diff --git a/Examples/Markov/CMakeLists.txt b/Examples/Markov/CMakeLists.txt
index 70d1f24da89a09fec817ee6f5babb50e25f56173..c4d85b17a0c5be52b1fa5970b4d3a6c916d83a1a 100755
--- a/Examples/Markov/CMakeLists.txt
+++ b/Examples/Markov/CMakeLists.txt
@@ -3,6 +3,8 @@ INCLUDE_REGULAR_EXPRESSION("^.*$")
SET(Markov_EXAMPLES ${CXX_TEST_PATH}/otbMarkovExamplesTests)
+#ADD_EXECUTABLE(testRandom testRandom.cxx )
+#TARGET_LINK_LIBRARIES(testRandom OTBMarkov OTBCommon OTBIO ITKNumerics ITKIO)
ADD_EXECUTABLE(MarkovRandomField1Example MarkovRandomField1Example.cxx )
TARGET_LINK_LIBRARIES(MarkovRandomField1Example OTBMarkov OTBCommon OTBIO ITKNumerics ITKIO)
@@ -10,6 +12,9 @@ TARGET_LINK_LIBRARIES(MarkovRandomField1Example OTBMarkov OTBCommon OTBIO ITKNum
ADD_EXECUTABLE(MarkovRandomField2Example MarkovRandomField2Example.cxx )
TARGET_LINK_LIBRARIES(MarkovRandomField2Example OTBMarkov OTBCommon OTBIO ITKNumerics ITKIO)
+# ADD_EXECUTABLE(MarkovRandomField3Example MarkovRandomField3Example.cxx )
+# TARGET_LINK_LIBRARIES(MarkovRandomField3Example OTBMarkov OTBCommon OTBIO ITKNumerics ITKIO)
+
IF( NOT OTB_DISABLE_CXX_TESTING AND NOT OTB_DISABLE_CXX_EXAMPLES_TESTING )
SET(BASELINE ${OTB_DATA_ROOT}/Baseline/Examples/Markov)
@@ -28,13 +33,22 @@ ADD_TEST(MarkovRandomField1ExampleTest ${EXE_TESTS}
${TEMP}/MarkovRandomField1.png
MarkovRandomField1ExampleTest
${INPUTDATA}/QB_Suburb.png
- ${INPUTDATA}/RandomImage.png
${TEMP}/MarkovRandomField1.png
1.0
- 30
+ 20
1.0
)
+ADD_TEST(MarkovRandomField2ExampleTest ${EXE_TESTS}
+ --compare-image ${EPSILON}
+ ${BASELINE}/MarkovRandomField2.png
+ ${TEMP}/MarkovRandomField2.png
+ MarkovRandomField1ExampleTest
+ ${INPUTDATA}/QB_Suburb.png
+ ${TEMP}/MarkovRandomField2.png
+ 1.0
+ 5
+)
INCLUDE_DIRECTORIES("${OTBTesting_BINARY_DIR}")
diff --git a/Examples/Markov/MarkovRandomField1Example.cxx b/Examples/Markov/MarkovRandomField1Example.cxx
index 1cad19411a8c3ea51878b48f8272adc3458e6f88..b07634766ab2a9a708ef348b60bb529b9097c1bb 100644
--- a/Examples/Markov/MarkovRandomField1Example.cxx
+++ b/Examples/Markov/MarkovRandomField1Example.cxx
@@ -25,24 +25,25 @@
#endif
// Software Guide : BeginCommandLineArgs
-// INPUTS: {QB_Suburb.png}, {RandomImage.png}
-// OUTPUTS: {MarkovClassification1.png}
-// 1.0 30 1.0
+// INPUTS: {QB_Suburb.png}
+// OUTPUTS: {MarkovRandomField1.png}
+// 1.0 20 1.0
// Software Guide : EndCommandLineArgs
// Software Guide : BeginLatex
//
-// This example illustrates the details of the \doxygen{otb}{MarkovClassificationFilter}.
+// This example illustrates the details of the \doxygen{otb}{MarkovRandomFieldFilter}.
// This filter is an application of the Markov Random Fields for classification,
// segmentation or restauration.
//
-// This example applies the \doxygen{otb}{MarkovClassificationFilter} to
+// This example applies the \doxygen{otb}{MarkovRandomFieldFilter} to
// classify an image into four classes defined by their mean and variance. The
// optimization is done using an Metropolis algorithm with a random sampler. The
// regularization energy is defined by a Potts model and the fidelity by a
// Gaussian model.
//
+//
// Software Guide : EndLatex
#include "otbImageFileReader.h"
@@ -54,7 +55,7 @@
// Software Guide : BeginLatex
//
-// The first step towards the use of this filter is the inclusion of the proper
+// The first step toward the use of this filter is the inclusion of the proper
// header files.
