a multi-threaded base class for Image to SparseImage finite difference processes. More...

#include <itkFiniteDifferenceSparseImageFilter.h>

Inheritance diagram for itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >:
Collaboration diagram for itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >:

List of all members.

Classes

struct  FDThreadStruct
struct  ThreadRegionType

Public Types

typedef SmartPointer< const SelfConstPointer
typedef
Superclass::FiniteDifferenceFunctionType 
FiniteDifferenceFunctionType
typedef
SparseOutputImageType::IndexType 
IndexType
typedef Superclass::InputImageType InputImageType
typedef
OutputNodeType::NodeDataType 
NodeDataType
typedef
SparseOutputImageType::NodeListType 
NodeListType
typedef
OutputNodeType::NodeValueType 
NodeValueType
typedef
SparseOutputImageType::NodeType 
OutputNodeType
typedef Superclass::PixelType PixelType
typedef SmartPointer< SelfPointer
typedef
FiniteDifferenceSparseImageFilter 
Self
typedef
SparseOutputImageType::SizeType 
SizeType
typedef
FiniteDifferenceSparseImageFunction
< SparseOutputImageType
SparseFunctionType
typedef Superclass::OutputImageType SparseOutputImageType
typedef
FiniteDifferenceImageFilter
< TInputImageType,
TSparseOutputImageType > 
Superclass
typedef Superclass::TimeStepType TimeStepType

Public Member Functions

virtual const char * GetNameOfClass () const
virtual bool GetPrecomputeFlag () const
virtual void SetPrecomputeFlag (bool _arg)
void SetSparseFunction (SparseFunctionType *sf)

Static Public Attributes

static const unsigned int ImageDimension = Superclass::ImageDimension

Protected Member Functions

virtual void AllocateUpdateBuffer ()
virtual void ApplyUpdate (const TimeStepType &dt)
virtual TimeStepType CalculateChange ()
virtual NodeDataType DataConstraint (const NodeDataType &data) const
 FiniteDifferenceSparseImageFilter ()
ThreadIdType GetSplitRegion (ThreadIdType i, ThreadIdType num, ThreadRegionType &splitRegion)
virtual void Initialize ()
virtual void PrecalculateChange ()
void PrintSelf (std::ostream &os, Indent indent) const
virtual void ThreadedApplyUpdate (const TimeStepType &dt, const ThreadRegionType &regionToProcess, ThreadIdType threadId)
virtual TimeStepType ThreadedCalculateChange (const ThreadRegionType &regionToProcess, ThreadIdType threadId)
virtual void ThreadedPrecalculateChange (const ThreadRegionType &regionToProcess, ThreadIdType threadId)
 ~FiniteDifferenceSparseImageFilter ()

Static Protected Member Functions

static ITK_THREAD_RETURN_TYPE ApplyUpdateThreaderCallback (void *arg)
static ITK_THREAD_RETURN_TYPE CalculateChangeThreaderCallback (void *arg)
static ITK_THREAD_RETURN_TYPE PrecalculateChangeThreaderCallback (void *arg)

Private Member Functions

 FiniteDifferenceSparseImageFilter (const Self &)
void operator= (const Self &)

Private Attributes

bool m_PrecomputeFlag
NodeListType::RegionListType m_RegionList
SparseFunctionTypem_SparseFunction

Detailed Description

template<class TInputImageType, class TSparseOutputImageType>
class itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >

This class implements a multi-threaded base class for Image to SparseImage finite difference processes.

This class implements a multi-threading mechanism for implementing finite difference PDE's on sparse image types. The sparse image is a image of pointers to node variables at valid pixel locations and null pointers at others. The node variable type must have the following members: m_Index, m_Data and m_Update.
This class also adds precomputing support to the finite difference image filter scheme. This support can be used by certain filters to speed up the processing. The m_PrecomputeFlag should be set to true to use this and the Function object must provide a PrecomputeSparseUpdate method.
INPUTS
The input to this filter is either a regular or sparse image. Subclasses should provide a way of copying this information to the output sparse image or initializing the output image nodes from the input image.
OUTPUTS
The output is a sparse image. The output will be in the m_Data members of the nodes of the sparse image.
IMPORTANT
The output sparse image type must be templated with a node type that at least has the following member variables: m_Index, m_Data and m_Update.

Definition at line 63 of file itkFiniteDifferenceSparseImageFilter.h.


Member Typedef Documentation

template<class TInputImageType, class TSparseOutputImageType>
typedef SmartPointer< const Self > itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::ConstPointer
template<class TInputImageType, class TSparseOutputImageType>
typedef Superclass::FiniteDifferenceFunctionType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::FiniteDifferenceFunctionType

The value type of the time step. This is distinct from PixelType because PixelType may often be a vector value, while the TimeStep is a scalar value.

Reimplemented from itk::FiniteDifferenceImageFilter< TInputImageType, TSparseOutputImageType >.

Reimplemented in itk::ImplicitManifoldNormalVectorFilter< TInputImage, TSparseOutputImage >.

