Demonstrate Available Threshold Algorithms

Synopsis

Demonstrate available threshold algorithms.

Results

Input Image

Output In QuickView

Code

C++

#include "itkLiThresholdImageFilter.h"
#include "itkHuangThresholdImageFilter.h"
#include "itkIntermodesThresholdImageFilter.h"
#include "itkIsoDataThresholdImageFilter.h"
#include "itkKittlerIllingworthThresholdImageFilter.h"
#include "itkMaximumEntropyThresholdImageFilter.h"
#include "itkMomentsThresholdImageFilter.h"
#include "itkOtsuThresholdImageFilter.h"
#include "itkRenyiEntropyThresholdImageFilter.h"
#include "itkShanbhagThresholdImageFilter.h"
#include "itkTriangleThresholdImageFilter.h"
#include "itkYenThresholdImageFilter.h"

#include "itkImageFileReader.h"
#include "itkImageFileWriter.h"

#include "itksys/SystemTools.hxx"
#include <sstream>
#include <map>
#ifdef ENABLE_QUICKVIEW
#  include "QuickView.h"
#endif

int
main(int argc, char * argv[])
{
  if (argc < 2)
  {
    std::cout << "Usage: " << argv[0];
    std::cout << " inputImageFile";
    std::cerr << std::endl;
    return EXIT_FAILURE;
  }

  using InputPixelType = short;
  using OutputPixelType = unsigned char;

  using InputImageType = itk::Image<InputPixelType, 2>;
  using OutputImageType = itk::Image<OutputPixelType, 2>;

  using LiFilterType = itk::LiThresholdImageFilter<InputImageType, OutputImageType>;
  using HuangFilterType = itk::HuangThresholdImageFilter<InputImageType, OutputImageType>;
  using IntermodesFilterType = itk::IntermodesThresholdImageFilter<InputImageType, OutputImageType>;
  using IsoDataFilterType = itk::IsoDataThresholdImageFilter<InputImageType, OutputImageType>;
  // TODO: Fix KittlerIllingworth Filter
  // using KittlerIllingworthFilterType = itk::KittlerIllingworthThresholdImageFilter<InputImageType, OutputImageType>;
  using LiFilterType = itk::LiThresholdImageFilter<InputImageType, OutputImageType>;
  using MaximumEntropyFilterType = itk::MaximumEntropyThresholdImageFilter<InputImageType, OutputImageType>;
  using MomentsFilterType = itk::MomentsThresholdImageFilter<InputImageType, OutputImageType>;
  using OtsuFilterType = itk::OtsuThresholdImageFilter<InputImageType, OutputImageType>;
  using RenyiEntropyFilterType = itk::RenyiEntropyThresholdImageFilter<InputImageType, OutputImageType>;
  using ShanbhagFilterType = itk::ShanbhagThresholdImageFilter<InputImageType, OutputImageType>;
  using TriangleFilterType = itk::TriangleThresholdImageFilter<InputImageType, OutputImageType>;
  using YenFilterType = itk::YenThresholdImageFilter<InputImageType, OutputImageType>;

  using ReaderType = itk::ImageFileReader<InputImageType>;

  ReaderType::Pointer reader = ReaderType::New();
  reader->SetFileName(argv[1]);

#ifdef ENABLE_QUICKVIEW
  QuickView viewer;
  viewer.AddImage(reader->GetOutput(), true, itksys::SystemTools::GetFilenameName(argv[1]));

  using FilterContainerType =
    std::map<std::string, itk::HistogramThresholdImageFilter<InputImageType, OutputImageType>::Pointer>;
  FilterContainerType filterContainer;

  filterContainer["Huang"] = HuangFilterType::New();
  filterContainer["Intermodes"] = IntermodesFilterType::New();
  filterContainer["IsoData"] = IsoDataFilterType::New();
  // filterContainer["KittlerIllingworth"] = KittlerIllingworthFilterType::New();
  filterContainer["Li"] = LiFilterType::New();
  filterContainer["MaximumEntropy"] = MaximumEntropyFilterType::New();
  filterContainer["Moments"] = MomentsFilterType::New();
  filterContainer["Otsu"] = OtsuFilterType::New();
  filterContainer["RenyiEntropy"] = RenyiEntropyFilterType::New();
  filterContainer["Shanbhag"] = ShanbhagFilterType::New();
  filterContainer["Triangle"] = TriangleFilterType::New();
  filterContainer["Yen"] = YenFilterType::New();

  auto it = filterContainer.begin();
  for (it = filterContainer.begin(); it != filterContainer.end(); ++it)
  {
    (*it).second->SetInsideValue(255);
    (*it).second->SetOutsideValue(0);
    (*it).second->SetInput(reader->GetOutput());
    (*it).second->SetNumberOfHistogramBins(100);
    (*it).second->Update();
    std::stringstream desc;
    desc << (*it).first << " threshold = " << (*it).second->GetThreshold();
    viewer.AddImage((*it).second->GetOutput(), true, desc.str());
  }


  viewer.Visualize();
#endif
  return EXIT_SUCCESS;
}

Classes demonstrated

template<typename TInputImage, typename TOutputImage, typename TMaskImage = TOutputImage>
class LiThresholdImageFilter : public itk::HistogramThresholdImageFilter<TInputImage, TOutputImage, TMaskImage>

Threshold an image using the Li Threshold.

This filter creates a binary thresholded image that separates an image into foreground and background components. The filter computes the threshold using the LiThresholdCalculator and applies that threshold to the input image using the BinaryThresholdImageFilter.

This implementation was taken from the Insight Journal paper:

https://www.insight-journal.org/browse/publication/811

Author

Richard Beare. Department of Medicine, Monash University, Melbourne, Australia.

Author

Gaetan Lehmann. Biologie du Developpement et de la Reproduction, INRA de Jouy-en-Josas, France.

See

HistogramThresholdImageFilter

ITK Sphinx Examples:

• All ITK Sphinx Examples

• Demonstrate Available Threshold Algorithms

See itk::LiThresholdImageFilter for additional documentation.