[Insight-users] Parallel iteration of 2D/3D image series

Mat matthias.dodt at mdc-berlin.de
Wed Feb 17 11:25:22 EST 2010


Hi Luis!

Yes, i realized that it was getting slow...

I got at first confused with the SliceIterator- thanks, that filter was 
exactly what i was looking for! Works now;-)

Best,

mat

Luis Ibanez schrieb:
> Hi Matt,
>
> The filters that must be applied in 2D should be instantiated using
> a 2D image type.
>
> When you select the slice (frame) that you want to process via one
> of these 2D filter, out of the (3D = 2D+time) volume that you have
> created, you should use the ExtractImageFilter, with the option  that
> converts a 1-slice 3D image into a 2D image.
>
> See the example in:
>
>          Insight/Examples/IO/ImageReadExtractWrite.cxx
>
>
> Don't try to trick a 3D filter into behaving like a 2D filter by simply
> manipulating its neighborhood settings. That will lead to terribly
> bad performance.
>
>
>      Regards,
>
>
>            Luis
>
>
>
> -----------------------------------------------------------------------------------------
> On Tue, Feb 2, 2010 at 5:12 AM, Mat <matthias.dodt at mdc-berlin.de> wrote:
>   
>> Hi!
>>
>> A simple question due to data processing. I have a basic image-analysis
>> pipeline which i would like to extend to work on a stack of images (video).
>> Some Filters should only be applied in 2D while others require 3D
>> information of the previous frames. I am using the ImageSeriesReader on
>> float , dim=3 images type to read an image stack. Normally i would do this
>> in my own filters by using a NeighborhoodIterator with defined radius.
>>
>> How do i restrict certain filters (for ex.
>> CurvatureAnisotropicDiffusionImageFilter) to work on each slice (in 2D)
>> while others should work in 3D? It doesnt seem to be possible to define a
>> radius or dim to work on for this filter...
>>
>> thanks,
>>
>> mat
>>
>> _____________________________________
>> 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
>>
>>     


More information about the Insight-users mailing list