[ITK-users] Geodesic 3D Shape guidance

[Dženan Zukić]

Hi Robert,

by default the precision for output does not include all digits in floats
and doubles. If there is a difference on 7th decimal place, it will cause
the exception but will not be visible in cout. Can you examine origin,
spacing and directions on the input images? I had that problem recently
(1.99999999 was different from 2.0).

HTH,
Dženan

On Mon, May 30, 2016 at 11:06 AM, Robert <scorpiuni at gmail.com> wrote:

> Dear Community,
> I encountered an error which I really can't get fixed.
> I wrote some code in order to do a segmentation with the geodesic function
> with shape guidance, but in 3D.
> I got my shape models, PCA data, etc, all done.
> Now in the geodesic code, I preprocess my image in which I want to find my
> structure (I need shape guidance, as the structure kind of melts to the
> background at one point, so I need the shape to restrict this).
> Preprocessing etc is all done, I now give my preprocessed image, and a
> levelset as well as the mean image and pc-images to my geodesic function
> (see link to code below).
> At one point, when testing my code, this error occurs:
> Description: itk::ERROR:
> GeodesicActiveContourShapePriorLevelSetImageFilter(0x115f970): Inputs do
> not
> occupy the same physical space!
> Now, as I already encountered this error once, I checked ALL the directions
> of ANY image in the pipeline (see code, console output usually marked with
> //*****DEBUG*****). They all give the same, correct value to the console.
> I really am at my wits end, I can't explain this error.
> I think it crashed at the writer called "writer", writer->SetInput(
> thresholder->GetOutput() );, as this one gets as input the output from the
> geodesic function (so, basically, the writer that will produce the final
> result).
> I commented this part out for testing purposes, and there is still a crash
> even later:
> "evalOutput5.nrrd"
> and
> "paraOutput6.nrrd" can never be reached, the same error message occurs
> here.
> I really cant find a solution, and the worst part is that due to this error
> I can't even test if the result is useful at all. Like I mentioned before,
> without shape guidance, the result is not satisfying, as "too much" of the
> structure is being segmented.
> Any help would be really appreciated.
> Greetings,
> Robert
>
>
>
> --
> View this message in context:
> http://itk-insight-users.2283740.n2.nabble.com/Geodesic-3D-Shape-guidance-tp7588926.html
> Sent from the ITK Insight Users mailing list archive at Nabble.com.
> _____________________________________
> 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.php
>
> 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://public.kitware.com/mailman/listinfo/insight-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/insight-users/attachments/20160530/54ccc8fc/attachment.html>