[Insight-users] Computing the Bounding Box of a Ellipse Spatial Object after applying an offset

[Lino Ramirez]

Hello All,

I am trying to compute the bounding box of a couple of ellipses in 3-D.
The radius in each dimension is set to 10. When the ellipses are centered
in the origin, the bounds are what I expect (from -10 to 10). If I apply
an offset only in the z axis to both ellipses (20 for the first one and 50
for the second one), the results are not what I expect ([-10, 10, 10, 10,
-10, 30] for the first ellipse and [-1, 10, 40, 10, -10, 60] for the
second one). The results I get are:

Ellipse 1 After Transform Bounding Box: [-10, 10, -10, 10, 10, 30]
Ellipse 2 After Transform Bounding Box: [-10, 10, -10, 10, 40, 60]

I am using a CVS version dated April 29, 2005. My compiler is Visual
Studio .Net. The portion of the code I used for the computation is located
at the end of this message.

I am not sure if this is a bug or if I am forgetting something in my code

Any help would be really appreciated

Regards,

Lino


//-----------
//Portion of the code

  typedef itk::EllipseSpatialObject< 3 >   EllipseType;
  EllipseType::Pointer ellipse1 = EllipseType::New();
  EllipseType::Pointer ellipse2 = EllipseType::New();
  ellipse1->SetRadius(  10.0  );
  ellipse2->SetRadius(  10.0  );

  ellipse1->ComputeBoundingBox();
  EllipseType::BoundingBoxType * ellipse1BoundingBox =
ellipse1->GetBoundingBox();
  std::cout << "Ellipse 1 Initial Bounding Box: " <<
ellipse1BoundingBox->GetBounds() << std::endl;

  ellipse2->ComputeBoundingBox();
  EllipseType::BoundingBoxType * ellipse2BoundingBox =
ellipse2->GetBoundingBox();
  std::cout << "Ellipse 2 Initial Bounding Box: " <<
ellipse2BoundingBox->GetBounds() << std::endl;


  EllipseType::TransformType::OffsetType soOffset;
  soOffset[ 0 ] =  0.0;
  soOffset[ 1 ] =  0.0;
  soOffset[ 2 ] =  20.0;

  ellipse1->GetObjectToParentTransform()->SetOffset(soOffset);
  ellipse1->ComputeObjectToWorldTransform();

  soOffset[ 0 ] =   0.0;
  soOffset[ 1 ] =   0.0;
  soOffset[ 2 ] =   50.0;
  ellipse2->GetObjectToParentTransform()->SetOffset(soOffset);
  ellipse2->ComputeObjectToWorldTransform();

  ellipse1->ComputeBoundingBox();
  std::cout << "Ellipse 1 After Transform Bounding Box: " <<
ellipse1BoundingBox->GetBounds() << std::endl;

  ellipse2->ComputeBoundingBox();
  std::cout << "Ellipse 2 After Transform Bounding Box: " <<
ellipse2BoundingBox->GetBounds() << std::endl;