Re: Scanning for Stitching with Vuescan

From: Dave Soemarko ^lt;>
Date: 02/21/04-12:13:52 PM Z
Message-id: <003901c3f8a6$76e7fb00$9729fea9@W>


Try to place the print in such a way so that in both passes there is an
overlapping area and this area is physically on approximately the same area
on the printer. This will ensure the exposure will be very close.

And in photoshop, don't try a perfect stitch. Place the images on two
different layers. After aligning them, select (from the top layer) the whole
image up to the halfway through the overlapping area), then soften the
selection (I used to do this a lot but it has been about 5 years now so I
can't remember the exact term in Photoshop, but basically you don't want to
selection to be a stiff rectangle but a soft rectangle); then set this layer
to be transparent. On one side you have only the top layer, the bottom layer
is blank, so setting the top layer to transparent will get you the original
image. In the overlapping area, however, since it is a soften selection, it
become a soften transparency, so the area nicely blend into the bottom
layer. After everything looks right, you can merge the two layers (to make
the file smaller).

As I said, I had done that a lot and never had any problem of visible
density difference. I don't know if I explained it clear enough though. If
there is any part that is not clear above, please ask again, and I will try
to describe. It's hard to describe this kind of thing with words, sometimes.

Dave S

----- Original Message -----
From: "Sandy King" <>
To: <>
Sent: Saturday, February 21, 2004 10:56 AM
Subject: Scanning for Stitching with Vuescan

> I am using the Vuescan software to scan 12X20 images with my Epson
> 836XL 12X17 scanner. This has to be done in two passes and I stitch
> the files together in Photoshop.
> I am having a problem. After the first scan I lock the exposure down
> so that in theory there should be no difference in density between
> the first and second scans. In practice, however, there is enough
> difference in density between the two scans so that when they are
> stitched together one side is visibly lighter than the other.
> Since there are a few Vuescan users on this list I am hoping that
> someone with more experience with the software than me can explain
> what I need to do to get equal exposure on both scans.
> Sandy King
Received on Sat Feb 21 12:14:14 2004

This archive was generated by hypermail 2.1.8 : 03/02/04-11:35:09 AM Z CST