[ITK-users] OtsuMultipleThresholdsImageFilter creates asymmetric classes for symmetric a distribution

[Dženan Zukić]

Hi Frederik,

I agree that the threshold should be larger than 2000, and should not
depend on whether the number of histogram bins is 127, 128 or 129 - with
this many bins it really shouldn't matter.

This example is simple enough that it would be worthwhile to step through
the code
<https://github.com/InsightSoftwareConsortium/ITK/blob/master/Modules/Filtering/Thresholding/include/itkOtsuMultipleThresholdsCalculator.hxx>
of OtsuMultipleThresholdsCalculator
<https://itk.org/Doxygen/html/classitk_1_1OtsuMultipleThresholdsCalculator.html>
to
see why it doesn't produce the result we expect. This might lead to a bug
fix.

Will you do it Frederik? Or wait for Francois or me to get around to it? If
you do, please report the findings on discourse <https://discourse.itk.org/>,
or submit a patch <https://itk.org/Wiki/ITK/Git/Develop> to Gerrit.

Regards,
Dženan

On Wed, Oct 18, 2017 at 3:03 AM, hellman <fredrik.hellman at gmail.com> wrote:

> Hi Dženan,
>
> With 127 bins I get: 2004.02
>
> With 129 bins I get: 1988.45
>
> Best wishes,
> Fredrik
>
>
>
> --
> Sent from: http://itk-users.7.n7.nabble.com/
> The ITK community is transitioning from this mailing list to
> discourse.itk.org. Please join us there!
> ________________________________
> 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/insight-users/attachments/20171018/133bf0c4/attachment.html>