[Insight-users] Kernel Transform issues
Luis Ibanez
luis.ibanez at kitware.com
Tue Apr 18 13:23:03 EDT 2006
Hi Gary,
How many landmark points were you passing to the ElasticBodySplineKernel
Transform when you notice these artifacts in the image resampling ?
About the registration issue that you report: that the final parameters
are the same as the initial ones, that looks more like an optimizer
issue, than an issue related to the Transform.
+ What optimizer were you using for those trials ?
+ How many iterations of the optimizer did you run ?
+ Did you plotted the values of the Metric at every iteration of
the optimizer ?
+ Were those values of the metric changing significantly ?
Please let us know,
Thanks
Luis
-----------------
Gary Kline wrote:
> We have been developing some software based on ITK, and ran into some
> problems with some of the kernel transforms. In particular, the elastic
> body spline kernel transform, when applied to an image, was introducing
> a lot of noise into the transformed image. Depending on the input
> image, the noise took the form of “speckle” noise, or lines in the
> output image.
>
>
>
> Has anyone else run into this problem?
>
>
>
> The other problem we have run into is trying to use the elastic body
> spline kernel, or the elastic body reciprocal spline kernel, in image
> and point set registration applications. When we use either of these
> two kernels, the final parameters reported after the registration
> process are the same as the initial parameters. We do not have this
> problem using the other kernel transforms.
>
>
>
> Has anyone else seen this problem?
>
>
>
> We are currently using both Windows with VS.NET 2003, and Linux with gcc
> 3.4(? – working from memory, my VM is not running).
>
>
>
> Thank you,
>
>
>
> Gary Kline
>
> AccuSoft Corp.
>
>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk.org
> http://www.itk.org/mailman/listinfo/insight-users
More information about the Insight-users
mailing list