//
// Software Guide : EndLatex
@@ -71,11 +72,12 @@
int main(int argc, char* argv[] )
{
- if( argc < 6 )
+ if( argc != 6 )
{
+ std::cerr << "Missing Parameters "<< argc << std::endl;
std::cerr << "Missing Parameters " << std::endl;
std::cerr << "Usage: " << argv[0];
- std::cerr << " inputImage inputInitialization output lambda iterations temperature" << std::endl;
+ std::cerr << " inputImage output lambda iterations temperature" << std::endl;
return 1;
}
@@ -84,7 +86,7 @@ int main(int argc, char* argv[] )
//
// Then we must decide what pixel type to use for the image. We
// choose to make all computations with double precision.
- // The labeled image is of type unsigned char to allow up to 256 different
+ // The labelled image is of type unsigned char which allows up to 256 different
// classes.
//
// Software Guide : EndLatex
@@ -112,19 +114,15 @@ int main(int argc, char* argv[] )
// Software Guide : BeginCodeSnippet
typedef otb::ImageFileReader< InputImageType > ReaderType;
- typedef otb::ImageFileReader< LabelledImageType > ReaderLabelledType;
typedef otb::ImageFileWriter< LabelledImageType > WriterType;
ReaderType::Pointer reader = ReaderType::New();
- ReaderLabelledType::Pointer reader2 = ReaderLabelledType::New();
WriterType::Pointer writer = WriterType::New();
const char * inputFilename = argv[1];
- const char * labelledFilename = argv[2];
- const char * outputFilename = argv[3];
+ const char * outputFilename = argv[2];
reader->SetFileName( inputFilename );
- reader2->SetFileName( labelledFilename );
writer->SetFileName( outputFilename );
// Software Guide : EndCodeSnippet
@@ -132,8 +130,8 @@ int main(int argc, char* argv[] )
// Software Guide : BeginLatex
//
// Finally, we define the different classes necessary for the Markov classification.
- // A \doxygen{otb}{MarkovClassificationFilter} is instanciated, this is the
- // main class which connects the others to do the Markov classification.
+ // A \doxygen{otb}{MarkovRandomFieldFilter} is instanciated, this is the
+ // main class which connect the other to do the Markov classification.
//
// Software Guide : EndLatex
@@ -149,7 +147,7 @@ int main(int argc, char* argv[] )
// An \doxygen{otb}{MRFSamplerRandomMAP}, which derives from the
// \doxygen{otb}{MRFSampler}, is instanciated. The sampler is in charge of
// proposing a modification for a given site. The
- // \doxygen{otb}{MRFSamplerRandomMAP}, randomly picks one possible value
+ // \doxygen{otb}{MRFSamplerRandomMAP}, randomly pick one possible value
// according to the MAP probability.
//
// Software Guide : EndLatex
@@ -178,9 +176,9 @@ int main(int argc, char* argv[] )
// Software Guide : BeginLatex
//
- // Two energies, deriving from the \doxygen{otb}{MRFEnergy} class needs to be instanciated. One energy
- // is required for the regularization, taking into account the relationship between neighborhing pixels
- // in the classified image. Here, this is done with the \doxygen{otb}{MRFEnergyPotts} which implements
+ // Two energy, deriving from the \doxygen{otb}{MRFEnergy} class need to be instanciated. One energy
+ // is required for the regularization, taking into account the relashionship between neighborhing pixels
+ // in the classified image. Here it is done with the \doxygen{otb}{MRFEnergyPotts} which implement
// a Potts model.
//
// The second energy is for the fidelity to the original data. Here it is done with an
@@ -216,7 +214,7 @@ int main(int argc, char* argv[] )
// Software Guide : BeginLatex
//
- // Variables for the \doxygen{otb}{MRFEnergyGaussianClassification} class, meand
+ // Parameter for the \doxygen{otb}{MRFEnergyGaussianClassification} class, meand
// and standard deviation are created.
//
// Software Guide : EndLatex
@@ -225,25 +223,6 @@ int main(int argc, char* argv[] )
// Software Guide : BeginCodeSnippet
unsigned int nClass = 4;
- itk::VariableLengthVector<double> mean;
- itk::VariableLengthVector<double> stdDev;
- mean.SetSize(nClass);
- stdDev.SetSize(nClass);
- mean[0]=10;mean[1]=80;mean[2]=150;mean[3]=220;
- stdDev[0]=10;stdDev[1]=10;stdDev[2]=10;stdDev[3]=10;
-
- // Software Guide : EndCodeSnippet
-
-
- // Software Guide : BeginLatex
- //
- // These parameters are passed to the different classs and the sampler, optimizer and
- // energies are connected with the Markov filter.
