Re: [Phys-l] Galaxies and clustering in astronomy

[John Denker <jsd@av8n.com>]

On 05/14/2011 12:27 AM, alex brown wrote:
> the actual problem is......  circle recognition in an image from a
> fabry perot.  I have used hough transform and my 3d array is in
> "hough space." The array is sparse and thus I thought it is analogous
> to the problem of stars and assigning them to galaxies.  I don't want
> to reinvent the wheel so....


The general topic of "cluster analysis" has been around for
a very long time, although not quite as long as those thingies 
that go round and round on axles.  See for example
  http://en.wikipedia.org/wiki/K-means_clustering
and references therein.

Some of the implementations are optimized for sparse data.
  http://www.google.com/search?q=k-means+sparse

More specifically we have:
  http://www.google.com/search?q=%22hough+space%22+clustering

If none of that is what you wanted, please ask a more specific 
question.