[Insight-users] question on transforming points

[Luis Ibanez]

Hi John,

Thanks for letting us know about the Transforms that you are using.

Particularly, the fact that you are using a Coarse BSpline Transform
followed by a Fine BSpline Transform.


Please note that if you initialize the Fine BSpline transform by using
the BSpline Transform that results from the Coarse registration level,
then you don't need to map the points through both Transforms.

       Instead, just using the Fine Transform should be enough.


For an example on how to initialize a finer BSpline Transform from
a finer one you may want to look at:

               Insight/Examples/Registration/
                      DeformableRegistration6.cxx
                      DeformableRegistration15.cxx


     Regards,


            Luis


----------------------------------------------------------------------------------
On Sat, Apr 17, 2010 at 8:54 PM, John Jan Drozd <john.drozd at gmail.com>wrote:

> Hi Luis,
>
> It's nice to talk with you again.
> I am using a Bspline deformable transform and I am initializing it with an
> affine transform. I am using a course bspline deformable transform followed
> by a finer resolution bspline deformable transform. This gives me greater
> accuracy and the code runs much faster using this multi-resolution approach.
> I used and modified code to work with 3 dimensional dicom brain images from
> deformableregistration8.cxx and deformableregistration15.cxx.  In mapping my
> points, I now realize that I have to transform the fixed point twice, first
> with the course transformation and then with the finer transformation. I was
> only transforming the point with the finer transformation and that's one
> possibility why my transformation of a point from the brain ventricles in
> the fixed image did not map that accurately to within ventricles in the
> moving image. I use these as a way of automatically generating seed points
> for a region growing algorithm.
>
> Thanks for clarifying that the bsplinetransform takes care of the origins.
>
> Take care,
> John
>
> Sent from my iPhone
>
> On 2010-04-17, at 6:57 PM, Luis Ibanez <luis.ibanez at kitware.com> wrote:
>
>
> Hi John,
>
>
> 1) What Transform are you using ?
> 2) How did you initialize it ?
>
>
> If you are following any of the BSplineDeformable
> Transform as shown in the examples in
>
>                Insight/Examples/Registration
>
> then the Transform will include the translation
> between the image origins.
>
>
>      Regards,
>
>
>             Luis
>
>
>
> -----------------------------------------------------------------------------
> On Thu, Apr 15, 2010 at 5:30 PM, John Drozd < <john.drozd at gmail.com>
> john.drozd at gmail.com> wrote:
>
>> Hello,
>>
>> I perform a multi-resolution deformable registration between 2 MRI images
>> from 2 different people using a multi-modal metric.
>> Then I map points from the fixed image to the moving image using the
>> transform obtained from the registration using TransformPoint();
>> Do I need to include the difference between the coordinates of the origins
>> of both or either images in my calculation, or does the TransformPoint take
>> care of this?
>>
>> Thank you,
>> John
>>
>> _____________________________________
>> Powered by <http://www.kitware.com>www.kitware.com
>>
>> Visit other Kitware open-source projects at
>>  <http://www.kitware.com/opensource/opensource.html>
>> http://www.kitware.com/opensource/opensource.html
>>
>> Kitware offers ITK Training Courses, for more information visit:
>>  <http://www.kitware.com/products/protraining.html>
>> http://www.kitware.com/products/protraining.html
>>
>> Please keep messages on-topic and check the ITK FAQ at:
>>  <http://www.itk.org/Wiki/ITK_FAQ>http://www.itk.org/Wiki/ITK_FAQ
>>
>> Follow this link to subscribe/unsubscribe:
>>  <http://www.itk.org/mailman/listinfo/insight-users>
>> http://www.itk.org/mailman/listinfo/insight-users
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20100418/03543cfb/attachment.htm>