- //
- // Software Guide : EndLatex
-
- // Software Guide : BeginCodeSnippet
-
energyFidelity->SetNumberOfParameters(2*nClass);
EnergyFidelityType::ParametersType parameters;
parameters.SetSize(energyFidelity->GetNumberOfParameters());
@@ -256,18 +235,33 @@ int main(int argc, char* argv[] )
parameters[6]=220.0;//Class 3 mean
parameters[7]=10.0; //Class 3 stde
energyFidelity->SetParameters(parameters);
- OptimizerType::ParametersType paramOpt(1);
- paramOpt.Fill(atof(argv[6]));
- optimizer->SetParameters(paramOpt);
- markovFilter->SetValueInsteadRandom(500); // Unable rand() calculation
+
+ // Software Guide : EndCodeSnippet
+
+
+ // Software Guide : BeginLatex
+ //
+ // Parameters are given to the different class an the sampler, optimizer and
+ // energies are connected with the Markov filter.
+ //
+ // Software Guide : EndLatex
+
+ // Software Guide : BeginCodeSnippet
+
+
+ optimizer->SetParameters(atoi(argv[5]));
markovFilter->SetNumberOfClasses(nClass);
- markovFilter->SetMaximumNumberOfIterations(atoi(argv[5]));
- markovFilter->SetErrorTolerance(-1.0);
- markovFilter->SetLambda(atof(argv[4]));
+ markovFilter->SetMaximumNumberOfIterations(atoi(argv[4]));
+ markovFilter->SetErrorTolerance(0.0);
+ markovFilter->SetLambda(atof(argv[3]));
markovFilter->SetNeighborhoodRadius(1);
-
+ markovFilter->SetEnergyRegularization(static_cast<MarkovRandomFieldFilterType::EnergyRegularizationPointer>(energyRegularization));
+ markovFilter->SetEnergyFidelity(static_cast<MarkovRandomFieldFilterType::EnergyFidelityPointer>(energyFidelity));
+ /** Pas de finction concordante avec l'appel EnergyPotts*/
+ /*
markovFilter->SetEnergyRegularization(energyRegularization);
markovFilter->SetEnergyFidelity(energyFidelity);
+ */
markovFilter->SetOptimizer(optimizer);
markovFilter->SetSampler(sampler);
@@ -276,13 +270,12 @@ int main(int argc, char* argv[] )
// Software Guide : BeginLatex
//
// The pipeline is connected. An \doxygen{itk}{RescaleIntensityImageFilter}
- // rescales the classified image before saving it.
+ // rescale the classified image before saving it.
//
// Software Guide : EndLatex
// Software Guide : BeginCodeSnippet
- markovFilter->SetTrainingInput(reader2->GetOutput());
markovFilter->SetInput(reader->GetOutput());
typedef itk::RescaleIntensityImageFilter
@@ -299,7 +292,7 @@ int main(int argc, char* argv[] )
// Software Guide : BeginLatex
//
- // Finally, the pipeline execution is triggered.
+ // Finally, the pipeline execution is trigerred.
//
// Software Guide : EndLatex
@@ -308,20 +301,6 @@ int main(int argc, char* argv[] )
writer->Update();
// Software Guide : EndCodeSnippet
-
- // Software Guide : BeginLatex
-// Figure \ref{fig:MARKOVCLASS1} shows the result of the MRF classification.
-// \begin{figure}
-// \center
-// \includegraphics[width=0.45\textwidth]{QB_Suburb.eps}
-// \includegraphics[width=0.45\textwidth]{MarkovClassification1.eps}
-// \itkcaption[Markov Random Field Image Classification]{Result of the MRF
-// classification using a Metropolis algorithm with a random sampler
-// and a Potts regularization energy and a Gaussian fidelity. Left: Original image. Right: image of classes.}
-// \label{fig:MARKOVCLASS1}
-// \end{figure}
-// Software Guide : EndLatex
-
return 0;
diff --git a/Examples/Markov/MarkovRandomField2Example.cxx b/Examples/Markov/MarkovRandomField2Example.cxx
index bc7e1c8ed388ce686c40bd43791056b32c0882d6..395ef409df926dd8ca1120ecd0f07a4ff0a652d0 100644
--- a/Examples/Markov/MarkovRandomField2Example.cxx
+++ b/Examples/Markov/MarkovRandomField2Example.cxx
@@ -25,23 +25,20 @@
#endif
// Software Guide : BeginCommandLineArgs
-// INPUTS: {QB_Suburb.png RandomImage.png}
-// OUTPUTS: {MarkovClassification2.png}
-// 1.0 30 1.0
+// INPUTS: {QB_Suburb.png}
+// OUTPUTS: {MarkovRandomField1.png}
+// 1.0 5
// Software Guide : EndCommandLineArgs
// Software Guide : BeginLatex
//
-// This example illustrates the details of the \doxygen{otb}{MarkovClassificationFilter}.
