[ITK] Problem with Mattes' mutual information

[Constantinus Spanakis]

Hello, i tried to modify the example  DeformableRegistration15.cxx by
replacing the Regular step gradient Descent optimizer with One plus
One optimizer. At first i had problem with having too many samples
outside moving image buffer. I corrected that problem by reducing the
number of spatial samples. Then i got this problem:

itk::ExceptionObject (00000000002FE348)
Location: "double __cdecl itk::MattesMutualInformationImageToImageMetric<class i
tk::Image<short,3>,class itk::Image<short,3> >::GetValue(const class itk::Optimi
zerParameters<double> &) const"
File: c:\itk\modules\registration\common\include\itkMattesMutualInformationImage
ToImageMetric.hxx
Line: 571
Description: itk::ERROR: MattesMutualInformationImageToImageMetric(00000000027FA
FF0): Joint PDF summed to zero
Image (0000000002815040)
  RTTI typeinfo:   class itk::Image<double,2>
  Reference Count: 2
  Modified Time: 462
  Debug: Off
  Object Name:
  Observers:
    none
  Source: (none)
  Source output name: (none)
  Release Data: Off
  Data Released: False
  Global Release Data: Off
  PipelineMTime: 0
  UpdateMTime: 0
  RealTimeStamp: 0 seconds
  LargestPossibleRegion:
    Dimension: 2
    Index: [0, 0]
    Size: [50, 50]
  BufferedRegion:
    Dimension: 2
    Index: [0, 0]
    Size: [50, 50]
  RequestedRegion:
    Dimension: 2
    Index: [0, 0]
    Size: [50, 50]
  Spacing: [83.0652, 46.913]
  Origin: [2, 2]
  Direction:
1 0
0 1

  IndexToPointMatrix:
83.0652 0
0 46.913

  PointToIndexMatrix:
0.0120387 0
0 0.021316

  Inverse Direction:
1 0
0 1

  PixelContainer:
    ImportImageContainer (0000000002777D40)
      RTTI typeinfo:   class itk::ImportImageContainer<unsigned long,double>
      Reference Count: 1
      Modified Time: 463
      Debug: Off
      Object Name:
      Observers:
        none
      Pointer: 0000000002911380
      Container manages memory: true
      Size: 2500
      Capacity: 2500

Could someone please explain the problem more analytically and help me solve it?

Thank you