Re: JPEG Header Information

From: Doug Tody <dtody-at-nrao.edu>
Date: Tue, 22 Apr 2003 16:24:19 -0600 (MDT)


Wil -

I think this is a good idea - at least I can't see any harm in it. If you do this I suggest trying to pattern it after current FITS practice as much as possible.

The problem with this is that it is format dependent, much software will not make use of or support the headers (as Boyd points out), and we have no VO standards for "external" data formats such as FITS headers. An example of an alternative approach is what is done in SIA. Here we assume that external data objects will have nonstandard metadata or no metadata at all, and we use a VOtable to pass the standard metadata for any image in a uniform way. It is less convenient and not as simple, but probably more reliable and allows the metadata to be standardized.

On Wed, 16 Apr 2003, Wil O'Mullane wrote:

> We have a lot of JPEG images about which lack anykind of infomration in them.
> For the SDSS images we are thinking of adding infomration to a comment block.
> Perhaps this would be good in other instances also.
>
> Currently we are thinking of adding something like
> Author="SDSS Collaboration "
> Info="http://skyservice.jhu.edu/develop/jpeghead/index.html"
> TITLE="SDSS Image "
> CRPIX1=744.5
> CRPIX2=1024.5
> CRTYPE1=RA--TAN
> CRTYPE2=DEC--TAN
> CRVAL1=179.948240689264
> CRVAL2=0.101381194295879
> CDVAL1=-3.8008997158114E-08
> CDVAL2=0.000109924630958157
> CDVAL3=0.000110062460780114
> CDVAL4=-4.78845056498477E-08
> NAXES=2
> NAXES1=2048
> NAXES2=1489
> EPOCH=J2000
>
> or basically the fits keywords.
>
> Slightly more info at http://skyservice/develop/jpeghead/index.html
>

> wil
Received on 2003-04-23Z10:27:27