-// This filter is an application of the Markov Random Fields for classification,
-// segmentation or restauration.
-//
-// This example applies the \doxygen{otb}{MarkovClassificationFilter} to
-// classify an image into four classes defined by their mean and variance. The
-// optimization is done using an ICM algorithm with a MAP estimator. The
-// regularization energy is defined by a Potts model and the fidelity by a
-// Gaussian model.
+// Using a similar structure as the previous program and the same energy
+// function, we are now going to slightly alter the program to use a
+// different sampler and optimizer. The proposed sample is proposed
+// randomly according to the MAP probability and the optimizer is the
+// ICM which accept the proposed sample if it enable a reduction of
+// the energy.
//
// Software Guide : EndLatex
@@ -54,47 +51,31 @@
// Software Guide : BeginLatex
//
-// The first step toward the use of this filter is the inclusion of the proper
-// header files.
+// First, we need to include header specific to these class:
//
// Software Guide : EndLatex
-// Software Guide : BeginCodeSnippet
+
#include "otbMRFEnergy.h"
#include "otbMRFEnergyPotts.h"
#include "otbMRFEnergyGaussianClassification.h"
-// #include "otbMRFEnergyPotts.h"
-// #include "otbMRFEnergyGaussianClassification.h"
-// #include "otbMRFOptimizerICM.h"
-// #include "otbMRFSamplerMAP.h"
+
+// Software Guide : BeginCodeSnippet
#include "otbMRFSamplerRandomMAP.h"
#include "otbMRFOptimizerICM.h"
-// #include "otbMRFSamplerRandom.h"
// Software Guide : EndCodeSnippet
-
+#include "otbMRFSamplerRandom.h"
int main(int argc, char* argv[] )
{
- if( argc < 6 )
+ if( argc != 5 )
{
std::cerr << "Missing Parameters " << std::endl;
std::cerr << "Usage: " << argv[0];
- std::cerr << " inputImage inputInitialization output lambda iterations temperature" << std::endl;
+ std::cerr << " inputImage output lambda iterations" << std::endl;
return 1;
}
-
-
- // Software Guide : BeginLatex
- //
- // Then we must decide what pixel type to use for the image. We
- // choose to make all computations with double precision.
- // The labelled image is of type unsigned char to allow up to 256 different
- // classes.
- //
- // Software Guide : EndLatex
-
- // Software Guide : BeginCodeSnippet
const unsigned int Dimension = 2;
@@ -103,78 +84,35 @@ int main(int argc, char* argv[] )
typedef otb::Image<InternalPixelType, Dimension> InputImageType;
typedef otb::Image<LabelledPixelType, Dimension> LabelledImageType;
- // Software Guide : EndCodeSnippet
-
-
- // Software Guide : BeginLatex
- //
- // We define a reader for the image to be classified, an initialisation for the
- // classification (which could be random) and a writer for the final
- // classification.
- //
- // Software Guide : EndLatex
-
- // Software Guide : BeginCodeSnippet
-
typedef otb::ImageFileReader< InputImageType > ReaderType;
- typedef otb::ImageFileReader< LabelledImageType > ReaderLabelledType;
typedef otb::ImageFileWriter< LabelledImageType > WriterType;
ReaderType::Pointer reader = ReaderType::New();
- ReaderLabelledType::Pointer reader2 = ReaderLabelledType::New();
WriterType::Pointer writer = WriterType::New();
const char * inputFilename = argv[1];
- const char * labelledFilename = argv[2];
- const char * outputFilename = argv[3];
+ const char * outputFilename = argv[2];
reader->SetFileName( inputFilename );
- reader2->SetFileName( labelledFilename );
writer->SetFileName( outputFilename );
- // Software Guide : EndCodeSnippet
-
- // Software Guide : BeginLatex
- //
- // Finally, we define the different classes necessary for the Markov classification.
- // A \doxygen{otb}{MarkovClassificationFilter} is instanciated, this is the
- // main class which connect the other to do the Markov classification.
- //
- // Software Guide : EndLatex
-
- // Software Guide : BeginCodeSnippet
-
typedef otb::MarkovRandomFieldFilter
<InputImageType,LabelledImageType> MarkovRandomFieldFilterType;
- // Software Guide : EndCodeSnippet
// Software Guide : BeginLatex
//
- // An \doxygen{otb}{MRFSamplerRandomMAP}, which derives from the
- // \doxygen{otb}{MRFSampler}, is instanciated. The sampler is in charge of
- // proposing a modification for a given site. The
- // \doxygen{otb}{MRFSamplerRandomMAP}, randomly pick one possible value
- // according to the MAP probability.
