Vuescan IR problem - anybody else experience this?

I did a google groups search to see if this has been discussed before but
didn't find any posts. Sorry if this has been discussed before.

I'm having a problem with Vuescan's IR filtering altering the scan where
it's not needed.

Below are links to some scans of a 35mm slide that show the problem.
Actually, the first 3 pictures are just small cropped sections of the full
scan of the slide, but are shown at full scan resolution (2400ppi). The
last picture is of the full slide, but the resolution has been reduced by
75% to make viewing easier.

http://i232.photobucket.com/albums/...622/IR-none.jpg

http://i232.photobucket.com/albums/...22/IR-light.jpg

http://i232.photobucket.com/albums/...622/IR-high.jpg

http://i232.photobucket.com/albums/...2/fullslide.jpg


The first one (IR-none) is scanned without any IR filter. The second (IR-
light) is scanned with the IR filter set to "light". As you can see, the
IR filter has messed with the edge between the white star and the blue
background. This problem shows up regardless of what level of IR filter I
use (the third photo is with IR set to "high").

I'm doing a whole bunch of fighter jet slides and am seeing this same
problem on other slides. I have seen this occur before (on older versions
of vuescan), but sporadically and I was able to work around (just scan w/o
IR and do extra touchup). But it's happening with too great a frequency on
these slides to do that.

I'm scanning using an Epson 4990 Photo flatbed. I'm using Vuescan 8.4.38.
I'm scanning the slides at 2400ppi.

Has anyone else experienced this problem? Any recommendations on
correcting? Thanks!

Would this happen to be Kodachrome? These kind of artifacts are well
known when IR cleaning is used with Kodachrome. As I recall (I
haven't scanned any Kodachrome in a while), the effect appears
somewhat inconsistently, and may be a function of the specific
Kodachrome emulsion, the type and strength of the IR cleaning used,
and the version of Vuescan. Others may be able to provide a more
detailed "solution" for you. If not, I can detail a simple workflow
that I have used for processing scanned images when I just HAD to have
the IR clean on, and these artifacts resulted.

Jeff

On Nov 20, 2:52 pm, me <m...@here.com> wrote:
> I did a google groups search to see if this has been discussed before but
> didn't find any posts. Sorry if this has been discussed before.
>
> I'm having a problem with Vuescan's IR filtering altering the scan where
> it's not needed.
>
> Below are links to some scans of a 35mm slide that show the problem.
> Actually, the first 3 pictures are just small cropped sections of the full
> scan of the slide, but are shown at full scan resolution (2400ppi). The
> last picture is of the full slide, but the resolution has been reduced by
> 75% to make viewing easier.
>
> http://i232.photobucket.com/albums/...622/IR-none.jpg
>
> http://i232.photobucket.com/albums/...22/IR-light.jpg
>
> http://i232.photobucket.com/albums/...622/IR-high.jpg
>
> http://i232.photobucket.com/albums/...2/fullslide.jpg
>
> The first one (IR-none) is scanned without any IR filter. The second (IR-
> light) is scanned with the IR filter set to "light". As you can see, the
> IR filter has messed with the edge between the white star and the blue
> background. This problem shows up regardless of what level of IR filter I
> use (the third photo is with IR set to "high").
>
> I'm doing a whole bunch of fighter jet slides and am seeing this same
> problem on other slides. I have seen this occur before (on older versions
> of vuescan), but sporadically and I was able to work around (just scan w/o
> IR and do extra touchup). But it's happening with too great a frequency on
> these slides to do that.
>
> I'm scanning using an Epson 4990 Photo flatbed. I'm using Vuescan 8.4.38.
> I'm scanning the slides at 2400ppi.
>
> Has anyone else experienced this problem? Any recommendations on
> correcting? Thanks!

> Would this happen to be Kodachrome?

Yes, they do happen to be Kodachrome


> These kind of artifacts are well
> known when IR cleaning is used with Kodachrome. As I recall (I
> haven't scanned any Kodachrome in a while), the effect appears
> somewhat inconsistently, and may be a function of the specific
> Kodachrome emulsion, the type and strength of the IR cleaning used,
> and the version of Vuescan. Others may be able to provide a more
> detailed "solution" for you.

So, if the reason is primarilly related to the type/manufacturer of the
slide, I take it I would likely have the same problem even if I sent the
slides out to one of the big guys like "Dig My Pics"?


> If not, I can detail a simple workflow
> that I have used for processing scanned images when I just HAD to have
> the IR clean on, and these artifacts resulted.


I would love to hear more

>
> Jeff
>

Thanks Jeff!

