[Insight-developers] itkVectorNearestNeighborInterpolateImageFunction

[Karthik Krishnan]

Bill Lorensen wrote:

> Karthik,
>
> While trying to track down a failing test, I noticed that the output 
> of ResampleImageFIlter9Test looked strange. It uses the subject image 
> function to do nearest neighbor interpolation of an RGB image. A 
> pattern of zero pixels is noticeable in the baseline test image. 

Yes. You are right. Nasty little bug that was:
http://www.itk.org/cgi-bin/viewcvs.cgi/Code/Common/itkVectorNearestNeighborInterpolateImageFunction.txx?root=Insight&r1=1.4&r2=1.5

> Upon further investigation, I looked at the code in the subject 
> filter. I'm not sure why it is so complicated. Why doesn't it use the 
> same logic as itkNearNeighborInterpolateImageFunction. If I drop that 
> code into the Vector code, the resulting image looks appropriate.
> Any ideas?

That's true. I was just borrowing code from the 
VectorLinearInterpolateImageFunction at some point and missed the 
functionality provided in the superclass.

http://www.itk.org/cgi-bin/viewcvs.cgi/Code/Common/itkVectorNearestNeighborInterpolateImageFunction.txx?root=Insight&r1=1.5&r2=1.6

Its been replaced and the baselines replaced. They look good now.

Thanks again
karthik

>
> Bill
>
> _______________________________________________
> Insight-developers mailing list
> Insight-developers at itk.org
> http://www.itk.org/mailman/listinfo/insight-developers
>