Reducing an existing catalog

[Jeffrey McCall]

Is there a way to take an existing astrometry.net index file and reduce its size by filtering out stars below a certain magnitude threshold? For example if I have an astrometry.net index file that contains stars up to magnitude 16, is there a way to filter out stars that have a magnitude > 10 so that I can drastically reduce the amount of stars in the file? I did try to see if there was a way to use the "build-astrometry-index" command to do this, but based on what I saw in the code it looks like if you use the "-B" option and also pass an index file with "-1", the code ignores the "-B" option completely. 

As a side note, I don't have access to the original catalog files used to create these index files.

If I can't do that, is there a way to take an index file that is at scale 4 and has a lot of stars and rebuild it with a much higher scale value like 19? When I've tried to do that, I get errors like the following:

.quad-builder.c:167:quadbuilder_create: quad-builder: failed to malloc qb->inbox or qb->pquads.  Nstars=827813

It looks like it's trying to allocate a huge amount of memory. Is there a workaround for this error?

I'd appreciate any help that anyone could provide with this.

Jeff M.

[Dustin Lang]

Hi,

The index files are built to contain a bright subset of stars selected at a given scale, AND "features" (micro-constellations) with a given range of scales.  So you can't just drop stars, you would also need to select new features at a new set of scales, likely.

So I think your question boils down to, if you only have an index file, can you get back to a list of RA,Dec,mag, from which you could build a new index file.

One issue with trying to do this: index files don't always *have* the magnitudes of the stars, that's an optional thing.

The RA,Dec positions are encoded in a rather complicated way, so reading them straight out of the index files will be a pain.

You might try using the "query-starkd" program to "query" all the stars in the index -- eg, query for RA,Dec of 0,0 with a radius of 180 degrees, or something like that.  I'm not certain that will work, but it might be your best bet.

cheers,

dustin

--
You received this message because you are subscribed to the Google Groups "astrometry" group.
To unsubscribe from this group and stop receiving emails from it, send an email to astrometry+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/astrometry/405632bd-8d5b-45b1-aa45-bf1f5b1dd241n%40googlegroups.com.