Definition at line 84 of file itkFiniteDifferenceSparseImageFilter.h.

template<class TInputImageType, class TSparseOutputImageType>
typedef SparseOutputImageType::IndexType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::IndexType

Typedefs from the sparse output image type.

Reimplemented in itk::ImplicitManifoldNormalVectorFilter< TInputImage, TSparseOutputImage >.

Definition at line 92 of file itkFiniteDifferenceSparseImageFilter.h.

template<class TInputImageType, class TSparseOutputImageType>
typedef Superclass::InputImageType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::InputImageType
template<class TInputImageType, class TSparseOutputImageType>
typedef OutputNodeType::NodeDataType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::NodeDataType

The type for the data variable of OutputNodeType.

Definition at line 98 of file itkFiniteDifferenceSparseImageFilter.h.

template<class TInputImageType, class TSparseOutputImageType>
typedef SparseOutputImageType::NodeListType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::NodeListType
template<class TInputImageType, class TSparseOutputImageType>
typedef OutputNodeType::NodeValueType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::NodeValueType

The basic scalar variable type used in OutputNodeType. Expected to be float or double. If NodeDataType is a scalar, then this is the same type as that.

Reimplemented in itk::ImplicitManifoldNormalVectorFilter< TInputImage, TSparseOutputImage >.

Definition at line 103 of file itkFiniteDifferenceSparseImageFilter.h.

template<class TInputImageType, class TSparseOutputImageType>
typedef SparseOutputImageType::NodeType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::OutputNodeType

Definition at line 94 of file itkFiniteDifferenceSparseImageFilter.h.

template<class TInputImageType, class TSparseOutputImageType>
typedef Superclass::PixelType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::PixelType
template<class TInputImageType, class TSparseOutputImageType>
typedef SmartPointer< Self > itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::Pointer
template<class TInputImageType, class TSparseOutputImageType>
typedef FiniteDifferenceSparseImageFilter itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::Self
template<class TInputImageType, class TSparseOutputImageType>
typedef SparseOutputImageType::SizeType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::SizeType

Definition at line 93 of file itkFiniteDifferenceSparseImageFilter.h.

template<class TInputImageType, class TSparseOutputImageType>
typedef FiniteDifferenceSparseImageFunction< SparseOutputImageType > itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::SparseFunctionType

The sparse image finite difference function type used in this class.

Definition at line 107 of file itkFiniteDifferenceSparseImageFilter.h.

template<class TInputImageType, class TSparseOutputImageType>
typedef Superclass::OutputImageType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::SparseOutputImageType
template<class TInputImageType, class TSparseOutputImageType>
typedef FiniteDifferenceImageFilter< TInputImageType, TSparseOutputImageType > itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::Superclass
template<class TInputImageType, class TSparseOutputImageType>
typedef Superclass::TimeStepType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::TimeStepType

Constructor & Destructor Documentation

template<class TInputImageType, class TSparseOutputImageType>
itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::FiniteDifferenceSparseImageFilter ( ) [protected]
template<class TInputImageType, class TSparseOutputImageType>
itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::~FiniteDifferenceSparseImageFilter ( ) [inline, protected]

Definition at line 116 of file itkFiniteDifferenceSparseImageFilter.h.

template<class TInputImageType, class TSparseOutputImageType>
itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::FiniteDifferenceSparseImageFilter ( const Self ) [private]

Member Function Documentation

template<class TInputImageType, class TSparseOutputImageType>
virtual void itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::AllocateUpdateBuffer ( ) [inline, protected, virtual]

This class does not use AllocateUpdateBuffer to allocate memory for its narrow band. All memory is handled through the SparseImage class.

Implements itk::FiniteDifferenceImageFilter< TInputImageType, TSparseOutputImageType >.

Definition at line 126 of file itkFiniteDifferenceSparseImageFilter.h.

template<class TInputImageType, class TSparseOutputImageType>
virtual void itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::ApplyUpdate ( const TimeStepType dt) [protected, virtual]

This function updates the m_Data variable in the output image nodes using the update values computed by CalculateChange.

template<class TInputImageType, class TSparseOutputImageType>
static ITK_THREAD_RETURN_TYPE itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::ApplyUpdateThreaderCallback ( void *  arg) [static, protected]

Multi-threaded implementation of ApplyUpdate.

template<class TInputImageType, class TSparseOutputImageType>
virtual TimeStepType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::CalculateChange ( ) [protected, virtual]

This method computes changes to the output image using the ComputeSparseUpdate method in the Sparse Function object.

Implements itk::FiniteDifferenceImageFilter< TInputImageType, TSparseOutputImageType >.

template<class TInputImageType, class TSparseOutputImageType>
static ITK_THREAD_RETURN_TYPE itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::CalculateChangeThreaderCallback ( void *  arg) [static, protected]

Multuthreaded implementation of CalculateChange

template<class TInputImageType, class TSparseOutputImageType>
virtual NodeDataType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::DataConstraint ( const NodeDataType data) const [inline, protected, virtual]

This function can be used to implements constraints on the range of data values. Default is no constraint.

