ITK/ExamplesBoneyard/KdTreeBasedKMeansClustering 2D

Description

Cluster a collection of measurements using the KMeans algorithm. The name "KdTreeBased" indicates that this is an efficient implementation which uses a KdTree.

Output

The input is shown on the left. It consists of a single collection of 2D points that lend themselves to easy clustering into 2 clusters. The output clusters are shown on the right. Points belonging to the same cluster as shown in the same color.

KdTreeBasedKMeansClustering_2D.cxx

<source lang="cpp">

1. include "itkDecisionRule.h"
2. include "itkVector.h"
3. include "itkListSample.h"
4. include "itkKdTree.h"
5. include "itkWeightedCentroidKdTreeGenerator.h"
6. include "itkKdTreeBasedKmeansEstimator.h"
7. include "itkMinimumDecisionRule2.h"
8. include "itkEuclideanDistanceMetric.h"
9. include "itkDistanceToCentroidMembershipFunction.h"
10. include "itkSampleClassifierFilter.h"
11. include "itkNormalVariateGenerator.h"

1. include "vtkActor.h"
2. include "vtkInteractorStyleTrackballCamera.h"
3. include "vtkPolyData.h"
4. include "vtkPolyDataMapper.h"
5. include "vtkProperty.h"
6. include "vtkRenderer.h"
7. include "vtkRenderWindow.h"
8. include "vtkRenderWindowInteractor.h"
9. include "vtkSmartPointer.h"
10. include "vtkVertexGlyphFilter.h"

int main(int, char *[]) {

 typedef itk::Vector< double, 2 > MeasurementVectorType;
 typedef itk::Statistics::ListSample< MeasurementVectorType > SampleType;
 SampleType::Pointer sample = SampleType::New();

 typedef itk::Statistics::NormalVariateGenerator NormalGeneratorType;
 NormalGeneratorType::Pointer normalGenerator = NormalGeneratorType::New();

 normalGenerator->Initialize( 101 );

 MeasurementVectorType mv;
 double mean = 100;
 double standardDeviation = 30;
 for ( unsigned int i = 0 ; i < 100 ; ++i )
   {
   mv[0] = ( normalGenerator->GetVariate() * standardDeviation ) + mean;
   mv[1] = ( normalGenerator->GetVariate() * standardDeviation ) + mean;
   sample->PushBack( mv );
   }

 normalGenerator->Initialize( 3024 );
 mean = 200;
 standardDeviation = 30;
 for ( unsigned int i = 0 ; i < 100 ; ++i )
   {
   mv[0] = ( normalGenerator->GetVariate() * standardDeviation ) + mean;
   mv[1] = ( normalGenerator->GetVariate() * standardDeviation ) + mean;
   sample->PushBack( mv );
   }

 typedef itk::Statistics::WeightedCentroidKdTreeGenerator< SampleType >
   TreeGeneratorType;
 TreeGeneratorType::Pointer treeGenerator = TreeGeneratorType::New();

 treeGenerator->SetSample( sample );
 treeGenerator->SetBucketSize( 16 );
 treeGenerator->Update();

 typedef TreeGeneratorType::KdTreeType TreeType;
 typedef itk::Statistics::KdTreeBasedKmeansEstimator<TreeType> EstimatorType;
 EstimatorType::Pointer estimator = EstimatorType::New();

 EstimatorType::ParametersType initialMeans(4);
 initialMeans[0] = 0.0; // Cluster 1, mean[0]
 initialMeans[1] = 0.0; // Cluster 1, mean[1]
 initialMeans[2] = 5.0; // Cluster 2, mean[0]
 initialMeans[3] = 5.0; // Cluster 2, mean[1]

 estimator->SetParameters( initialMeans );
 estimator->SetKdTree( treeGenerator->GetOutput() );
 estimator->SetMaximumIteration( 200 );
 estimator->SetCentroidPositionChangesThreshold(0.0);
 estimator->StartOptimization();

 EstimatorType::ParametersType estimatedMeans = estimator->GetParameters();

 for ( unsigned int i = 0 ; i < 4 ; i+=2 )
   {
   std::cout << "cluster[" << i << "] " << std::endl;
   std::cout << "    estimated mean : " << estimatedMeans[i] << " , " << estimatedMeans[i+1] << std::endl;
   }

 typedef itk::Statistics::DistanceToCentroidMembershipFunction< MeasurementVectorType >
   MembershipFunctionType;
 typedef MembershipFunctionType::Pointer                      MembershipFunctionPointer;

 typedef itk::Statistics::MinimumDecisionRule2 DecisionRuleType;
 DecisionRuleType::Pointer decisionRule = DecisionRuleType::New();

 typedef itk::Statistics::SampleClassifierFilter< SampleType > ClassifierType;
 ClassifierType::Pointer classifier = ClassifierType::New();

 classifier->SetDecisionRule(decisionRule);
 classifier->SetInput( sample );
 classifier->SetNumberOfClasses( 2 );

 typedef ClassifierType::ClassLabelVectorObjectType               ClassLabelVectorObjectType;
 typedef ClassifierType::ClassLabelVectorType                     ClassLabelVectorType;
 typedef ClassifierType::MembershipFunctionVectorObjectType       MembershipFunctionVectorObjectType;
 typedef ClassifierType::MembershipFunctionVectorType             MembershipFunctionVectorType;

 ClassLabelVectorObjectType::Pointer  classLabelsObject = ClassLabelVectorObjectType::New();
 classifier->SetClassLabels( classLabelsObject );

 ClassLabelVectorType &  classLabelsVector = classLabelsObject->Get();
 classLabelsVector.push_back( 100 );
 classLabelsVector.push_back( 200 );

