ITK  4.0.0
Insight Segmentation and Registration Toolkit
Public Types | Public Member Functions | Static Public Member Functions | Static Public Attributes | Protected Member Functions | Private Member Functions | Private Attributes
itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage > Class Template Reference

Morphological watershed transform from markers. More...

#include <itkMorphologicalWatershedFromMarkersImageFilter.h>

Inheritance diagram for itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >:
Collaboration diagram for itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >:

List of all members.

Public Types

typedef SmartPointer< const SelfConstPointer
typedef LabelImageType::IndexType IndexType
typedef
InputImageType::ConstPointer 
InputImageConstPointer
typedef InputImageType::PixelType InputImagePixelType
typedef InputImageType::Pointer InputImagePointer
typedef InputImageType::RegionType InputImageRegionType
typedef TInputImage InputImageType
typedef
LabelImageType::ConstPointer 
LabelImageConstPointer
typedef LabelImageType::PixelType LabelImagePixelType
typedef LabelImageType::Pointer LabelImagePointer
typedef LabelImageType::RegionType LabelImageRegionType
typedef TLabelImage LabelImageType
typedef SmartPointer< SelfPointer
typedef
MorphologicalWatershedFromMarkersImageFilter 
Self
typedef ImageToImageFilter
< TInputImage, TLabelImage > 
Superclass

Public Member Functions

virtual ::itk::LightObject::Pointer CreateAnother (void) const
const LabelImageTypeGetMarkerImage () const
virtual const char * GetNameOfClass () const
void SetInput1 (const TInputImage *input)
void SetInput2 (const TLabelImage *input)
void SetMarkerImage (const TLabelImage *input)
virtual void SetFullyConnected (bool _arg)
virtual const bool & GetFullyConnected ()
virtual void FullyConnectedOn ()
virtual void FullyConnectedOff ()

Static Public Member Functions

static Pointer New ()

Static Public Attributes

static const unsigned int ImageDimension = TInputImage::ImageDimension

Protected Member Functions

void EnlargeOutputRequestedRegion (DataObject *)
void GenerateData ()
void GenerateInputRequestedRegion ()

Private Member Functions

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

Private Attributes

bool m_FullyConnected
bool m_MarkWatershedLine
virtual void SetMarkWatershedLine (bool _arg)
virtual const bool & GetMarkWatershedLine ()
virtual void MarkWatershedLineOn ()
virtual void MarkWatershedLineOff ()
 MorphologicalWatershedFromMarkersImageFilter ()
 ~MorphologicalWatershedFromMarkersImageFilter ()
void PrintSelf (std::ostream &os, Indent indent) const

Detailed Description

template<class TInputImage, class TLabelImage>
class itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >

Morphological watershed transform from markers.

The watershed transform is a tool for image segmentation that is fast and flexible and potentially fairly parameter free. It was originally derived from a geophysical model of rain falling on a terrain and a variety of more formal definitions have been devised to allow development of practical algorithms. If an image is considered as a terrain and divided into catchment basins then the hope is that each catchment basin would contain an object of interest.

The output is a label image. A label image, sometimes referred to as a categorical image, has unique values for each region. For example, if a watershed produces 2 regions, all pixels belonging to one region would have value A, and all belonging to the other might have value B. Unassigned pixels, such as watershed lines, might have the background value (0 by convention).

The simplest way of using the watershed is to preprocess the image we want to segment so that the boundaries of our objects are bright (e.g apply an edge detector) and compute the watershed transform of the edge image. Watershed lines will correspond to the boundaries and our problem will be solved. This is rarely useful in practice because there are always more regional minima than there are objects, either due to noise or natural variations in the object surfaces. Therefore, while many watershed lines do lie on significant boundaries, there are many that don't. Various methods can be used to reduce the number of minima in the image, like thresholding the smallest values, filtering the minima and/or smoothing the image.

This filter use another approach to avoid the problem of over segmentation: it let the user provide a marker image which mark the minima in the input image and give them a label. The minima are imposed in the input image by the markers. The labels of the output image are the label of the marker image.

The morphological watershed transform algorithm is described in Chapter 9.2 of Pierre Soille's book "Morphological Image Analysis: Principles and Applications", Second Edition, Springer, 2003.