+ // And to declare these new type:
//
// Software Guide : EndLatex
// Software Guide : BeginCodeSnippet
typedef otb::MRFSamplerRandomMAP< InputImageType, LabelledImageType> SamplerType;
+// typedef otb::MRFSamplerRandom< InputImageType, LabelledImageType> SamplerType;
-
// Software Guide : EndCodeSnippet
- // Software Guide : BeginLatex
- //
- // An \doxygen{otb}{MRFOptimizerMetropoli}, which derives from the
- // \doxygen{otb}{MRFOptimizer}, is instanciated. The optimizer is in charge
- // of accepting or rejecting the value proposed by the sampler. The
- // \doxygen{otb}{MRFSamplerRandomMAP}, accept the proposal according to the
- // variation of energy it causes and a temperature parameter.
- //
- // Software Guide : EndLatex
// Software Guide : BeginCodeSnippet
@@ -182,74 +120,18 @@ int main(int argc, char* argv[] )
// Software Guide : EndCodeSnippet
- // Software Guide : BeginLatex
- //
- // Two energy, deriving from the \doxygen{otb}{MRFEnergy} class need to be instanciated. One energy
- // is required for the regularization, taking into account the relashionship between neighborhing pixels
- // in the classified image. Here it is done with the \doxygen{otb}{MRFEnergyPotts} which implement
- // a Potts model.
- //
- // The second energy is for the fidelity to the original data. Here it is done with an
- // \doxygen{otb}{MRFEnergyGaussianClassification} class, which defines a gaussian model for the data.
- //
- // Software Guide : EndLatex
-
- // Software Guide : BeginCodeSnippet
-
typedef otb::MRFEnergyPotts
<LabelledImageType, LabelledImageType> EnergyRegularizationType;
typedef otb::MRFEnergyGaussianClassification
<InputImageType, LabelledImageType> EnergyFidelityType;
- // Software Guide : EndCodeSnippet
-
- // Software Guide : BeginLatex
- //
- // The different filters composing our pipeline are created by invoking their
- // \code{New()} methods, assigning the results to smart pointers.
- //
- // Software Guide : EndLatex
-
- // Software Guide : BeginCodeSnippet
-
MarkovRandomFieldFilterType::Pointer markovFilter = MarkovRandomFieldFilterType::New();
EnergyRegularizationType::Pointer energyRegularization = EnergyRegularizationType::New();
EnergyFidelityType::Pointer energyFidelity = EnergyFidelityType::New();
OptimizerType::Pointer optimizer = OptimizerType::New();
SamplerType::Pointer sampler = SamplerType::New();
- // Software Guide : EndCodeSnippet
-
- // Software Guide : BeginLatex
- //
- // Parameter for the \doxygen{otb}{MRFEnergyGaussianClassification} class, meand
- // and standard deviation are created.
- //
- // Software Guide : EndLatex
-
-
- // Software Guide : BeginCodeSnippet
-
unsigned int nClass = 4;
- itk::VariableLengthVector<double> mean;
- itk::VariableLengthVector<double> stdDev;
- mean.SetSize(nClass);
- stdDev.SetSize(nClass);
- mean[0]=10;mean[1]=80;mean[2]=150;mean[3]=220;
- stdDev[0]=10;stdDev[1]=10;stdDev[2]=10;stdDev[3]=10;
-
- // Software Guide : EndCodeSnippet
-
-
- // Software Guide : BeginLatex
- //
- // Parameters are given to the different class an the sampler, optimizer and
- // energies are connected with the Markov filter.
- //
- // Software Guide : EndLatex
-
- // Software Guide : BeginCodeSnippet
-
energyFidelity->SetNumberOfParameters(2*nClass);
EnergyFidelityType::ParametersType parameters;
parameters.SetSize(energyFidelity->GetNumberOfParameters());
@@ -262,30 +144,26 @@ int main(int argc, char* argv[] )
parameters[6]=220.0;//Class 3 mean
parameters[7]=10.0; //Class 3 stde
energyFidelity->SetParameters(parameters);
-
- markovFilter->SetNumberOfClasses(nClass);
- markovFilter->SetMaximumNumberOfIterations(atoi(argv[5]));
- markovFilter->SetErrorTolerance(-1.0);
- markovFilter->SetLambda(atof(argv[4]));
- markovFilter->SetNeighborhoodRadius(1);
-
- markovFilter->SetEnergyRegularization(energyRegularization);
- markovFilter->SetEnergyFidelity(energyFidelity);
- markovFilter->SetOptimizer(optimizer);
- markovFilter->SetSampler(sampler);
- // Software Guide : EndCodeSnippet
// Software Guide : BeginLatex
//
- // The pipeline is connected. An \doxygen{itk}{RescaleIntensityImageFilter}
- // rescale the classified image before saving it.
