psf matching and negative bits around stars
Stin. Kerr wrote on May 01, 2009
Hi all,
I'm using psfmatch -- pretty much out of the box, which is likely to be the problem.
OK, I'm doing the obvious stuff, but around my brightest stars, I get small sections of negative values. They are always offset from the centre of the stars either due-x or due y.
I'm using convolution mode = image.
Does anyone have any idea about what the origin of this problem could be? Right now, figuring it out seems beyond me.
Cheers, and thanks in advance!
I'm using psfmatch -- pretty much out of the box, which is likely to be the problem.
OK, I'm doing the obvious stuff, but around my brightest stars, I get small sections of negative values. They are always offset from the centre of the stars either due-x or due y.
I'm using convolution mode = image.
Does anyone have any idea about what the origin of this problem could be? Right now, figuring it out seems beyond me.
Cheers, and thanks in advance!
Ray Butler wrote on May 01, 2009
It's been several years since I used psfmatch, so from memory...I expect that what is happening is due to the limitations of its use of *pre-extracted* PSFs from somewhere else (you're probably giving it PSFs from DAOphot?). Have a close look at your PSF...does it have any negative residuals?
Not to disrespect the IRAF/immatch package - which pioneered the way in 1995 - but you can get significantly better results for this kind of image-matching and subtraction work if you use Alard's ISIS package, first released in 1999. ISIS is inherently both more accurate and based on a smarter, more integrated algorithm. It's the industry standard. Although it's stand-alone, not an IRAF addon package, you can feasibily embed calls to it from within IRAF scripts if that's what you want to do.
Not to disrespect the IRAF/immatch package - which pioneered the way in 1995 - but you can get significantly better results for this kind of image-matching and subtraction work if you use Alard's ISIS package, first released in 1999. ISIS is inherently both more accurate and based on a smarter, more integrated algorithm. It's the industry standard. Although it's stand-alone, not an IRAF addon package, you can feasibily embed calls to it from within IRAF scripts if that's what you want to do.
Last post on May 01, 2009