This code was contributed in the Insight Journal paper: "The watershed transform in ITK - discussion and new developments" by Beare R., Lehmann G. http://hdl.handle.net/1926/202 http://www.insight-journal.org/browse/publication/92

Author:
Gaetan Lehmann. Biologie du Developpement et de la Reproduction, INRA de Jouy-en-Josas, France.
Richard Beare. Department of Medicine, Monash University, Melbourne, Australia.
See also:
WatershedImageFilter, MorphologicalWatershedImageFilter

Definition at line 80 of file itkMorphologicalWatershedFromMarkersImageFilter.h.


Member Typedef Documentation

template<class TInputImage , class TLabelImage >
typedef SmartPointer< const Self > itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::ConstPointer
template<class TInputImage , class TLabelImage >
typedef LabelImageType::IndexType itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::IndexType
template<class TInputImage , class TLabelImage >
typedef InputImageType::ConstPointer itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::InputImageConstPointer
template<class TInputImage , class TLabelImage >
typedef InputImageType::PixelType itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::InputImagePixelType
template<class TInputImage , class TLabelImage >
typedef InputImageType::Pointer itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::InputImagePointer
template<class TInputImage , class TLabelImage >
typedef InputImageType::RegionType itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::InputImageRegionType
template<class TInputImage , class TLabelImage >
typedef TInputImage itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::InputImageType

Some convenient typedefs.

Reimplemented from itk::ImageToImageFilter< TInputImage, TLabelImage >.

Definition at line 91 of file itkMorphologicalWatershedFromMarkersImageFilter.h.

template<class TInputImage , class TLabelImage >
typedef LabelImageType::ConstPointer itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::LabelImageConstPointer
template<class TInputImage , class TLabelImage >
typedef LabelImageType::PixelType itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::LabelImagePixelType
template<class TInputImage , class TLabelImage >
typedef LabelImageType::Pointer itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::LabelImagePointer
template<class TInputImage , class TLabelImage >
typedef LabelImageType::RegionType itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::LabelImageRegionType
template<class TInputImage , class TLabelImage >
typedef TLabelImage itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::LabelImageType
template<class TInputImage , class TLabelImage >
typedef SmartPointer< Self > itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::Pointer
template<class TInputImage , class TLabelImage >
typedef MorphologicalWatershedFromMarkersImageFilter itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::Self

Standard class typedefs.

Reimplemented from itk::ImageToImageFilter< TInputImage, TLabelImage >.

Definition at line 85 of file itkMorphologicalWatershedFromMarkersImageFilter.h.

template<class TInputImage , class TLabelImage >
typedef ImageToImageFilter< TInputImage, TLabelImage > itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::Superclass

Constructor & Destructor Documentation

template<class TInputImage , class TLabelImage >
itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::MorphologicalWatershedFromMarkersImageFilter ( ) [protected]

Set/Get whether the watershed pixel must be marked or not. Default is true. Set it to false do not only avoid writing watershed pixels, it also decrease algorithm complexity.

template<class TInputImage , class TLabelImage >
itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::~MorphologicalWatershedFromMarkersImageFilter ( ) [inline, protected]

Set/Get whether the watershed pixel must be marked or not. Default is true. Set it to false do not only avoid writing watershed pixels, it also decrease algorithm complexity.

Definition at line 162 of file itkMorphologicalWatershedFromMarkersImageFilter.h.

template<class TInputImage , class TLabelImage >
itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::MorphologicalWatershedFromMarkersImageFilter ( const Self ) [private]

Member Function Documentation

template<class TInputImage , class TLabelImage >
virtual::itk::LightObject::Pointer itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::CreateAnother ( void  ) const [virtual]

Create an object from an instance, potentially deferring to a factory. This method allows you to create an instance of an object that is exactly the same type as the referring object. This is useful in cases where an object has been cast back to a base class.

Reimplemented from itk::Object.

template<class TInputImage , class TLabelImage >
void itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::EnlargeOutputRequestedRegion ( DataObject ) [protected, virtual]

This filter will enlarge the output requested region to produce all of the output.

See also:
ProcessObject::EnlargeOutputRequestedRegion()

Reimplemented from itk::ProcessObject.

template<class TInputImage , class TLabelImage >
virtual void itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::FullyConnectedOff ( ) [virtual]

Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. For objects that are 1 pixel wide, use FullyConnectedOn.

template<class TInputImage , class TLabelImage >
virtual void itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::FullyConnectedOn ( ) [virtual]

Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. For objects that are 1 pixel wide, use FullyConnectedOn.

template<class TInputImage , class TLabelImage >
void itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::GenerateData ( ) [protected, virtual]

The filter is single threaded.

