[Insight-users] Label Disconnection

[Sergio Vera]

Pietro,

Is your segmentation really disconnected or is an issue with visualization?
I mean, marching cubes isovolumes often do not work extremelly well with
thin structures, and they generate apparently unconnected structures while
in the binary 27-connected neighborhood everithing is connected. This is
specially true on thin or small regions.

If objects are really unconnected some options are hole filling methods
(morphologycal closing) or iterativeVotingHoleFill filters

Regards

On Wed, May 23, 2012 at 11:41 AM, Pietro Nardelli <pie.nardelli at gmail.com>wrote:

> Hi everybody,
>
> I have a question: does anyone know how could fix disconnected label?
> I explain: I'm developing a module to segment the airway of the lung
> starting from a DICOM dataset, and in order to do so I'm using a
> threshold connected method. Now, since this method use a
> 26-neighborhood region growing, I get some disconnected pieces of
> branches (I've attached a picture of what I mean). Does something
> exist to create a piece of label in order to fix these gaps?
>
>
> Thanks so much,
>
> Pietro
>
> _____________________________________
> 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://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/20120523/6b69a5f4/attachment.htm>