+ // As the \doxygen{otb}{MRFOptimizerICM} does not have any parameters,
+ // the call to \code{optimizer->SetParameters()} must be removed
//
// Software Guide : EndLatex
- // Software Guide : BeginCodeSnippet
+ markovFilter->SetNumberOfClasses(nClass);
+ markovFilter->SetMaximumNumberOfIterations(atoi(argv[4]));
+ markovFilter->SetErrorTolerance(0.0);
+ markovFilter->SetLambda(atof(argv[3]));
+ markovFilter->SetNeighborhoodRadius(1);
+
+ markovFilter->SetEnergyRegularization(static_cast<MarkovRandomFieldFilterType::EnergyRegularizationPointer>(energyRegularization));
+ markovFilter->SetEnergyFidelity(static_cast<MarkovRandomFieldFilterType::EnergyFidelityPointer>(energyFidelity));
+ markovFilter->SetOptimizer(optimizer);
+ markovFilter->SetSampler(sampler);
- markovFilter->SetTrainingInput(reader2->GetOutput());
markovFilter->SetInput(reader->GetOutput());
typedef itk::RescaleIntensityImageFilter
@@ -298,20 +176,14 @@ int main(int argc, char* argv[] )
writer->SetInput( rescaleFilter->GetOutput() );
- // Software Guide : EndCodeSnippet
+ writer->Update();
// Software Guide : BeginLatex
//
- // Finally, the pipeline execution is trigerred.
+ // Apart from these, no further modification is required.
//
// Software Guide : EndLatex
- // Software Guide : BeginCodeSnippet
-
- writer->Update();
-
- // Software Guide : EndCodeSnippet
-
return 0;
}
diff --git a/Examples/Markov/MarkovRandomField3Example.cxx b/Examples/Markov/MarkovRandomField3Example.cxx
new file mode 100755
index 0000000000000000000000000000000000000000..6ec7d6bfac84581a799cb09af3525f5f8d040ee6
--- /dev/null
+++ b/Examples/Markov/MarkovRandomField3Example.cxx
@@ -0,0 +1,310 @@
+
+/*=========================================================================
+
+ Program: ORFEO Toolbox
+ Language: C++
+ Date: $Date$
+ Version: $Revision$
+
+
+ Copyright (c) Centre National d'Etudes Spatiales. All rights reserved.
+ See OTBCopyright.txt for details.
+
+
+ This software is distributed WITHOUT ANY WARRANTY; without even
+ the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ PURPOSE. See the above copyright notices for more information.
+
+=========================================================================*/
+#if defined(_MSC_VER)
+#pragma warning ( disable : 4786 )
+#endif
+
+#ifdef __BORLANDC__
+#define ITK_LEAN_AND_MEAN
+#endif
+
+// Software Guide : BeginCommandLineArgs
+// INPUTS: {QB_Suburb.png RandomImage.png}
+// OUTPUTS: {MarkovRandomField1.png}
+// 1.0 30 1.0
+// Software Guide : EndCommandLineArgs
+
+
+// Software Guide : BeginLatex
+//
+// This example illustrates the details of the \doxygen{otb}{MarkovRandomFieldFilter}.
+// This filter is an application of the Markov Random Fields for classification,
+// segmentation or restauration.
+//
+// This example applies the \doxygen{otb}{MarkovRandomFieldFilter} to
+// classify an image into four classes defined by their mean and variance. The
+// optimization is done using an Metropolis algorithm with a random sampler. The
+// regularization energy is defined by a Potts model and the fidelity by a
+// Gaussian model.
+//
+// The starting state of classification is given by the image proposed as a second
+// parameter. It could be a random image.
+//
+// Software Guide : EndLatex
+
+#include "otbImageFileReader.h"
+#include "otbImageFileWriter.h"
+#include "otbImage.h"
+#include "otbMarkovRandomFieldFilter.h"
+#include "itkRescaleIntensityImageFilter.h"
+
+
+// Software Guide : BeginLatex
+//
+// The first step toward the use of this filter is the inclusion of the proper
+// header files.
+//
+// Software Guide : EndLatex
+
+// Software Guide : BeginCodeSnippet
+#include "otbMRFEnergy.h"
+#include "otbMRFEnergyPotts.h"
+#include "otbMRFEnergyGaussianClassification.h"
+#include "otbMRFOptimizerMetropolis.h"
+#include "otbMRFSamplerRandom.h"
+// Software Guide : EndCodeSnippet
+
+
+int main(int argc, char* argv[] )
+{
+
+ if( argc != 7 )
+ {
+ std::cerr << "Missing Parameters " << std::endl;
+ std::cerr << "Usage: " << argv[0];
+ std::cerr << " inputImage inputInitialization output lambda iterations temperature" << std::endl;
+ return 1;
+ }
+
+
+ // Software Guide : BeginLatex
+ //
+ // Then we must decide what pixel type to use for the image. We
+ // choose to make all computations with double precision.
+ // The labelled image is of type unsigned char to allow up to 256 different
+ // classes.
