iraf.net

HiBefore I begin it must be noted I am rather new to Iraf.I have been using another program that is not iraf (in fact its a task in starlink's kappa called "setmagic" just in case anyone has heard of it) inorder to set any pixels in a fits image with a value of 0 to what it calls bad (or magic) and assumed iraf would also understand the bad value it set. Indeed when the image is opened with Gaia (the graphics utility I use) these pixels are displayed as "blank" or "nan". I am now concerned, having been using stsdas's ellipse task for sometime on the images, if iraf is in fact interpreting the pixels as bad or flagged as ellipse calls it. Can anyone enlighten me on a task in iraf to determine how many "bad" pixels it is detecting in an image? and also, if possible, does anyone know how ellipse deals with "bad" pixels (ie are they flagged when it is doing its fitting).CheersJames

From what little I understand of the ELLIPSE task, the "bad" pixels are determined using a pixel mask having the same root name as the image, e.g. the image 'foo.fits' would use the mask 'foo.pl' in the same directory. These pixels should be excluded automatically, but you may want to contact STScI with detailed questions about the algorith (help@stsci.edu) if you don't get a response here.To create the pixel mask you could use a task like IMEXPR, e.g. to create a mask where all of the zero values are flagged as the bad pixels try[code:1:4e0408f9eb]cl> imexpr "a == 0" foo.pl a=foo.fits[/code:1:4e0408f9eb]See the help page for details, but in general the blank/nan values aren't automatically understood to be "bad" values in IRAF in the same sense that starlink/gaia use them.Cheers,
-Mike