Reimplemented from itk::ImageSource< TLabelImage >.

template<class TInputImage , class TLabelImage >
void itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::GenerateInputRequestedRegion ( ) [protected, virtual]

MorphologicalWatershedFromMarkersImageFilter needs to request the entire input images.

Reimplemented from itk::ImageToImageFilter< TInputImage, TLabelImage >.

template<class TInputImage , class TLabelImage >
virtual const bool& itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::GetFullyConnected ( ) [virtual]

Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. For objects that are 1 pixel wide, use FullyConnectedOn.

template<class TInputImage , class TLabelImage >
const LabelImageType* itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::GetMarkerImage ( ) const [inline]

Get the marker image

Definition at line 123 of file itkMorphologicalWatershedFromMarkersImageFilter.h.

References itk::ProcessObject::GetInput().

template<class TInputImage , class TLabelImage >
virtual const bool& itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::GetMarkWatershedLine ( ) [virtual]

Set/Get whether the watershed pixel must be marked or not. Default is true. Set it to false do not only avoid writing watershed pixels, it also decrease algorithm complexity.

template<class TInputImage , class TLabelImage >
virtual const char* itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::GetNameOfClass ( ) const [virtual]

Runtime information support.

Reimplemented from itk::ImageToImageFilter< TInputImage, TLabelImage >.

template<class TInputImage , class TLabelImage >
virtual void itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::MarkWatershedLineOff ( ) [virtual]

Set/Get whether the watershed pixel must be marked or not. Default is true. Set it to false do not only avoid writing watershed pixels, it also decrease algorithm complexity.

template<class TInputImage , class TLabelImage >
virtual void itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::MarkWatershedLineOn ( ) [virtual]

Set/Get whether the watershed pixel must be marked or not. Default is true. Set it to false do not only avoid writing watershed pixels, it also decrease algorithm complexity.

template<class TInputImage , class TLabelImage >
static Pointer itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::New ( ) [static]

Standard New method.

Reimplemented from itk::Object.

template<class TInputImage , class TLabelImage >
void itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::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::ImageToImageFilter< TInputImage, TLabelImage >.

template<class TInputImage , class TLabelImage >
void itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::PrintSelf ( std::ostream &  os,
Indent  indent 
) const [protected, virtual]

Set/Get whether the watershed pixel must be marked or not. Default is true. Set it to false do not only avoid writing watershed pixels, it also decrease algorithm complexity.

Reimplemented from itk::ImageToImageFilter< TInputImage, TLabelImage >.

template<class TInputImage , class TLabelImage >
virtual void itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::SetFullyConnected ( bool  _arg) [virtual]

Set/Get whether the connected components are defined strictly by face connectivity or by face+edge+vertex connectivity. Default is FullyConnectedOff. For objects that are 1 pixel wide, use FullyConnectedOn.

template<class TInputImage , class TLabelImage >
void itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::SetInput1 ( const TInputImage *  input) [inline]

Set the input image

Definition at line 130 of file itkMorphologicalWatershedFromMarkersImageFilter.h.

template<class TInputImage , class TLabelImage >
void itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::SetInput2 ( const TLabelImage *  input) [inline]

Set the marker image

Definition at line 136 of file itkMorphologicalWatershedFromMarkersImageFilter.h.

template<class TInputImage , class TLabelImage >
void itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::SetMarkerImage ( const TLabelImage *  input) [inline]

Set the marker image

Definition at line 116 of file itkMorphologicalWatershedFromMarkersImageFilter.h.

template<class TInputImage , class TLabelImage >
virtual void itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::SetMarkWatershedLine ( bool  _arg) [virtual]

Set/Get whether the watershed pixel must be marked or not. Default is true. Set it to false do not only avoid writing watershed pixels, it also decrease algorithm complexity.


Member Data Documentation

template<class TInputImage , class TLabelImage >
const unsigned int itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::ImageDimension = TInputImage::ImageDimension [static]

ImageDimension constants

Definition at line 106 of file itkMorphologicalWatershedFromMarkersImageFilter.h.

template<class TInputImage , class TLabelImage >
bool itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::m_FullyConnected [private]
template<class TInputImage , class TLabelImage >
bool itk::MorphologicalWatershedFromMarkersImageFilter< TInputImage, TLabelImage >::m_MarkWatershedLine [private]

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