I have completed a test case study of a technique for extracting radiance values out of ROLO lunar photometry image.  It is illustrated in a PDF file - http://members.csolutions.net/fisherka/astronote/observed/LCROSS/20090927mps...   http://tinyurl.com/yaub5sb In summary, an image processing protocol to extract camera values for a 4 x 3 pixel region-of-interest near Cabeus A from a 512 x512 ROLO archive lunar image taken 2001-02-13 UT09:59 at 550nm from the ROLO facility near Flagstaff, Arizona is illustrated. The median radiance of the 12 pixel region-of-interest near Cabeus A is reduced as 0.0456363 Watts/m^2 sr nm with a standard deviation of 0.00189995 Watts/m^2 sr nm (4.1 %) at 550nm. Project materials, including the Mathematica 7 file used to perform image analysis and reduction, can be found in directory: http://members.csolutions.net/fisherka/astronote/observed/LCROSS/20090927mps... Part 2 of this process will be to convert the radiance value extracted from the ROLO image into a stellar V-magnitude and magnitudes-per-square-arcsec (mpsas).  The pixels in the ROLO camera are 4.08250 arcsecs per pixel.  The high-bound LCROSS apparent brightness estimate is 4.0 mpsas. I'd appreciate the group's comments on the proper algorithm for the next step to get from radiance (Watts/m^2 sr nm) to mpsas. Congrats on Selenology Today 16. Clear Skies - Kurt