Definition at line 130 of file itkFiniteDifferenceSparseImageFilter.h.

template<class TInputImageType, class TSparseOutputImageType>
virtual const char* itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::GetNameOfClass ( ) const [virtual]
template<class TInputImageType, class TSparseOutputImageType>
virtual bool itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::GetPrecomputeFlag ( ) const [virtual]
template<class TInputImageType, class TSparseOutputImageType>
ThreadIdType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::GetSplitRegion ( ThreadIdType  i,
ThreadIdType  num,
ThreadRegionType splitRegion 
) [protected]

This function returns a single region for use in multi-threading.

template<class TInputImageType, class TSparseOutputImageType>
virtual void itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::Initialize ( ) [protected, virtual]

This method splits the active pixels of the sparse image into equal size lists for multi-threading. These lists remain constant throughout the operation of this filter.

Reimplemented from itk::FiniteDifferenceImageFilter< TInputImageType, TSparseOutputImageType >.

Reimplemented in itk::ImplicitManifoldNormalVectorFilter< TInputImage, TSparseOutputImage >.

template<class TInputImageType, class TSparseOutputImageType>
void itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::operator= ( const Self ) [private]

PushBackInput(), PushFronInput() in the public section force the input to be the type expected by an ImageToImageFilter. However, these methods end of "hiding" the versions from the superclass (ProcessObject) whose arguments are DataObjects. Here, we re-expose the versions from ProcessObject to avoid warnings about hiding methods from the superclass.

Reimplemented from itk::FiniteDifferenceImageFilter< TInputImageType, TSparseOutputImageType >.

Reimplemented in itk::ImplicitManifoldNormalVectorFilter< TInputImage, TSparseOutputImage >.

template<class TInputImageType, class TSparseOutputImageType>
virtual void itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::PrecalculateChange ( ) [protected, virtual]

This method provides a means of performing a first pass for computing the change and storing intermediate values that will then be used by CalculateChange. This can be used to speed up certain update rules.

template<class TInputImageType, class TSparseOutputImageType>
static ITK_THREAD_RETURN_TYPE itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::PrecalculateChangeThreaderCallback ( void *  arg) [static, protected]

Multi-threaded implementation of PrecalculateChange

template<class TInputImageType, class TSparseOutputImageType>
void itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::PrintSelf ( std::ostream &  os,
Indent  indent 
) const [protected, virtual]

Methods invoked by Print() to print information about the object including superclasses. Typically not called by the user (use Print() instead) but used in the hierarchical print process to combine the output of several classes.

Reimplemented from itk::FiniteDifferenceImageFilter< TInputImageType, TSparseOutputImageType >.

Reimplemented in itk::ImplicitManifoldNormalVectorFilter< TInputImage, TSparseOutputImage >.

template<class TInputImageType, class TSparseOutputImageType>
virtual void itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::SetPrecomputeFlag ( bool  _arg) [virtual]
template<class TInputImageType, class TSparseOutputImageType>
void itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::SetSparseFunction ( SparseFunctionType sf)

Sets the function object that will be called for computing updates.

template<class TInputImageType, class TSparseOutputImageType>
virtual void itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::ThreadedApplyUpdate ( const TimeStepType dt,
const ThreadRegionType regionToProcess,
ThreadIdType  threadId 
) [protected, virtual]
template<class TInputImageType, class TSparseOutputImageType>
virtual TimeStepType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::ThreadedCalculateChange ( const ThreadRegionType regionToProcess,
ThreadIdType  threadId 
) [protected, virtual]
template<class TInputImageType, class TSparseOutputImageType>
virtual void itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::ThreadedPrecalculateChange ( const ThreadRegionType regionToProcess,
ThreadIdType  threadId 
) [protected, virtual]

Member Data Documentation

template<class TInputImageType, class TSparseOutputImageType>
const unsigned int itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::ImageDimension = Superclass::ImageDimension [static]

Dimensionality of input and output data is assumed to be the same. It is inherited from the superclass.

Reimplemented from itk::FiniteDifferenceImageFilter< TInputImageType, TSparseOutputImageType >.

Reimplemented in itk::ImplicitManifoldNormalVectorFilter< TInputImage, TSparseOutputImage >.

Definition at line 89 of file itkFiniteDifferenceSparseImageFilter.h.

template<class TInputImageType, class TSparseOutputImageType>
bool itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::m_PrecomputeFlag [private]

Flag to let the class know whether or not to call PrecalculateChange.

Definition at line 189 of file itkFiniteDifferenceSparseImageFilter.h.

template<class TInputImageType, class TSparseOutputImageType>
NodeListType::RegionListType itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::m_RegionList [private]

A list of subregions of the active set of pixels in the sparse image which are passed to each thread for parallel processing.

Definition at line 196 of file itkFiniteDifferenceSparseImageFilter.h.

template<class TInputImageType, class TSparseOutputImageType>
SparseFunctionType* itk::FiniteDifferenceSparseImageFilter< TInputImageType, TSparseOutputImageType >::m_SparseFunction [private]

The Sparse function type.

Definition at line 192 of file itkFiniteDifferenceSparseImageFilter.h.


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