ðòïåëôù 


  áòèé÷ 


Apache-Talk @lexa.ru 

Inet-Admins @info.east.ru 

Filmscanners @halftone.co.uk 

Security-alerts @yandex-team.ru 

nginx-ru @sysoev.ru 

  óôáôøé 


  ðåòóïîáìøîïå 


  ðòïçòáííù 



ðéûéôå
ðéóøíá












     áòèé÷ :: Filmscanners
Filmscanners mailing list archive (filmscanners@halftone.co.uk)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[filmscanners] Re: JPEG2000 - is it lossless?



Julian,

Thanks a bunch for the workflow.

But looking at the workflow, I think it might also be usefull to merge an
underexposed image with an overexposed one.

Cheers,
Andre


----- Original Message -----
From: "Julian Robinson" <jrobinso@pcug.org.au>
To: <am1000@videotron.ca>
Sent: Monday, February 03, 2003 4:19 PM
Subject: [filmscanners] Re: JPEG2000 - is it lossless?


Andre,

>So for me, JPEG2000 is okay. Only one question remains: is it truly
lossless?

Yes!  Try this (same as what Robert said, but in a little more detail, for
a windows machine).

Using Photoshop,  have the original (say TIFF) image in one window and open
the compressed file into another.

Copy and paste the compressed image into the original window. It will then
be a second layer above the original layer.

i.e.    /Select the compressed image (ctrl-A), copy (ctrl-C), then paste
(ctrl-v)
         /into the original window

Select the top layer, change the mode from 'normal' to 'difference' (drop
down at top left of layers palette).

You will get a largely black result. If the two images are identical, the
result will be exactly and entirely black.  Any non-black pixels are
differences between the two images.  If they are not identical but close,
you probably will not be able to see the differences (it will look all
black) but you can use the histogram to quantify the difference.

Merge these two layers. (layers / merge visible, or flatten, or
shift-ctrl-E) so you have a single layer image.

Now edit / histogram.  This will show you the stats and distribution of the
'difference' image.  If the images are identical you will read 0 for the
mean.  ANything other than 0 means the original two images were not
identical.

The way my machine is set up, I cannot see the faint difference image at
the black end, so I invert the resulting difference image, and can see a
faint near-white image which represents the areas of difference.

When I tried the j2k plug-in, the lossless image was around 50% of the
original in size, but was indeed lossless as tested by this method.  Try a
lossy save and you will see the result is obvious using this method.

Julian



Julian
Canberra, Australia

Satellite maps of fire situation Canberra and Snowy Montains
http://members.austarmetro.com.au/~julian/cbfires/fires.htm

----------------------------------------------------------------------------
------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title
or body

----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe 
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or 
body



 




Copyright © Lexa Software, 1996-2009.