[Insight-users] Transform

[Pandu Jugnu]

Hi Luis,

As per you recommendation I am using
ScaleSkewVersor3DTransform. I had few questions about
it. How many parameters does it support? 9 or 12?
Also, while setting up the optimier using the
following code:

typename OptimizerType::ScalesType scales( 
  m_Transform->GetNumberOfParameters() );
  scales.Fill( 1.0);
  
 for ( int j = 4; j < 7; j++ ) 
    {
    scales[j] = m_TranslationScale;
    }

I am a bit confused if i have to change the "for" loop
to run from 6 to 9 or 9 to 12.

I am using the GradientDescentOptimizer along with
this transform. Is this is a good choice.

Thanks Luis
Pandu 



--- Luis Ibanez <luis.ibanez at kitware.com> wrote:

> 
> Hi Pandu,
> 
> In that case you may want to use the
> ScaleSkewVersor3DTransform
> 
>
http://www.itk.org/Insight/Doxygen/html/classitk_1_1ScaleSkewVersor3DTransform.html
> 
> 
> This transform supports the combination of rigid and
> scaling transforms.
> 
> 
> Regards,
> 
> 
>     Luis
> 
> 
> -----------------------------------------------
> Pandu Jugnu wrote:
> 
> > Hi Luis,
> > 
> > Thanks for your last email. It really helped clear
> my
> > doubts. I have one more question. 
> > I am trying to incorporate ScaleTransform along
> with
> > the QuaternionRigidTransform in
> MultiResMIRegistration
> > application so that I can take care of the scaling
> in
> > addition to tranlation and rotation. 
> > I went over the Doxygen files for ScaleTransform.
> But
> > I am not sure how to implement it along with
> > QuaternionRigidTransform in the application.
> > 
> > Can you guide me as how to accomplish this?
> > 
> > I appreciate your help Luis.
> > Thanks
> > Pandu
> > --- Luis Ibanez <luis.ibanez at kitware.com> wrote:
> > 
> > 
> >>Hi Pandu,
> >>
> >>This MultiResMIRegistration application uses a
> >>
> >>          QuaternionRigidTransform
> >>
> >>for performing the registration. The presence of
> >>the AffineTransform is required for converting
> >>the QuaternionRigidTransform into an Affine
> >>transform and returning it for being used
> elsewhere.
> >>
> >>An Affine transform can represent a rigid
> transform
> >>without any lose of precision. In practice you can
> >>use the Affine transform as a 'generic' transform
> >>that can internally represt
> >>
> >>     - Rotation transforms
> >>     - Scaling transforms
> >>     - Translation transforms
> >>     - Rigid transforms
> >>     - Similarity transforms
> >>     - Shearing transforms
> >>     - Reflection transforms
> >>
> >>In this way, no matter which one of the above
> >>transforms
> >>in actually used for performing the registration,
> >>you could
> >>always return the results represented in an Affine
> >>transform.
> >>
> >>
> >>No reason to worry,
> >>Just continue enjoying your reading of the code.
> >>
> >>
> >>Please let us know if you have further questions.
> >>
> >>
> >>
> >>   Thanks
> >>
> >>
> >>     Luis
> >>
> >>
> >>-----------------------
> >>Pandu Jugnu wrote:
> >>
> >>
> >>>Hi Luis
> >>>
> >>>I was going over the code for
> >>
> >>MultiResMIRegistration
> >>
> >>>application. I read in the guide that this
> >>
> >>application
> >>
> >>>uses "QuaternionRigidTransform" but I noticed the
> >>>presence of "Affine Transform" in the code too. 
> >>>Can you put some light on this as of why are we
> >>
> >>using
> >>
> >>>the Affine transform in this example.
> >>>
> >>>Thanks in advance.
> >>>Pandu
> >>>
> >>>
> >>>		
> >>
> >>
> >>
> >>
> >>
> > 
> > 
> > 
> > 		
> > _______________________________
> > Do you Yahoo!?
> > Express yourself with Y! Messenger! Free. Download
> now. 
> > http://messenger.yahoo.com
> > _______________________________________________
> > Insight-users mailing list
> > Insight-users at itk.org
> > http://www.itk.org/mailman/listinfo/insight-users
> > 
> 
> 
> 
> 
> 



	
		
__________________________________
Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage!
http://promotions.yahoo.com/new_mail