[Insight-users] itkScalarChanAndVeseSparseLevelSetImageFilter - RMS Change

[Sgouritsa, Eleni]

Hello Kishore,

thank you for your answer. Let me clarify myself: I am currently just working on the sparse case. My problem is that I get very high values of RMS Change between
the iterations of the level set. As you said, RMS values are used for indicating the convergence of the evolution. I would expect them to be around lets say 0.1 near convergence but I get all the time very high changes during the evolution, RMS change values like 263450, meaning huge numbers. 
Do you know why is this happening?
Thank you very much.

Regards,
Eleni


-----Original Message-----
From: Kishore Mosaliganti [mailto:kishoreraom at gmail.com]
Sent: Tue 8/18/2009 9:07 AM
To: Sgouritsa, Eleni ; insight-users at itk.org
Subject: Re: [Insight-users] itkScalarChanAndVeseSparseLevelSetImageFilter - RMS Change
 
Hi Eleni,

I am back at work :).

Ok, we made several corrections to the code in the past two months and you
can download the latest CVS version. Many thanks for your feedback on using
the code. It helped us identify bugs and clean up the code further.

Your question was regarding the RMS Error values computed by the Dense and
Sparse filters on the same image producing different values.

The dense filter works on the levelset function defined on the whole image
domain. As a result, the RMS values are computed on changes in the function
everywhere.

The sparse filter works only locally to the 0 level-set (hence the name
sparse). The RMS values look at a small band of pixels. Therefore, the
values in this case are at a different range.

Irrespective, the RMS values are used for indicating the convergence of the
evolution. Is there any reason why you want them to be at the same range ?

Best,
Kishore

On Mon, Aug 17, 2009 at 6:37 PM, Sgouritsa, Eleni
<esgourit at central.uh.edu>wrote:

>  Hello,
> I hope everything went well concerning the embassy.
> Do you have any updates for the high RMS values? Was this problem fixed?
>
> Regards,
> Eleni
>
>
>
> -----Original Message-----
> From: alex gouaillard [mailto:alexandre_gouaillard at hms.harvard.edu<alexandre_gouaillard at hms.harvard.edu>
> ]
> Sent: Thu 7/2/2009 2:32 PM
> To: Sgouritsa, Eleni
> Cc: insight-users at itk.org
> Subject: Re: [Insight-users] itkScalarChanAndVeseSparseLevelSetImageFilter
> - RMS Change
>
> hi eleni,
>
> Kishore has been release by the US embassy in India (without any
> ransom shall I had) and should be back in town july 9th.
>
> We will adress this issue right after that and keep you informed.
>
> thanks again for your feedback.
>
> alex.
>
>
> On Jul 2, 2009, at 12:53 PM, Sgouritsa, Eleni wrote:
>
> > Hello,
> > I still have a problem with very high values of RMS Change between
> > the iterations of the level set. Any help or direction on how to
> > solve this would be appreciated a lot.
> > Thank you very much.
> >
> > Best regards,
> > Eleni
> >
> > <ATT00001.txt>
>
>
>
> _____________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.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
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20090818/df11eed1/attachment.htm>