solution for matching problems

[Stupendous Man <richmond@a188-l009.rit.edu>]

  I finally found time to look at the problem.  Thanks to Tom for
providing some disks with sample data sets which exhibit the problem.
The matching problems are caused by two factors:

       1. the existing reference catalog doesn't contain a great
              many stars in the proper magnitude range.

       2. the pipeline was picking the brightest N stars from the
              reference catalog in the B-band, and using those
              to match against the detected stars in V-band and I-band.
              Whoops.  This one was really my fault.

  By making a small change to the code in "astrom.tcl" and the 
"astrom.param" file, I addressed number 2.  By making a somewhat
larger version of the astrometric reference catalog, I addressed number 1.
After making both changes, I was able to match all V-band and I-band
images in Tom's disk set for 020205, starting with Hxra2310540.fits.

  I'll distribute the fixes in the near future.  In the meantime,
one very quick and easy change which alleviates the problem is 
to change 

               astrom_mag

in the file "astrom.param" from "3" to "4".  That indicates that the
V-band magnitudes from the reference catalog should be used to pick
the brightest N in the area of the TASS image (rather than B-band).
It may permit an extra 50% of the I-band images to pass the matching step.

                                         Michael Richmond