template<typename TInputImage, bool doDilate, typename TOutputImage = TInputImage>
class itk::ParabolicErodeDilateImageFilter< TInputImage, doDilate, TOutputImage >
Parent class for morphological operations with parabolic structuring elements.
Parabolic structuring elements are the morphological counterpart of the gaussian function in linear processing. Parabolic structuring elements are dimensionally decomposable and fast algorithms are available for computing erosions and dilations along lines. This class implements the "point of contact" algorithm and the "intersection" algorithm. The contact point algorithm is faster at small scales. The intersection algorithm was rediscovered by Felzenszwalb & Huttenlocher, but was actually described and tested by van den Boomgaard many years earlier. The intersection algorithm is faster for scales > 1, and independent of scale. It also seems to perform significantly better in the distance transform application. The intersection algorithm is also used in the IJ article on generalised distance transforms.
Parabolic structuring functions can be used as a fast alternative to the "rolling ball" structuring element classically used in background estimation, for example in ImageJ, have applications in image sharpening and distance transform computation.
This class uses an internal buffer of RealType pixels for each line. This line is cast to the output pixel type when written back to the output image. Since the filter uses dimensional decomposition this approach could result in inaccuracy as pixels are cast back and forth between low and high precision types. Use a high precision output type and cast manually if this is a problem.
Boomgaard, R. van den and Dorst, L. and Makram-Ebeid, L.S. and Schavemaker, J. Quadratic structuring functions in mathematical morphology. Mathematical Morphology and its Applications to Image and Signal Processing.
Felzenszwalb, P.F. & Huttenlocher, D.P. Distance Transforms of Sampled Functions. Techreport: Cornell Computing and Information Science, 2004.
This filter is threaded. Threading mechanism derived from SignedMaurerDistanceMap extensions by Gaetan Lehman
Core methods described in the InsightJournal article: "Morphology with parabolic structuring elements"
http://hdl.handle.net/1926/1370
- Author
- Richard Beare, Department of Medicine, Monash University, Australia. Richa.nosp@m.rd.B.nosp@m.eare@.nosp@m.mona.nosp@m.sh.ed.nosp@m.u
Definition at line 85 of file itkParabolicErodeDilateImageFilter.h.
|
| void | DynamicThreadedGenerateData (const OutputImageRegionType &outputRegionForThread) override |
| |
| void | EnlargeOutputRequestedRegion (DataObject *output) override |
| |
| void | GenerateData (void) override |
| |
| void | GenerateInputRequestedRegion () override throw ( InvalidRequestedRegionError ) |
| |
| unsigned int | SplitRequestedRegion (unsigned int i, unsigned int num, OutputImageRegionType &splitRegion) override |
| |
|
| | ParabolicErodeDilateImageFilter () |
| |
| | ~ParabolicErodeDilateImageFilter () override |
| |
| void | PrintSelf (std::ostream &os, Indent indent) const override |
| |
| virtual void | CallCopyInputRegionToOutputRegion (OutputImageRegionType &destRegion, const InputImageRegionType &srcRegion) |
| |
| virtual void | CallCopyOutputRegionToInputRegion (InputImageRegionType &destRegion, const OutputImageRegionType &srcRegion) |
| |
| | ImageToImageFilter () |
| |
| void | VerifyInputInformation () ITKv5_CONST override |
| |
| | ~ImageToImageFilter () override |
| |
| void | PushBackInput (const DataObject *input) override |
| |
| void | PushFrontInput (const DataObject *input) override |
| |
| virtual void | AfterThreadedGenerateData () |
| |
| virtual void | AllocateOutputs () |
| |
| virtual void | BeforeThreadedGenerateData () |
| |
