[ITK] [ITK-users] Watershed for 3D Image Segmentation - One Seed Point -

Richard Beare richard.beare at gmail.com
Sat Feb 11 16:41:54 EST 2017


The usual thing I do when testing watersheds in a new application is to
manually create marker images using an image segmentation tool, such as
itksnap or fslview. The format of the image doesn't matter, as long as ITK
can read it. The key thing is to make sure you save a label image - i.e one
with each marker having a different voxel value. This gives you an idea of
how sensitive the problem is to marker placement and what sort of control
image you might need.

On Sun, Feb 12, 2017 at 2:58 AM, Abdelkhalek Bakkari <
bakkari.abdelkhalek at hotmail.fr> wrote:

> Dear Dr Richard Beare,
>
>
> I am trying your code source which is published in the insight journal
> about the " The watershed transform in ITK - discussion and new
> developments".
>
>
> I debugged the perf3D and mperf3D. However, in the mperf3D, it needs two
> input images. Could you please explain for me the way of adopting it for a
> .mha file ?
>
>
> Thank you in advance.
>
>
> Best regards,
>
>
> Abdelkhalek Bakkari
> Ph.D candidate in Computer Science
> Institute of Applied Computer Science
> Lodz University of Technology, Poland
>
>
>
>
> ------------------------------
> *From:* Richard Beare <richard.beare at gmail.com>
> *Sent:* 10 February 2017 22:00
> *To:* Abdelkhalek Bakkari
> *Cc:* insight-users
> *Subject:* Re: [ITK-users] Watershed for 3D Image Segmentation - One Seed
> Point -
>
> The greedy approaches, like MorphologicalWatershedFromMarkers, require 2
> seeds. One is the background seed. Sometimes that can be something very
> simple like the image boundary. The advantage of greedy approaches is that
> you don't need to specify stopping conditions. The cost is obviously the
> extra seed. Another advantage is that most  of the methods naturally
> support lots of independent seeds, if you want them.
>
> On Sat, Feb 11, 2017 at 7:45 AM, Abdelkhalek Bakkari <
> bakkari.abdelkhalek at hotmail.fr> wrote:
>
>> Dear ITK users,
>>
>>
>> I Would like to know if there is an algorithm in ITK that is proposed for a semi-automatic segmentation of three-dimensional images.
>>
>> One seed location should be chosen like in Region Growing.
>>
>> Thank you in advance.
>>
>> Best regards,
>>
>>
>> Abdelkhalek Bakkari
>> Ph.D candidate in Computer Science
>> Institute of Applied Computer Science
>> Lodz University of Technology, Poland
>>
>>
>>
>> _____________________________________
>> 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/community/attachments/20170212/86092dcf/attachment.html>
-------------- next part --------------
_____________________________________
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


More information about the Community mailing list