Feedbacks about AVM tags in Aladin

August (Gus) Muench gmuench at cfa.harvard.edu
Tue Nov 25 07:20:03 PST 2008


Doug,

according to AVM 1.1 the entire FITS header would be appended as free text in 
the AVM tag, Spatial.FITSheader, XMP tag <avm:Spatial.FITSheader>.  I like to 
think of it as giving a unspecified comment field a semantically useful name.

Best,

	 - gus


Doug Mink wrote:
> It was interesting to hear about Pierre's problems with AVM tags, which
> I had hoped to add support for in WCSTools.  Our default way of dealing
> with WCS in JPEG images--and GIFs and PNGs for that matter--has been to
> simply extract the FITS header using WCSTools imhead -f (which preserves
> all of the blanks and keeps 2880-byte blocks) and append it to the JPEG
> file after other processing (which hopefully preserves pixel locations)
> has been completed.  The WCSTools header reading software knows to look
> at the end of a GIF, JPEG, or PNG file for the FITS header for pixel to
> WCS or WCS to pixel conversion or WCS coverage computation.  This may
> not be a good standard, but it is a convenient convention and is
> independent of AVM standard changing.  In addition, *all* of the header
> information is present, so you can find out other things about the image
> than its WCS.  I added this feature in version 3.6.4, in May 2006, but I
> haven't publicized it very much as it is more a convenience than a
> standard, but at least it is a stable convenience!
> 
> For a future standard, I think it is important to have a way to put the
> entire FITS header, or a reasonably inclusive subset, in any
> astronomical image.
> 
> -Doug Mink
>   Telescope Data Center
>   Smithsonian Astrophysical Observatory
>   Cambridge, MA USA

-- 
August Muench
Research Associate
Harvard College Observatory
gmuench at cfa.harvard.edu
(617) 495-7979



More information about the interop mailing list