| void | ClassicMultiThread (ThreadFunctionType callbackFunction) |
| |
virtual const
ImageRegionSplitterBase * | GetImageRegionSplitter () const |
| |
| | ImageSource () |
| |
| | ~ImageSource () override=default |
| |
| virtual void | ThreadedGenerateData (const OutputImageRegionType &outputRegionForThread, ThreadIdType threadId) |
| |
| virtual bool | GetDynamicMultiThreading () const |
| |
| virtual void | SetDynamicMultiThreading (bool _arg) |
| |
| virtual void | DynamicMultiThreadingOn () |
| |
| virtual void | DynamicMultiThreadingOff () |
| |
| virtual void | AddInput (DataObject *input) |
| |
| void | AddOptionalInputName (const DataObjectIdentifierType &) |
| |
| void | AddOptionalInputName (const DataObjectIdentifierType &, DataObjectPointerArraySizeType idx) |
| |
| virtual void | AddOutput (DataObject *output) |
| |
| bool | AddRequiredInputName (const DataObjectIdentifierType &) |
| |
| bool | AddRequiredInputName (const DataObjectIdentifierType &, DataObjectPointerArraySizeType idx) |
| |
| virtual void | CacheInputReleaseDataFlags () |
| |
| virtual void | GenerateOutputInformation () |
| |
| virtual void | GenerateOutputRequestedRegion (DataObject *output) |
| |
| DataObject * | GetInput (const DataObjectIdentifierType &key) |
| |
| const DataObject * | GetInput (const DataObjectIdentifierType &key) const |
| |
virtual const
DataObjectPointerArraySizeType & | GetNumberOfRequiredInputs () const |
| |
virtual const
DataObjectPointerArraySizeType & | GetNumberOfRequiredOutputs () const |
| |
| bool | IsIndexedInputName (const DataObjectIdentifierType &) const |
| |
| bool | IsIndexedOutputName (const DataObjectIdentifierType &) const |
| |
| bool | IsRequiredInputName (const DataObjectIdentifierType &) const |
| |
| DataObjectPointerArraySizeType | MakeIndexFromInputName (const DataObjectIdentifierType &name) const |
| |
| DataObjectPointerArraySizeType | MakeIndexFromOutputName (const DataObjectIdentifierType &name) const |
| |
| DataObjectIdentifierType | MakeNameFromInputIndex (DataObjectPointerArraySizeType idx) const |
| |
| DataObjectIdentifierType | MakeNameFromOutputIndex (DataObjectPointerArraySizeType idx) const |
| |
| | ProcessObject () |
| |
| virtual void | PropagateResetPipeline () |
| |
| virtual void | ReleaseInputs () |
| |
| virtual void | RemoveInput (const DataObjectIdentifierType &key) |
| |
| virtual void | RemoveInput (DataObjectPointerArraySizeType) |
| |
| virtual void | RemoveOutput (const DataObjectIdentifierType &key) |
| |
| virtual void | RemoveOutput (DataObjectPointerArraySizeType idx) |
| |
| bool | RemoveRequiredInputName (const DataObjectIdentifierType &) |
| |
| virtual void | RestoreInputReleaseDataFlags () |
| |
| virtual void | SetInput (const DataObjectIdentifierType &key, DataObject *input) |
| |
| virtual void | SetNthInput (DataObjectPointerArraySizeType num, DataObject *input) |
| |
| virtual void | SetNthOutput (DataObjectPointerArraySizeType num, DataObject *output) |
| |
| void | SetNumberOfIndexedInputs (DataObjectPointerArraySizeType num) |
| |
| void | SetNumberOfIndexedOutputs (DataObjectPointerArraySizeType num) |
| |
| virtual void | SetNumberOfRequiredInputs (DataObjectPointerArraySizeType) |
| |
| virtual void | SetNumberOfRequiredOutputs (DataObjectPointerArraySizeType _arg) |
| |
| virtual void | SetOutput (const DataObjectIdentifierType &key, DataObject *output) |
| |
| virtual void | SetPrimaryInput (DataObject *input) |
| |
| virtual void | SetPrimaryOutput (DataObject *output) |
| |
| void | SetRequiredInputNames (const NameArray &) |
| |
| virtual void | VerifyPreconditions () ITKv5_CONST |
| |
| | ~ProcessObject () override |
| |
| DataObject * | GetInput (DataObjectPointerArraySizeType idx) |
| |
| const DataObject * | GetInput (DataObjectPointerArraySizeType idx) const |
| |
| DataObject * | GetPrimaryInput () |
