[Insight-users] Agglomerative Clustering?

[David Doria]

I don't see anything in the ITK software guide about agglomerative (or
divisive) clustering. I would like to take a set of N-d points, treat them
each as their own cluster, and then merge clusters until some "continue
merging?" tolerance is met. I don't know the number of clusters so I can't
use K-means clustering and this type of clustering has been my fall back -
but the memory usage of my implementation grows quickly out of control. Does
ITK have anything like this?

Thanks,

David
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20090728/3d547e09/attachment.htm>