 MembershipFunctionVectorObjectType::Pointer membershipFunctionsObject =
   MembershipFunctionVectorObjectType::New();
 classifier->SetMembershipFunctions( membershipFunctionsObject );

 MembershipFunctionVectorType &  membershipFunctionsVector = membershipFunctionsObject->Get();

 MembershipFunctionType::CentroidType origin( sample->GetMeasurementVectorSize() );
 int index = 0;
 for ( unsigned int i = 0 ; i < 2 ; i++ )
   {
   MembershipFunctionPointer membershipFunction = MembershipFunctionType::New();
   for ( unsigned int j = 0 ; j < sample->GetMeasurementVectorSize(); j++ )
     {
     origin[j] = estimatedMeans[index++];
     }
   membershipFunction->SetCentroid( origin );
   membershipFunctionsVector.push_back( membershipFunction.GetPointer() );
   }

 classifier->Update();

 const ClassifierType::MembershipSampleType* membershipSample = classifier->GetOutput();
 ClassifierType::MembershipSampleType::ConstIterator iter = membershipSample->Begin();

 while ( iter != membershipSample->End() )
   {
   std::cout << "measurement vector = " << iter.GetMeasurementVector()
             << "class label = " << iter.GetClassLabel()
             << std::endl;
   ++iter;
   }

 // Visualize
 vtkSmartPointer<vtkPoints> points1 =
   vtkSmartPointer<vtkPoints>::New();
 vtkSmartPointer<vtkPoints> points2 =
   vtkSmartPointer<vtkPoints>::New();

 iter = membershipSample->Begin();
 while ( iter != membershipSample->End() )
   {
   if(iter.GetClassLabel() == 100)
     {
     points1->InsertNextPoint(iter.GetMeasurementVector()[0], iter.GetMeasurementVector()[1], 0);
     }
   else
     {
     points2->InsertNextPoint(iter.GetMeasurementVector()[0], iter.GetMeasurementVector()[1], 0);
     }
   ++iter;
   }

 vtkSmartPointer<vtkPolyData> polyData1 =
   vtkSmartPointer<vtkPolyData>::New();
 polyData1->SetPoints(points1);
 vtkSmartPointer<vtkVertexGlyphFilter> glyphFilter1 =
   vtkSmartPointer<vtkVertexGlyphFilter>::New();
 glyphFilter1->SetInputConnection(polyData1->GetProducerPort());
 glyphFilter1->Update();
 vtkSmartPointer<vtkPolyDataMapper> mapper1 =
   vtkSmartPointer<vtkPolyDataMapper>::New();
 mapper1->SetInputConnection(glyphFilter1->GetOutputPort());
 vtkSmartPointer<vtkActor> actor1 =
   vtkSmartPointer<vtkActor>::New();
 actor1->GetProperty()->SetColor(0,1,0);
 actor1->GetProperty()->SetPointSize(3);
 actor1->SetMapper(mapper1);

 vtkSmartPointer<vtkPolyData> polyData2 =
   vtkSmartPointer<vtkPolyData>::New();
 polyData2->SetPoints(points2);
 vtkSmartPointer<vtkVertexGlyphFilter> glyphFilter2 =
   vtkSmartPointer<vtkVertexGlyphFilter>::New();
 glyphFilter2->SetInputConnection(polyData2->GetProducerPort());
 glyphFilter2->Update();
 vtkSmartPointer<vtkPolyDataMapper> mapper2 =
   vtkSmartPointer<vtkPolyDataMapper>::New();
 mapper2->SetInputConnection(glyphFilter2->GetOutputPort());
 vtkSmartPointer<vtkActor> actor2 =
   vtkSmartPointer<vtkActor>::New();
 actor2->GetProperty()->SetColor(1,0,0);
 actor2->GetProperty()->SetPointSize(3);
 actor2->SetMapper(mapper2);

 vtkSmartPointer<vtkRenderWindow> renderWindow =
   vtkSmartPointer<vtkRenderWindow>::New();
 renderWindow->SetSize(300,300);

 vtkSmartPointer<vtkRenderer> renderer =
   vtkSmartPointer<vtkRenderer>::New();
 renderWindow->AddRenderer(renderer);

 renderer->AddActor(actor1);
 renderer->AddActor(actor2);
 renderer->ResetCamera();
 renderer->Render();

 vtkSmartPointer<vtkRenderWindowInteractor> renderWindowInteractor =
   vtkSmartPointer<vtkRenderWindowInteractor>::New();
 vtkSmartPointer<vtkInteractorStyleTrackballCamera> style =
   vtkSmartPointer<vtkInteractorStyleTrackballCamera>::New();

 renderWindowInteractor->SetInteractorStyle(style);

 renderWindowInteractor->SetRenderWindow(renderWindow);
 renderWindowInteractor->Initialize();

 renderWindowInteractor->Start();

 return EXIT_SUCCESS;

} </source>

CMakeLists.txt

<source lang="cmake"> cmake_minimum_required(VERSION 2.6)

PROJECT(KdTreeBasedKMeansClustering_2D)

FIND_PACKAGE(VTK REQUIRED) INCLUDE(${VTK_USE_FILE})

FIND_PACKAGE(ITK REQUIRED) INCLUDE(${ITK_USE_FILE})

ADD_EXECUTABLE(KdTreeBasedKMeansClustering_2D KdTreeBasedKMeansClustering_2D.cxx) TARGET_LINK_LIBRARIES(KdTreeBasedKMeansClustering_2D ITKBasicFilters ITKCommon ITKIO ITKStatistics vtkHybrid)

</source>