Hi, all
I have a pair of fixed image and moving image. And the transform is itk::AffineTransform, which includes translation, rotation, isotropic scale. In ITK-SNAP, it's like:
My question is, what does translation mean?
I found the source code of ITK-SNAP. Here it is:
// ----------------------------------------------------------------------
// Pinv: a matrix,
SVD result of AffineMat(3x3)
// R: a matrix, SVD result of AffineMat(3x3)
// crot: the center of rotation
// cimg: the center of moving image
Vec3 v_tran = Pinv * (b - cimg) + cimg + R * crot - crot;
// ----------------------------------------------------------------------
I don't understand what this translation is. Is it one of the followings:
1) The translation from the fixed image center to the moving image center
2) The translation from the fixed image origin to the moving image origin
3) Something else?