Ok, here's what I worked out for processing Kodachrome scans, last
time I did this:

1) If necessary, first clean the film (beginning with compressed air,
and moving on to wet cleaning as need be)
2) Scan and save the file as you normally would with IR cleaning ON
(probably set to "light").
3) Immediately re-set IR clean to "none" and then save the non-cleaned
version to a different file name. This is actually very quick - you
just resave the file without the IR clean (it does not require a new
scan).
4) Bring these two files into photoshop as separate layers, with the
IR cleaned version as the top layer, and create a layer mask for the
top layer.
5) Wherever you see IR cleaning artifacts, simply use the brush tool
to paint black onto the mask. This reveals the non-cleaned layer
below which doesn't have the artifacts.

Alternatively, you can reverse the layers in the above process, so
that the IR cleaned layer is on the bottom. In this case you just
mask off the dust spots to reveal the cleaned version below it.

The key is that unless you have a really dirty slide, it is unlikely
that you will have dust spots exactly where the IR artifacts are.
Even if you occasionally do, you can still spot-clean them as you
normally would in Photoshop, but relying on IR clean to do most of the
work.



On Nov 21, 7:18 am, me <m...@here.com> wrote:
> > Would this happen to be Kodachrome?
>
> Yes, they do happen to be Kodachrome
>
> > These kind of artifacts are well
> > known when IR cleaning is used with Kodachrome. As I recall (I
> > haven't scanned any Kodachrome in a while), the effect appears
> > somewhat inconsistently, and may be a function of the specific
> > Kodachrome emulsion, the type and strength of the IR cleaning used,
> > and the version of Vuescan. Others may be able to provide a more
> > detailed "solution" for you.
>
> So, if the reason is primarilly related to the type/manufacturer of the
> slide, I take it I would likely have the same problem even if I sent the
> slides out to one of the big guys like "Dig My Pics"?
>
> > If not, I can detail a simple workflow
> > that I have used for processing scanned images when I just HAD to have
> > the IR clean on, and these artifacts resulted.
>
> I would love to hear more
>
>
>
> > Jeff
>
> Thanks Jeff!

Ok, here's what I worked out for processing Kodachrome scans, last
time I did this:

1) If necessary, first clean the film (beginning with compressed air,
and moving on to wet cleaning as need be)
2) Scan and save the file as you normally would with IR cleaning ON
(probably set to "light").
3) Immediately re-set IR clean to "none" and then save the non-cleaned
version to a different file name. This is actually very quick - you
just resave the file without the IR clean (it does not require a new
scan).
4) Bring these two files into photoshop as separate layers, with the
IR cleaned version as the top layer, and create a layer mask for the
top layer.
5) Wherever you see IR cleaning artifacts, simply use the brush tool
to paint black onto the mask. This reveals the non-cleaned layer
below which doesn't have the artifacts.

Alternatively, you can reverse the layers in the above process, so
that the IR cleaned layer is on the bottom. In this case you just
mask off the dust spots to reveal the cleaned version below it.

The key is that unless you have a really dirty slide, it is unlikely
that you will have dust spots exactly where the IR artifacts are.
Even if you occasionally do, you can still spot-clean them as you
normally would in Photoshop, but relying on IR clean to do most of the
work.



On Nov 21, 7:18 am, me <m...@here.com> wrote:
> > Would this happen to be Kodachrome?
>
> Yes, they do happen to be Kodachrome
>
> > These kind of artifacts are well
> > known when IR cleaning is used with Kodachrome. As I recall (I
> > haven't scanned any Kodachrome in a while), the effect appears
> > somewhat inconsistently, and may be a function of the specific
> > Kodachrome emulsion, the type and strength of the IR cleaning used,
> > and the version of Vuescan. Others may be able to provide a more
> > detailed "solution" for you.
>
> So, if the reason is primarilly related to the type/manufacturer of the
> slide, I take it I would likely have the same problem even if I sent the
> slides out to one of the big guys like "Dig My Pics"?
>
> > If not, I can detail a simple workflow
> > that I have used for processing scanned images when I just HAD to have
> > the IR clean on, and these artifacts resulted.
>
> I would love to hear more
>
>
>
> > Jeff
>
> Thanks Jeff!

