Example usage:

./LabelMapToVectorData Input/labelImage_UnsignedChar.tif Output/LabelMapToVectorData.shp

Example source code (LabelMapToVectorData.cxx):

// \index{otb::LabelMapToVectorDataFilter}
// This class vectorizes a LabelObject to a VectorData.
// The \doxygen{otb}{LabelMapToVectorDataFilter} converts an \doxygen{itk}{LabelMap} to an
// \doxygen{otb}{VectorData} where all the pixels get the attribute
// value of the label object they belong to.
// It uses the class \doxygen{otb}{LabelObjectToPolygonFunctor} which
// follows a finite state machine described in \cite{Francis2000}.
//  Only polygon conversion is available yet.

#include "otbImageFileReader.h"
#include "otbVectorDataFileWriter.h"
#include "otbVectorData.h"
#include "otbVectorDataProjectionFilter.h"

#include <fstream>
#include <iostream>

#include "otbImage.h"

// These are the main header files which need to be included:
#include "otbLabelMapToVectorDataFilter.h"
#include "otbAttributesMapLabelObject.h"
#include "itkLabelImageToLabelMapFilter.h"

int main(int argc, char* argv[])
  /** Use the labelObjecttopolygon functor (not thread safe) only polygon conversion is available yet*/
  if (argc != 3)
    std::cerr << "Usage: " << argv[0];
    std::cerr << " inputImageFile outputVectorfile(shp)" << std::endl;
    return EXIT_FAILURE;
  const char* infname  = argv[1];
  const char* outfname = argv[2];

  // The image types are defined using pixel types and
  // dimension. The input image is defined as an \doxygen{itk}{Image},
  // the output is a \doxygen{otb}{VectorData}.

  const unsigned int Dimension = 2;
  using LabelType              = unsigned short;
  using LabeledImageType       = otb::Image<LabelType, Dimension>;
  using VectorDataType         = otb::VectorData<double, 2>;

  // We instantiate reader and writer types
  using LabeledReaderType = otb::ImageFileReader<LabeledImageType>;
  using WriterType        = otb::VectorDataFileWriter<VectorDataType>;

  // Label map typedef
  // The Attribute Label Map is
  // instantiated using the image pixel types as template parameters.
  // The LabelObjectToPolygonFunctor is instantiated with LabelObjectType and PolygonType.

  using LabelObjectType    = otb::AttributesMapLabelObject<LabelType, Dimension, double>;
  using LabelMapType       = itk::LabelMap<LabelObjectType>;
  using LabelMapFilterType = itk::LabelImageToLabelMapFilter<LabeledImageType, LabelMapType>;

  LabeledReaderType::Pointer lreader = LabeledReaderType::New();
  WriterType::Pointer        writer  = WriterType::New();

  //  Now the reader and writer are instantiated and
  //  the input image is set and a name is given to the output image.


  //  Then, the input image is converted to a map of label objects.
  //  Here each white region connected regions are converted. So the background is define all zero pixels.

  LabelMapFilterType::Pointer labelMapFilter = LabelMapFilterType::New();

  //  Then, the \doxygen{otb}{LabelMapToVectorDataFilter} is instantiated. This is
  // the main filter which performs the vectorization.

  using LabelMapToVectorDataFilterType = otb::LabelMapToVectorDataFilter<LabelMapType, VectorDataType>;

  LabelMapToVectorDataFilterType::Pointer MyFilter = LabelMapToVectorDataFilterType::New();


  //  The output can be passed to a writer.


  //  The invocation of the \code{Update()} method on the writer triggers the
  //  execution of the pipeline.  As usual, it is recommended to place update calls in a
  //  \code{try/catch} block in case errors occur and exceptions are thrown.
    return EXIT_SUCCESS;
  catch (itk::ExceptionObject& excep)
    std::cerr << "Exception caught !" << std::endl;
    std::cerr << excep << std::endl;
  catch (...)
    std::cout << "Unknown exception !" << std::endl;
    return EXIT_FAILURE;

  return EXIT_SUCCESS;