+ //
+ // Software Guide : EndLatex
+
+ // Software Guide : BeginCodeSnippet
+
+ const unsigned int Dimension = 2;
+
+ typedef double InternalPixelType;
+ typedef unsigned char LabelledPixelType;
+ typedef otb::Image<InternalPixelType, Dimension> InputImageType;
+ typedef otb::Image<LabelledPixelType, Dimension> LabelledImageType;
+
+ // Software Guide : EndCodeSnippet
+
+
+ // Software Guide : BeginLatex
+ //
+ // We define a reader for the image to be classified, an initialisation for the
+ // classification (which could be random) and a writer for the final
+ // classification.
+ //
+ // Software Guide : EndLatex
+
+ // Software Guide : BeginCodeSnippet
+
+ typedef otb::ImageFileReader< InputImageType > ReaderType;
+ typedef otb::ImageFileReader< LabelledImageType > ReaderLabelledType;
+ typedef otb::ImageFileWriter< LabelledImageType > WriterType;
+
+ ReaderType::Pointer reader = ReaderType::New();
+ ReaderLabelledType::Pointer reader2 = ReaderLabelledType::New();
+ WriterType::Pointer writer = WriterType::New();
+
+ const char * inputFilename = argv[1];
+ const char * labelledFilename = argv[2];
+ const char * outputFilename = argv[3];
+
+ reader->SetFileName( inputFilename );
+ reader2->SetFileName( labelledFilename );
+ writer->SetFileName( outputFilename );
+
+ // Software Guide : EndCodeSnippet
+
+ // Software Guide : BeginLatex
+ //
+ // Finally, we define the different classes necessary for the Markov classification.
+ // A \doxygen{otb}{MarkovRandomFieldFilter} is instanciated, this is the
+ // main class which connect the other to do the Markov classification.
+ //
+ // Software Guide : EndLatex
+
+ // Software Guide : BeginCodeSnippet
+
+ typedef otb::MarkovRandomFieldFilter
+ <InputImageType,LabelledImageType> MarkovRandomFieldFilterType;
+
+ // Software Guide : EndCodeSnippet
+
+ // Software Guide : BeginLatex
+ //
+ // An \doxygen{otb}{MRFSamplerRandomMAP}, which derives from the
+ // \doxygen{otb}{MRFSampler}, is instanciated. The sampler is in charge of
+ // proposing a modification for a given site. The
+ // \doxygen{otb}{MRFSamplerRandomMAP}, randomly pick one possible value
+ // according to the MAP probability.
+ //
+ // Software Guide : EndLatex
+
+ // Software Guide : BeginCodeSnippet
+
+ typedef otb::MRFSamplerRandom< InputImageType, LabelledImageType> SamplerType;
+
+ // Software Guide : EndCodeSnippet
+
+ // Software Guide : BeginLatex
+ //
+ // An \doxygen{otb}{MRFOptimizerMetropoli}, which derives from the
+ // \doxygen{otb}{MRFOptimizer}, is instanciated. The optimizer is in charge
+ // of accepting or rejecting the value proposed by the sampler. The
+ // \doxygen{otb}{MRFSamplerRandomMAP}, accept the proposal according to the
+ // variation of energy it causes and a temperature parameter.
+ //
+ // Software Guide : EndLatex
+
+ // Software Guide : BeginCodeSnippet
+
+ typedef otb::MRFOptimizerMetropolis OptimizerType;
+
+ // Software Guide : EndCodeSnippet
+
+ // Software Guide : BeginLatex
+ //
+ // Two energy, deriving from the \doxygen{otb}{MRFEnergy} class need to be instanciated. One energy
+ // is required for the regularization, taking into account the relashionship between neighborhing pixels
+ // in the classified image. Here it is done with the \doxygen{otb}{MRFEnergyPotts} which implement
+ // a Potts model.
+ //
+ // The second energy is for the fidelity to the original data. Here it is done with an
+ // \doxygen{otb}{MRFEnergyGaussianClassification} class, which defines a gaussian model for the data.
+ //
+ // Software Guide : EndLatex
+
+ // Software Guide : BeginCodeSnippet
+
+ typedef otb::MRFEnergyPotts
+ <LabelledImageType, LabelledImageType> EnergyRegularizationType;
+ typedef otb::MRFEnergyGaussianClassification
+ <InputImageType, LabelledImageType> EnergyFidelityType;
+
+ // Software Guide : EndCodeSnippet
+
+ // Software Guide : BeginLatex
+ //
+ // The different filters composing our pipeline are created by invoking their
+ // \code{New()} methods, assigning the results to smart pointers.
