[Insight-users] A question about Spatial Objects Transformations: ObjectToNode Transform and ObjectToParent Transform

[Haiyong Xu]

Hello,

I have read the section in the ITK software guide PDF about Spatial
Objects, there are two transformations that are not clear to me:
ObjectToNode Transform and ObjectToParent Transform.

Please refer to the Figure 5.1 in ITK software guild PDF. In this
figure, it seems that an object (spatial object) has a coordinate
system, and Node has a coordinate system too. What's the difference
between them? After reading the source code in
itk::AffineGeometryFrame, it looks like that the ObjectToNode
transform will always be identity. Can I regard those two coordinate
systems as one?

Also, the name ObjectToParent transform is ambiguous. Is this a
ObjectToParentNode or ObjectToParentObject?

BTW, please correct the following paragraph in the ITK software guide
PDF (page 106), there are errors:

"... As one can see from the figure 5.1, the ObjectToParentTransform
is composed of two transforms: ObjectToNodeTransform and
NodeToParentNodeTransform.
The ObjectToNodeTransform is not applied to the children, but the
ObjectToNodeTransformis.
Therefore, if one sets the ObjectToParentTransform, the ObjectToNodeTransform
is modified."

Thanks.

Haiyong