imarith and chpixtype
ifriad wrote on Oct 27, 2008
Hi all,
I am trying to change the pixel type using imarith and chpixtype,
but I get the following error ERROR (502, "floating point invalid operation")
when I try and change the pixel type to short.
I got a set of images in three bands J, H and K, the H and K works fine but the J not.
One more strange thing the same code works well in other platforms.
I am trying to get it working in Ubuntu.
Thanks For any idea.
I am trying to change the pixel type using imarith and chpixtype,
but I get the following error ERROR (502, "floating point invalid operation")
when I try and change the pixel type to short.
I got a set of images in three bands J, H and K, the H and K works fine but the J not.
One more strange thing the same code works well in other platforms.
I am trying to get it working in Ubuntu.
Thanks For any idea.
Mike Fitzpatrick wrote on Oct 27, 2008
You're changing your image to type short, but from what? Using IMARITH, what is the expression you're using? Is the 'calctype' parameter set explicity? How? What version of IRAF is this? Are these FITS files?
I can't really say what the problem is without seeing the data and knowing specifically how you called the task. It could be something peculiar to specific data values (e.g. fluxes of 10e-14 or so), or perhaps a bug in the FITS kernel when scaling the data to type short. Please post a URL to the problem J image, or feel free to put it in the anonftp archive on ftp://iraf.noao.edu/pub.
-Mike
I can't really say what the problem is without seeing the data and knowing specifically how you called the task. It could be something peculiar to specific data values (e.g. fluxes of 10e-14 or so), or perhaps a bug in the FITS kernel when scaling the data to type short. Please post a URL to the problem J image, or feel free to put it in the anonftp archive on ftp://iraf.noao.edu/pub.
-Mike
Last post on Oct 27, 2008