[Insight-users] Normalized MI

Karthik Krishnan Karthik.Krishnan at kitware.com
Thu Apr 28 15:58:04 EDT 2005


Try looking at the evolution of the joint histograms as iterations
progress. The HistogramToEntropyImageFilter and derivatives do this. See
example Registration/ImageRegistrationHistogramPlotter.cxx. This plots
the joint histogram at every iteration. This will tell you if you are
overbinning or underbinning or if there is enough information to drive
MI. You should see that MI tries to re-group the bins. Sometimes you
might just have a few bins occupied and that may not be sufficient to
drive MI.

Run the example runs on BrainT1SliceBorder20 and BrainProtonDensitySliceShfited13x17y.png,
and you will see the joint histograms grouping.

Thanks
karthik


Yu Qi wrote:

> I copied some code from
> Testing/Code/Algorithms/itkNormalizedMutualInformationHistogramImageToImageMetricTest.cxx
> into my application of registration of MRI and fMRI.
> But the result is not good.
> I have two questions:
> 1. do i need metric->SetInterpolater();
> metric->SetTransform();
> metric->SetFixedImageRegion();,
> as I set those in registration->SetXXXXXX() ?
> 2. Which parameters are most important in normalized histogram MI
> metric? and how to adjust them to get a good result?
> BTW, I found there is no rotation in my result.
> Thanks!
> Best regards,
> Yu Qi
>
>
> ------------------------------------------------------------------------
> *Do You Yahoo!?*
> 150万曲MP3疯狂搜,带您闯入音乐殿堂 <http://music.yisou.com>
> 美女明星应有尽有,搜遍美图、艳图和酷图 <http://image.yisou.com>
> 1G就是1000兆,雅虎电邮自助扩容!
> <http://cn.rd.yahoo.com/mail_cn/tag/1g/*http://cn.mail.yahoo.com/event/mail_1g/>
>
>
>------------------------------------------------------------------------
>
>_______________________________________________
>Insight-users mailing list
>Insight-users at itk.org
>http://www.itk.org/mailman/listinfo/insight-users
>  
>



More information about the Insight-users mailing list