[Insight-users] Resampling the moving image?

[Luis Ibanez]

Hi Motes,

When you say that you "apply the transform" to the fixed image,
do you mean that you are resampling the Fixed image by using
the Transform ?

If so,
there is no surprise that the outcome looks like the inverse transform.

Once again,
the Transform that ITK computes is intended to represent the mapping
of points from the Fixed image coordinate system, to the Moving image
coordinate system, because that is the mapping that is required for
resampling the Moving image intensity values into the space of the Fixed
image.

You should not resample, the Fixed image,
you should resample the Moving image.


   Regards,


      Luis


-------------------------------------------------
On Thu, Nov 12, 2009 at 6:16 PM, motes motes <mort.motes at gmail.com> wrote:
> On Sun, Nov 8, 2009 at 7:47 PM, Luis Ibanez <luis.ibanez at kitware.com> wrote:
>> Hi Motes,
>>
>> If you have a Fixed image "F" and a Moving image "M",
>> and the output of your registration process is the transform
>> "T",
>>
>> Then applying "T" to the *points* of the image F will give you
>> the coordinates of the corresponding points in the image M.
>>
>> Following your example,
>> if the transform turns out to be a translation:
>>
>>                                      T = (1,5)
>>
>> That means that if we take a point P from the Fixed image
>> space, and P has coordinates P=(x,y), then we can find the
>> coordinates of its corresponding point Q in the Moving image
>> by the following operation:
>>
>>                        Q = T(P) = ( x+1, y+5 )
>>
>> The expression T(F) could be interpreted as "mapping all
>> the points from the fixed image".
>>
>> T is indeed the transform that is used for resampling the
>> Moving image M into the coordinates of the Fixed image F.
>>
>>
>> The way resampling works, is by:
>>
>>    A)  visiting all the pixels of the Fixed image,
>>    A.1) for each pixel, compute its physical coordinates
>>           in the space of the Fixed image
>>    A.2) Use the transform T to compute the corresponding
>>            point Q = T(P) in the moving image space
>>    A.3) convert Q from the physical coordinates of the moving
>>            image into the grid coordinates of the moving image
>>    A.4)  Apply an interpolator as needed.
>>
>
> Ok but that is the exact same thing that goes on in the registration
> method in the metric so I don't see any difference between how the
> transform is applied in the registration method and how its used in
> the resampler.
>
> Is it because the moving image is in fact passed to the resampler as
> input and is now treated as the fixed image?
>
> I have tried to compute a know transformation of the fixed image using
> the BSplineWarping1.cxx where I specify the fixed image as input for
> both the fixed image and the moving image. The result is a deformed
> version of the fixed image which I use as the moving image in the
> registration process.
>
> When the registration process is complete I take the final transform
> parameters and apply those to the original fixed image. Basically I
> just  run the BSplineWarping1.cxx example again but with the paramters
> produced in the registration method. Below is the result:
>
> http://37133.vs.webtropia.com/apache2-default/test/grid.jpg
>
> Now this looks wierd. When I apply the transform from the registration
> process to the fixed image I get something that looks like the
> "inverted" version of the moving image.
>
> Should the result not look like the deformed image?
>