+ //
+ // Software Guide : EndLatex
+
+ // Software Guide : BeginCodeSnippet
+
+ MarkovRandomFieldFilterType::Pointer markovFilter = MarkovRandomFieldFilterType::New();
+ EnergyRegularizationType::Pointer energyRegularization = EnergyRegularizationType::New();
+ EnergyFidelityType::Pointer energyFidelity = EnergyFidelityType::New();
+ OptimizerType::Pointer optimizer = OptimizerType::New();
+ SamplerType::Pointer sampler = SamplerType::New();
+
+ // Software Guide : EndCodeSnippet
+
+ // Software Guide : BeginLatex
+ //
+ // Parameter for the \doxygen{otb}{MRFEnergyGaussianClassification} class, meand
+ // and standard deviation are created.
+ //
+ // Software Guide : EndLatex
+
+
+ // Software Guide : BeginCodeSnippet
+
+ unsigned int nClass = 4;
+ energyFidelity->SetNumberOfParameters(2*nClass);
+ EnergyFidelityType::ParametersType parameters;
+ parameters.SetSize(energyFidelity->GetNumberOfParameters());
+ parameters[0]=10.0; //Class 0 mean
+ parameters[1]=10.0; //Class 0 stdev
+ parameters[2]=80.0;//Class 1 mean
+ parameters[3]=10.0; //Class 1 stdev
+ parameters[4]=150.0; //Class 2 mean
+ parameters[5]=10.0; //Class 2 stdev
+ parameters[6]=220.0;//Class 3 mean
+ parameters[7]=10.0; //Class 3 stde
+ energyFidelity->SetParameters(parameters);
+
+ // Software Guide : EndCodeSnippet
+
+
+ // Software Guide : BeginLatex
+ //
+ // Parameters are given to the different class an the sampler, optimizer and
+ // energies are connected with the Markov filter.
+ //
+ // Software Guide : EndLatex
+
+ // Software Guide : BeginCodeSnippet
+
+
+ optimizer->SetParameters(atoi(argv[6]));
+ markovFilter->SetNumberOfClasses(nClass);
+ markovFilter->SetMaximumNumberOfIterations(atoi(argv[5]));
+ markovFilter->SetErrorTolerance(0.0);
+ markovFilter->SetLambda(atof(argv[4]));
+ markovFilter->SetNeighborhoodRadius(1);
+
+ markovFilter->SetEnergyRegularization(static_cast<MarkovRandomFieldFilterType::EnergyRegularizationPointer>(energyRegularization));
+ markovFilter->SetEnergyFidelity(static_cast<MarkovRandomFieldFilterType::EnergyFidelityPointer>(energyFidelity));
+ markovFilter->SetOptimizer(optimizer);
+ markovFilter->SetSampler(sampler);
+
+ // Software Guide : EndCodeSnippet
+
+ // Software Guide : BeginLatex
+ //
+ // The pipeline is connected. An \doxygen{itk}{RescaleIntensityImageFilter}
+ // rescale the classified image before saving it.
+ //
+ // Software Guide : EndLatex
+
+ // Software Guide : BeginCodeSnippet
+
+ markovFilter->SetTrainingInput(reader2->GetOutput());
+ markovFilter->SetInput(reader->GetOutput());
+
+ typedef itk::RescaleIntensityImageFilter
+ < LabelledImageType, LabelledImageType > RescaleType;
+ RescaleType::Pointer rescaleFilter = RescaleType::New();
+ rescaleFilter->SetOutputMinimum(0);
+ rescaleFilter->SetOutputMaximum(255);
+
+ rescaleFilter->SetInput( markovFilter->GetOutput() );
+
+ writer->SetInput( rescaleFilter->GetOutput() );
+
+ // Software Guide : EndCodeSnippet
+
+ // Software Guide : BeginLatex
+ //
+ // Finally, the pipeline execution is trigerred.
+ //
+ // Software Guide : EndLatex
+
+ // Software Guide : BeginCodeSnippet
+
+ writer->Update();
+
+ // Software Guide : EndCodeSnippet
+
+ return 0;
+
+}
+
diff --git a/Examples/Markov/otbMarkovExamplesTests.cxx b/Examples/Markov/otbMarkovExamplesTests.cxx
index a7bbca718a4112bd17787acdc8c114c4c623cfaa..d37d9c63fa8d3f31bd586af3f9e026c4f294548d 100644
--- a/Examples/Markov/otbMarkovExamplesTests.cxx
+++ b/Examples/Markov/otbMarkovExamplesTests.cxx
@@ -25,12 +25,17 @@
void RegisterTests()
{
-REGISTER_TEST(MarkovRandomField1ExampleTest);
+ REGISTER_TEST(MarkovRandomField1ExampleTest);
+ REGISTER_TEST(MarkovRandomField2ExampleTest);
}
#undef main
#define main MarkovRandomField1ExampleTest
#include "MarkovRandomField1Example.cxx"
+#undef main
+#define main MarkovRandomField2ExampleTest
+#include "MarkovRandomField2Example.cxx"
+