[Insight-users] itkDemonsRegistration: Demon points selection?

[Luis Ibanez]

Hi Subrahmanyam,

 The term "Demons" covers a very large family of algorithms.

In general, any algorithm that is based on computation of local
effects could be considered a version of a Demons-like algorithm.

The implementation of the DemonsRegistrationFilter in ITK uses
all pixels in the image as Demon points, and computes a dense
deformation field from them.

As you pointed out, you could, of course, implement a smarter
way of selecting demons points, and then use interpolation (or
regularization) in order to estimate the deformation field in the
spaces where no demon points were selected.

Should you decide to go about implementing this variation of
Demons, it will be great if you could share your code with the
ITK community by posting it as a paper to the Insight Journal.  :-)


Please let us know if you have further questions,


     Thanks


           Luis


--------------------------------------------------------------------------------------------------
On Thu, Feb 25, 2010 at 4:37 AM, Subrahmanyam Gorthi
<subrahmanyam.gorthi at gmail.com> wrote:
> Dear ITK users,
>
> I am not sure I correctly understand the ITK demons force implementation. My
> question is regarding the choice of demon points. Thirion suggested a
> threshold on the image gradient norm as criterion for demon point selection,
> that is, not all points of moving and/or fixed image are used for demons
> force computation.
>
> In the itkDemonsRegistrationFunction.txx,  it seems that all index points of
> the fixed image or mapped points of the moving image are used and only
> updating values are dropped depending on denominator values (with a
> threshold on the image intensity differences).
>
> But I could not find anything related to a demon points selection upper a
> certain value of the gradient image intensity. I probably missed something,
> can you please clarify me about this?
>
> Thanks in advance.
>
>
> _____________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Kitware offers ITK Training Courses, for more information visit:
> 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
>
> Follow this link to subscribe/unsubscribe:
> http://www.itk.org/mailman/listinfo/insight-users
>
>