[Insight-users] ImageRegionIterator problem

[Ali Mouizina]

I think, I figured it out why I have this problem. In the png attached to
this mail you can have a precise idea of what this function is supposed to
do. I think that the output image yield by this function has the same size
as the input region and this is not what I want. How can I fix this problem
?

 

Regards,

 

Ali

 

De : Ali Mouizina [mailto:ali.mouizina at gmail.com] 
Envoyé : lundi 30 janvier 2012 13:05
À : insight-users at itk.org
Objet : ImageRegionIterator problem 

 

Hi,

 

I’m currently having a problem using ImageRegionIterator. I copy
successfully a region of my input image but when I want to process it
through a filter it crashes. Here’s the code below :

 

// In the main program


.

squarediff->SetInput1(read);

squarediff->SetInput2(imshift(deltax,deltay,deltaz));

squarediff->Update(); // -> It crashes here


..

 

// imshift program

 

ImageType::Pointer Weights::imshift(int x, int y, int z){

       

       int x1s=max(1,x+1);

       int x2s=min(nb_col,nb_col+x); //nb_col is the number of columns of my
input image

       int y1s=max(1,y+1);

       int y2s=min(nb_ln,nb_ln+y);

       int z1s=max(1,z+1);

       int z2s=min(nb_slice,nb_slice+z);

 

       int x1=max (1,-x+1);

       int x2=min(nb_col,nb_col-x);

       int y1=max(1,-y+1);

       int y2=min(nb_ln,nb_ln-y);

       int z1=max(1,-z+1);

       int z2=min(nb_slice,nb_slice-z);

 

       ImageType::Pointer image = ImageType::New();

 

       typedef itk::ImageRegionConstIterator< ImageType > ConstIteratorType;

       typedef itk::ImageRegionIterator< ImageType> IteratorType;

 

       // Input region definition

       ImageType::RegionType inputRegion;

       ImageType::RegionType::IndexType inputStart;

       ImageType::RegionType::SizeType size;

       inputStart[0] = x1s-1;

       inputStart[1] = y1s-1;

       inputStart[1] = z1s-1;

       size[0] = x2s-x1s+1;

       size[1] = y2s-y1s+1;

       size[2] = z2s-z1s+1;

       inputRegion.SetSize( size );

       inputRegion.SetIndex( inputStart );

 

       // Output region definition

       ImageType::RegionType outputRegion;

       ImageType::RegionType::IndexType outputStart;

       outputStart[0] = x1-1;

       outputStart[1] = y1-1;

       outputStart[1] = z1-1;

       outputRegion.SetSize( size );

       outputRegion.SetIndex( outputStart );

       

       // !!!  Warning : Spacing and origin are not defined as we don't
expect to do image registration

       image->SetRegions( outputRegion );

       image->Allocate();

       image->FillBuffer(0);

 

       // Copying 

       ConstIteratorType inputIt( read, inputRegion );

       IteratorType outputIt( image, outputRegion );

 

       for ( inputIt.GoToBegin(), outputIt.GoToBegin(); !inputIt.IsAtEnd();
++inputIt, ++outputIt)

             outputIt.Set( inputIt.Get() );

 

       return image;

}

 

This function copies all the values properly as I checked it. Hence, I don’t
see why can’t I process it through a filter. By the way this functions aims
at copying the pixel values from an input image into an output image by
shifting their positions according to x,y and z values. 

Regards,

 

Ali

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20120130/facb7e6a/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: example.png
Type: image/png
Size: 15063 bytes
Desc: not available
URL: <http://www.itk.org/pipermail/insight-users/attachments/20120130/facb7e6a/attachment.png>