[Insight-users] why use 3D watershed segmentation is very slow?

Dongfeng Han handongfeng at gmail.com
Tue Mar 22 09:34:12 EDT 2011


Hi Sergio,
Thank you for your advice.
It seems faster than before.
Now I segment a liver image with size 300x300x150 , voxel spacing is
0.5x0.5x1.5.
I set threshold =  0.01 level =  0.1.
The running time is about 5 min and the total number of region is about
37000.
Is that reasonable? and do you have any suggestion on the parameters and the
running time?
thanks
Best
dongfeng

On Tue, Mar 22, 2011 at 12:43 AM, Sergio Vera <sergio.vera at alma3d.com>wrote:

> Hi there
>
> The first check is to make sure you are using a release version of ITK
> libraries.
> Debug versions are several times slower (6-10 times in my experience)
> Once this is checked, we can start looking at your parameters.
>
> Regards
>
> On Mon, Mar 21, 2011 at 9:29 PM, Dongfeng Han <handongfeng at gmail.com>wrote:
>
>> hello everyone,
>> I use watershed method as shown the example code in ITK, to segment 3D
>> image with size of 300x300x100
>> The pixel type is unsigned short.
>> It's very very slow. may be >20 or 30 min.
>> Any one know how to make it faster?
>> May be I set wrong parameters.
>>
>> best
>> dongfeng
>>
>> _____________________________________
>> 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
>>
>>
>
>
> --
> Sergio Vera
>
>  Alma IT Systems
>  C/ Vilana, 4B, 4º 1ª
>  08022 Barcelona
>  T. (+34) 932 380 592
>  www.alma3d.com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20110322/67d49fd8/attachment.htm>


More information about the Insight-users mailing list