[daip] beam size in pixels

[Eric Greisen]

Casey Law writes:

 > 	I have a simple question that I can't find the answer to in the
 > help files.  I am trying to calculate the beam size in pixels for an
 > arbitrary image.  AIPS does this with IMSTAT, but only for VLA images.  
 > This is probably a header thing, but really I'd just like to know the
 > general form.
 > 	My naive calculation is like this:
 >  pix_area/beam_area = pixsize^2/(pi*bmaj/2*bmin/2)
 > 
 > but this gives an answer that is smaller than the value output by IMSTAT
 > by a factor of 0.69.  For example, in one case, I calculated that the beam
 > area as 25.7 pixels, but IMSTAT says "Beam area = 37.09 pixels".
 > 	Can you tell me the proper way to calculate the beam area in
 > pixels?
 > 
The beam area is the area under a Gaussian integrated to infinity.
This works out to be 1.1331 * Bmaj * Bmin (in say, arc sec ** 2).  To
convert to pixels divide by the cell size squared (in arc sec/cell).

Eric Greisen