DenverDad <the_applegates@comcast.net> wrote in news:35b4cdf3-8e47-4296-
85f1-01b8998998fe@i12g2000prf.googlegroups.com:

> Ok, here's what I worked out for processing Kodachrome scans, last
> time I did this:
>
> 1) If necessary, first clean the film (beginning with compressed air,
> and moving on to wet cleaning as need be)
> 2) Scan and save the file as you normally would with IR cleaning ON
> (probably set to "light").
> 3) Immediately re-set IR clean to "none" and then save the non-cleaned
> version to a different file name. This is actually very quick - you
> just resave the file without the IR clean (it does not require a new
> scan).
> 4) Bring these two files into photoshop as separate layers, with the
> IR cleaned version as the top layer, and create a layer mask for the
> top layer.
> 5) Wherever you see IR cleaning artifacts, simply use the brush tool
> to paint black onto the mask. This reveals the non-cleaned layer
> below which doesn't have the artifacts.
>
> Alternatively, you can reverse the layers in the above process, so
> that the IR cleaned layer is on the bottom. In this case you just
> mask off the dust spots to reveal the cleaned version below it.
>
> The key is that unless you have a really dirty slide, it is unlikely
> that you will have dust spots exactly where the IR artifacts are.
> Even if you occasionally do, you can still spot-clean them as you
> normally would in Photoshop, but relying on IR clean to do most of the
> work.
>

BRILLIANT! Thanks!

On Nov 22, 8:46 am, me <m...@here.com> wrote:
> DenverDad <the_applega...@comcast.net> wrote in news:35b4cdf3-8e47-4296-
> 85f1-01b899899...@i12g2000prf.googlegroups.com:
>
>
>
>
>
> > Ok, here's what I worked out for processing Kodachrome scans, last
> > time I did this:
>
> > 1) If necessary, first clean the film (beginning with compressed air,
> > and moving on to wet cleaning as need be)
> > 2) Scan and save the file as you normally would with IR cleaning ON
> > (probably set to "light").
> > 3) Immediately re-set IR clean to "none" and then save the non-cleaned
> > version to a different file name. This is actually very quick - you
> > just resave the file without the IR clean (it does not require a new
> > scan).
> > 4) Bring these two files into photoshop as separate layers, with the
> > IR cleaned version as the top layer, and create a layer mask for the
> > top layer.
> > 5) Wherever you see IR cleaning artifacts, simply use the brush tool
> > to paint black onto the mask. This reveals the non-cleaned layer
> > below which doesn't have the artifacts.
>
> > Alternatively, you can reverse the layers in the above process, so
> > that the IR cleaned layer is on the bottom. In this case you just
> > mask off the dust spots to reveal the cleaned version below it.
>
> > The key is that unless you have a really dirty slide, it is unlikely
> > that you will have dust spots exactly where the IR artifacts are.
> > Even if you occasionally do, you can still spot-clean them as you
> > normally would in Photoshop, but relying on IR clean to do most of the
> > work.
>
> BRILLIANT! Thanks!- Hide quoted text -
>
> - Show quoted text -

If I could add a suggestion, to save time and avoid mis-alignment:
instead of doing 2 physical scans with Vuescan, output a 64 bit raw
file at the first (and only) scan. Then do a couple of scan-from-
disks: one with and one without cleaning. You can either set your crop
at Vuescan, or later in Photoshop or other editor. In either case be
careful to crop exactly the same size/position.

With my Scan Elite 5400 I find:

1. Vuescan's cleaning is a distant second to my scanner's OEM software
ICE

2. Using the Grain Dissolver, I have no problems with artifacts when
scanning Kodachrome.

Mendel,

You might have misunderstood. The method I suggested involves only
one physical scan. When the scan (WITH IR cleaning) is complete,
simply set IR cleaning to "none" and hit the SAVE button. This saves
a new file which is the "uncorrected" version of the data. As long as
you haven't changed the crop between saves, you don't have to worry
about how you cropped the imges - the two files are exactly the same
except for cleaning.

Jeff

On Nov 27, 10:37 pm, Mendel Leisk <mendel_leisknos...@hotmail.com>
wrote:
> If I could add a suggestion, to save time and avoid mis-alignment:
> instead of doing 2 physical scans with Vuescan, output a 64 bit raw
> file at the first (and only) scan. Then do a couple of scan-from-
> disks: one with and one without cleaning. You can either set your crop
> at Vuescan, or later in Photoshop or other editor. In either case be
> careful to crop exactly the same size/position.
>
> With my Scan Elite 5400 I find:
>
> 1. Vuescan's cleaning is a distant second to my scanner's OEM software
> ICE
>
> 2. Using the Grain Dissolver, I have no problems with artifacts when
> scanning Kodachrome.- Hide quoted text -
>
> - Show quoted text -

Ok, that's essentially the same as scan-from-disk: except you're using
the scan data in ram twice. So, no mis-alignment. Still, it's cheap
insurance to save the raw file, say in case the cleaning level isn't
satisfactory, or you miscrop. And then delete it I suppose, once
you're happy with the outcome.