[Insight-users] Level Set Segmentation

[Benjamin Smith]

Good morning Maria,

I have an ITK implementation of the Chan-Vese "Active Contours without Edges
for Vector Valued Images" paper (
http://www.math.ucla.edu/~lvese/PAPERS/JVCIR2000.pdf).  This paper details a
level set method to create a segmentation which can be based on spectral
information (take a look at some of the examples in the paper).  If this
might be applicable to your problem, I would be interested in sharing the
code and / or discussing further.

Cheers,

Benjamin Smith


------------------------------
>
> Message: 3
> Date: Mon, 8 Jan 2007 00:31:04 +0100
> From: "Maria Cira Avvinto" <millarkaa at gmail.com>
> Subject: [Insight-users] Level Set Segmentation
> To: insight-users at itk.org
> Message-ID:
>         <c26d75cf0701071531j6718c2bdk715c73dd87245c37 at mail.gmail.com>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> Hi,
> I have a two important question:
> 1)Can i use level set segmentation for multispectral image?
>
> 2)Where can i collocate level set method in this scheme?
>
>                        Image Segmentation Method
>                                /                              \
>                           Single Contrast              Multispectral
>                               |
>      |_______
>                           Threshold                                 /
>              \
>                               |
> Supervised Unsupervised
>                          Boundary Tracing                       |
>                              |                                  Pattern
>                          Edge Detector                  Recognition
>                             |                                            |
>                          Seed Growing                  Artificial
>                             |                                  Neural
>                         Template Model                Networks
>                             |
>                        Random Fields
>
>
>
> Please, help me.
> Thanks a lot  to everybody.
>
> Mary
>
>
> ------------------------------
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://public.kitware.com/pipermail/insight-users/attachments/20070108/fd0cb9f1/attachment.html