[Insight-users] Compute of the normalized mutual Information ITK3/ITK4

[Paolo Zaffino]

Hi,
I try to update my situation:
I get a different NMI value between ITK3 and ITK4 just if I set the 
upper and/or the lower intensity bound.
I think that the bounds are correctly setted because in debug mode I can 
see the passed values and they are ok.
I'm using this code:

  if (min_value_defined) {
         MetricType::MeasurementVectorType lower_bounds;
         #ifdef ITK4
         lower_bounds.SetSize(2);
         #endif
         lower_bounds.Fill(min_value);
         metric->SetLowerBound(lower_bounds);
     }

if (max_value_defined) {
         MetricType::MeasurementVectorType upper_bounds;
         #ifdef ITK4
         upper_bounds.SetSize(2);
         #endif
         upper_bounds.Fill(max_value);
         metric->SetUpperBound(upper_bounds);
     }

Can anybody help me to figure out what is happening?
Thanks a lot.

Paolo





On 02/07/2013 14:40, Paolo Zaffino wrote:
> Dear ITK community,
> I'm using NormalizedMutualInformationHistogramImageToImageMetric 
> function to compute the normalized mutual information.
> I compile it on ITK 3.20.1 and on ITK 4.1 with no problem.
> The strange thing is that I get a different value between the binary 
> compiled against ITK3 and the binary compiled against ITK4.
> It is "normal"? Is it a bug or between the two ITK versions is changed 
> the way to compute this value?
> Thanks a lot.
> Paolo
> _____________________________________
> 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
>