| |
| const DataObject * | GetPrimaryInput () const |
| |
| virtual void | SetPrimaryInputName (const DataObjectIdentifierType &key) |
| |
| virtual const char * | GetPrimaryInputName () const |
| |
| DataObject * | GetOutput (const DataObjectIdentifierType &key) |
| |
| const DataObject * | GetOutput (const DataObjectIdentifierType &key) const |
| |
| virtual void | SetPrimaryOutputName (const DataObjectIdentifierType &key) |
| |
| virtual const char * | GetPrimaryOutputName () const |
| |
| DataObject * | GetOutput (DataObjectPointerArraySizeType idx) |
| |
| const DataObject * | GetOutput (DataObjectPointerArraySizeType idx) const |
| |
| DataObject * | GetPrimaryOutput () |
| |
| const DataObject * | GetPrimaryOutput () const |
| |
| | Object () |
| |
| bool | PrintObservers (std::ostream &os, Indent indent) const |
| |
| virtual void | SetTimeStamp (const TimeStamp &time) |
| |
| | ~Object () override |
| |
| virtual LightObject::Pointer | InternalClone () const |
| |
| | LightObject () |
| |
| virtual void | PrintHeader (std::ostream &os, Indent indent) const |
| |
| virtual void | PrintTrailer (std::ostream &os, Indent indent) const |
| |
| virtual | ~LightObject () |
| |
template<typename TInputImage, bool doDilate, typename TOutputImage = TInputImage>
If an imaging filter can be implemented as a multithreaded algorithm, the filter will provide an implementation of ThreadedGenerateData() or DynamicThreadedGenerateData(). This superclass will automatically split the output image into a number of pieces, spawn multiple threads, and call (Dynamic)ThreadedGenerateData() in each thread. Prior to spawning threads, the BeforeThreadedGenerateData() method is called. After all the threads have completed, the AfterThreadedGenerateData() method is called. If an image processing filter cannot support threading, that filter should provide an implementation of the GenerateData() method instead of providing an implementation of (Dynamic)ThreadedGenerateData(). If a filter provides a GenerateData() method as its implementation, then the filter is responsible for allocating the output data. If a filter provides a (Dynamic)ThreadedGenerateData() method as its implementation, then the output memory will allocated automatically by this superclass. The (Dynamic)ThreadedGenerateData() method should only produce the output specified by "outputThreadRegion" parameter. (Dynamic)ThreadedGenerateData() cannot write to any other portion of the output image (as this is responsibility of a different thread).
DynamicThreadedGenerateData() is the newer variant without threadId, and is the preferred signature, which is called by default. This variant can split the requested region into different number of pieces depending on current multi-processing load, which allows better load balancing. The non-dynamic (also known as classic) ThreadedGenerateData() signature has threadId, and number of pieces to be split into is known in advance. It is activated by calling this->DynamicMultiThreadingOff(); in derived class constructor. It should be used when the multi-threaded algorithm needs to pre-allocate some data structure with size dependent on the number of pieces (also known as chunks, work units, and sometimes also incorrectly as threads). Only PlatformMultiThreader guarantees that each piece will be processed in its own specific thread. Pool and TBB multi-threaders maintain a pool of threads (normally equal to number of processing cores) which they use to process the pieces. This normally results in a single thread being reused to process multiple work units.
- See Also
- GenerateData(), SplitRequestedRegion()
Reimplemented from itk::ImageSource< TOutputImage >.
Inheritance diagram for itk::ParabolicErodeDilateImageFilter< TInputImage, doDilate, TOutputImage >:
1.8.5
