[Insight-users] metric->SetUseNormalization(true);

[M.Staring at lumc.nl]

Hi Motes,

the documentation of that function reads:

 /** Set/Get whether to normalize the mean squares measure.
   * This divides the MeanSquares by a factor (range/10)^2,
   * where range represents the maximum gray value range of the
   * images. Based on the ad hoc assumption that range/10 is the
   * maximum average difference that will be observed. 
   * Dividing by range^2 sounds less ad hoc, but will yield
   * very small values. */

So, normalisation means in this case dividing by the expected (assumed)
maximal difference. Therefore, the range of metric values is not in
[0,1].

hth,

Marius

> -----Original Message-----
> From: insight-users-bounces at itk.org 
> [mailto:insight-users-bounces at itk.org] On Behalf Of motes motes
> Sent: vrijdag 11 december 2009 12:56
> To: Insight Users Mailing List
> Subject: [Insight-users] metric->SetUseNormalization(true);
> 
> I use the metric:
> 
> template < class TFixedImage, class TMovingImage >
> class AdvancedMeanSquaresImageToImageMetric :
>     public AdvancedImageToImageMetric< TFixedImage, TMovingImage>
> {
> public:
> 
> 
> from the elastix library. Its possible to set
> 
>     metric->SetUseNormalization(true);
> 
> 
> But when my registration process I typically get metric values from:
> 1.12079 to 0.32345. Should the range not be strictly in the range
> [0;1] since I have choosen to normalize the values?
> _____________________________________
> 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.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
>