image orientation on the screen

[Phil Hodge]

Michael Newberry (mnewberry at axres.com) writes:

> It would be really useful if there were two keywords that would go
> into an image header at the point of origin that could later be used
> to give the correct display in the "mirror" sense. This would leave
>	...
> Personally, I like XREVERSE and YREVERSE.

I don't think it's a good idea to create FITS keywords which depend
on the convention for displaying images.  A separate keyword for each
axis is also not very useful for images taken from spacecraft.  The
parity of an image is easily obtained from the sign of the determinant
of the CD matrix (or the PC matrix multiplied by the CDELTi, in Greisen
and Calabretta's notation).  The interpretation of the sign depends on
how you display the image, of course.  The sign is negative for images
that will be displayed correctly (i.e. without mirror reversal, but
possibly rotated) if the first pixel is at the lower left, with the
sample direction running from left to right.

						Phil