Re: [Gimp-user] Fw: scan-mirrage-effect-Problem with postcard

[Gene Heskett <gheskett shentel net>]

On Monday 10 May 2021 23:55:17 Liam R E Quin wrote:

> On Mon, 2021-05-10 at 23:22 -0400, Gene Heskett wrote:
> > The moire pattern is without a doubt, from scanning a litho print
> > which
> > is microscopic patterns of different size colored dots that mix in
> > the
> > eye to make the color. 
>
> Did you look at the image?

No, I just got interested in the thread as it seemed to be going on for a 
longer time than most, and the fact that I've shared chemical darkroom 
space with a print shop so I am somewhat familiar with the tech that 
probably printed that postcard.

Whats the date/time of the post containing the image?

And, most importantly, is that image raw, uncompressed staight out of the 
scanners cable, very unlikely, or has it been passed thru a compression 
utility, in the latter case all bets are off. Simply put, any scan good 
enough to publish out of my scanner will be jpegged or png'd because the 
raw scan is going to be half a gigabyte or more, and will be compressed 
to under 300k base64'd for transport over the net. gimp is very good at 
that, but it is not without lots of artifacts to the trained eye.

> Although there's evidence of dot screen 
> atrefacts, that'snot what is being referred to.  In addition, the
> stripes change direction on the second scan, suggesting they are not
> coming from the actual postcard.

in that event, I'd blame that on sloppiness of the scanners head 
transport linkage. A .1mm inconsistency of square would change a moire 
pattern, a lot. or moving the card on the glass by replacing it on the 
glass for the 2nd scan. Even if the lid is not raised, the scanner could 
easily displace the second scan enough just from temperature effects to 
change the pattern. Unsynchronized data clocking would also see to that. 
In fact combining multiple scans with a 32 or 64 bit addition, and 
decimating it back to 16 or even 8 bit depth by throwing away the lsbits 
might well be the cure because it would cancel the moire pattern by 
averaging the sums. But thats not practical to do on a large scale with 
todays computers, but when we get 256 bit cpu's and terabytes of fast 
ram it will be routine.

Astronomers with access to supercomputers do that stacking and 
synchronizing of images all the time to Hubbles images to get the 
amazing pix we get out of it. That scope is a time machine, looking back 
in time as much as 13.8 billion years. To a time when the universe was 
new, maybe 200,000 years after the big bang. We can't see any further 
back because it was opaque before then. It had to cool a while before 
light became uncoupled and could escape to begin its journey to Hubbles 
single photon detectors.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page <http://geneslinuxbox.net:6309/gene>