----- Original Message -----
Sent: Tuesday, November 06, 2001 8:07
PM
Subject: RE: [APML] Scaling Images,
javascript toggle
Matt,
Jpeg compression is spatially
lossy. It carves the image into 8x8 pixel
blocks
and then does a fourier transform on each, keeping the low order
large amplitude coefficients and discarding the small
amplitude, high
frequency ones. The result is
that an exact pixel coordinate of a point
source, like
a star, will not be preserved within the 8x8 block, depending
on how much gets thrown away.
If one jpeg is larger, it indicates that more of the detail is
retained.
I would expect its positions to be more true
to the original. I don't
know what the Genuine
Fractal algorithm does beforehand, so it's not
clear
which is "truer".
BTW, I agree that your image toggle is a very powerful
presentation
tool. It inspired me to assemble a
pair of images, not nearly so dramatic
as your
day-night scenes, but intriguing in its own way. Here is
a link to Many Glacier Lights, ala BenDaniel:
http://nightscapes2.home.att.net/photos/startrails/OdysseyTrails/ManyGlacierLightsOnOff.html
Thor
Thor
Olson
Color Imaging
Scientist
voice: +1 651 365 5252
EFI-Minnesota
email: mailto:thor.olson@efi.com
web-res astrophotos at:
http://home.att.net/~nightscapes
"Man has ten, computers have two, Nature has e
fingers."
-----Original Message-----
From: Matt
BenDaniel [mailto:matt@starmatt.com]
Sent: Tuesday, November 06, 2001 9:49 PM
To:
astro-photo@seds.org
Subject: Re: [APML] Scaling
Images (was: Cocoon Nebula)
Mike,
I used Photoshop JPEG quality=8 for both renditions.
In GF I used their lossless STN encoding, which purports to
capture and retain 100.0% of the information in the image.
JPEG *does* affect pixel layout. It is obvious in Photoshop
Save As JPEG Dialog and toggle the preview on and off. It is more obvious at
lower quality levels. It helps to zoom to 200% to see the subtle changes when
higher quality is used.
The web image toggling capability is very useful, huh? You can
do the same thing in PS by switching layers on and off, but it's hard to show
that to someone else on the internet.
Matt
At 10:40 PM 11/6/2001 -0500, Mike Cook, AF9Y wrote:
> This is really interesting. What
settings did you use for the
>jpeg and GF
compression? I can see a grid in both images,
>but less so in the GF. The grid lines appear to be the
point
>of shifting. I thought that jpeg
compression was just
>in the color data and
produced no image displacement.
>
>
>Date
sent:
Tue, 06 Nov 2001 22:26:27 -0500
>To:
astro-photo@seds.org
>From:
Matt BenDaniel <matt@starmatt.com>
>Subject:
[APML] Scaling Images (was: Cocoon Nebula)
>Send
reply to:
astro-photo@seds.org
>
>> At 06:16 PM 11/6/2001 +0100, Mikkel Steine wrote:
>> >...I know from my days in desktop
>> >publishing that Photoshop is not always the
preferred tool for scaling images.
>> >There
are numerous techniques for resampling an image, and it would be fun to
>> >try out some of them...
>>
>> Mikkel,
>>
>> I downloaded the free
evaluation version of Genuine Fractals. Using GF I scaled the image
>> down to the exact same size (703x640) as my PS JPG
and uploaded it to my web site:
>>
>> http://people.ne.mediaone.net/mbendaniel1/gallery/astro/cocoon.html
>>
>> You can toggle to
the PS scaled by moving mouse over image.
>>
>> The pixels patterns resulting from GF are
slightly different than Photoshop bicubic. The
>> GF stars are sharper and less smeared. The JPG size of the GF
image is larger, which may
>> be an indication
that GF preserves more information from the original image.
>>
>> It is very curious how the
image annoyingly shifts around when you toggle between the two.
>> Could PS or GF be screwing up its scaling? -- Matt
BenDaniel http://starmatt.com
>>
>>
>> -- APML Archives at <http://astro.umsystem.edu/apml/>
---
>>
Unsubscribe at <majordomo@seds.org>
>>
>
>
>
>Mike Cook, AF9Y
>http://www.webcom.com/af9y
>mwcook@cris.com
>
>
>
>
>-- APML Archives at <http://astro.umsystem.edu/apml/>
---
>
Unsubscribe at <majordomo@seds.org>
--
Matt BenDaniel
http://starmatt.com
-- APML Archives at <http://astro.umsystem.edu/apml/>
---
Unsubscribe at <majordomo@seds.org>