Thanks, Bill, for your answer, and for the interesting improvements you announce for the future, concerning new possibilities in contrast/brightnes adjustment scheme.

Cheers,

Bertrand

----- Original Message -----
From: Bill J Gray
To: guide-user@yahoogroups.com
Sent: Saturday, July 30, 2005 6:50 PM
Subject: Re: [guide-user] Contrast of images


Hi Bertrand, all,

(I've been away for the last week, and am catching up on some e-mails.)

As Denis mentioned, the beta version does fix some of these problems.
And as Bertrand mentioned, there are some odd complications in all of this.

You can't (usually) get decent contrast or histogram data from a FITS
header. The beta version gets such data from the header when it can;
otherwise, it gets about a thousand semi-randomly selected pixels from
the image and gets quite a good estimate of the distribution that way.
(At present, only the RealSky images extracted from the second CD-ROM
have the right sort of contrast data in the header.)

I did want all of the CD-based images to be simultaneously adjusted,
since (usually) they have similar pixel values. However, there are some
regions where the pixel values are wildly off from what they ought to be.
I may further revise the contrast/brightness adjustment scheme to allow
for this. (I'd also like to include schemes such as logarithmic vs.
square-root vs. histogram-equalization, and maybe some pseudo-color
effects.)

-- Bill


[Non-text portions of this message have been removed]