[Insight-users] Can metrics handle nodata/void data values ?

[Luis Ibanez]

Hi Carolyn,

I probably missunderstood what you meant by "void" points,
or "void" pixels. I assumed that those were pixels where the
data is not valid. For example is the image is the result of resampling
another image, and some areas didn't got any pixel data. That's common
with ultrasound images.

Could you explain why the "void" pixels will keep changing ?

Shouldn't they be the same during the entire registration process ?



Thanks


   Luis


-----------------------------
Carolyn Johnston wrote:

> Hi Luis,
>
> I think that the itkSpatialObject solution would solve only half of 
> the problem. A mask object could be constructed using the void data 
> points of the fixed image, but the metric would also need to be aware 
> of the void points of the moving image, the location in space of which 
> would be changing all the time. Of course, it is the void points in 
> the moving image which cause problems with mean squares metric 
> normalization anyway, since as the moving image moves, the number of 
> void points in the fixed image region  (at least in the case of the 
> mean squares metric -- we could probably make it work for the 
> normalized correlation metric).
>
> So I will  keep thinking about this. I hope your busy week isn't too 
> crazy.
>
> :) Carolyn
>
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk . org
> http://www . itk . org/mailman/listinfo/insight-users
>