RE: Calibrating Scanning for Preservation? Really?

Just to clarify – for other posters (hopefully). By having a Calibrated System – and the operant word here is SYSTEM – I mean all of it. Color Management is certainly an important part – but to give an example – a calibrated system would ensure that the resolution is the same across the ENTIRE imaging area – just because a scanner says it is x by y resolution does not mean necessarily that the same exact resolution is the same across all areas of the image – optics can have a huge role in that – there can be all sorts of distortion – geometric – chromatic and so forth. The same can be said of the bit depth – 48 bits for example doesn’t  tell you anything about the gamma curve of the imaging chip. How are those 48 bits worth of information distributed over the color spectrum space – are they closer together in some areas then others – or are they linear from the brightest bright to the lowest low – and how are those defined anyhow? No I mean a truly calibrated SYSTEM. If you don’t do that – and have metadata to support that – and the delta’s that might exist – then I don’t know how you can call it preservation scanning in the first place because you really don’t know what you have. Frankly, just because Epson or Nikon tells me that a product does such and such in a lab – maybe once – does not mean that the specific product that I am holding in my lab has the same results at all, and even if it did – unless I can prove that – where am I? It isn’t preservation at all. Just a scan – maybe good – maybe not – but certainly not preservation. Do I have it wrong somehow??

 

Jim Lindner

Media Matters LLC. is a technical consultancy specializing in archival audio and video material. We provide advice, analysis, and products to media archives that apply the beneficial advances in technology to collection management.

 


From: Archives & Archivists [mailto:[log in to unmask]] On Behalf Of Blake, Tom
Sent: Thursday, June 15, 2006 10:48 AM
To: [log in to unmask]
Subject: Re: Calibrating Scanning for Preservation? Really?

 

I'm only speaking from the perspective of digitizing still images, but I believe that the one crucial point that you are missing is that a good color management (not calibration) workflow can compensate for many of the idiosynchracies that you describe. Creating a good input profile using appropriate software and standard targets is an important first step in characterizing any imaging system. Scanning/shooting a target with a particular system under particular conditions will produce a file that can then be compared to the known values of that target. The profile that is created from this procedure basically says, "this is the green I'm getting, but this is the green I'm supposed to be getting, so here are the corrections that need to be made to achieve that target green." If you are concerned about your devices' performances fluctuating over time, create a new input profile every month, week, or more appropriately -- with every change that you make to the system and its material subjects. The visual corrections that you make on a raw scan to an 8 x 10 glossy photograph will be very different than the corrections you will need to make to a 4 x 5 transparency. Albumen prints require different corrections than resin-coated black and white papers. All of these visual corrections theoretically should be described in your input profile. Realistically, you set up for a batch of materials that all seem to be similar and then all of a sudden you get to one that just doesn't look right on screen (see metamerism) and you have to tweak your curves or levels or whatever to make it look right -- and now you're not using the same settings that your input profile describes anymore. That's when we ask ourselves, "what the heck is the goal of 'preservation scanning' anyway?" Is it to capture and save the purest, rawest data without any subjective determinations about what the thing should look like? If that's the case, then all digital captures should be performed the same way using a linear response and that looks downright ugly to the human eye/mind. Or is it to preserve the appearance of something that we all see a little differently anyway (the time of day and your mood can actually affect the way you see color -- not to mention the ~10% of men who perceive color inaccurately) and might even change its appearance by the time we get around to looking at the file anyway (yellowing papers, migrating color dyes). And then you go through all this trouble to make the thing look like the original on your calibrated screen in your strictly color-managed environment and it goes off onto the web where people are going to look at it with old CRT monitors in rooms awash in daylight and office fluorescence. So that's really the problem when it comes to talking about "preservation scanning" in my mind. Somebody someday will want to know what the thing actually looks like and that determination -- even when made in a measurable environment -- had to be made by somebody's subjective physiological response to a small range of wavelengths along the electromagnetic spectrum.

-Tom


-----Original Message-----
From: Archives & Archivists on behalf of James Lindner
Sent: Thu 6/15/2006 9:33 AM
To: [log in to unmask]
Subject: Calibrating Scanning for Preservation? Really?

Not calibrating just scanners - but calibration of the entire process - as
in a system approach...



I have been working on turning the video migration process into a calibrated
process for 4 years. It hasn't been easy.. My personal interests and
curiosity has started pulling me into areas of document scanning -
multi-spectral and otherwise. It seems to me that while there are many
people scanning all sorts of things - that in almost all cases there really
isn't a calibrated process, and that in fact people are relying on the
software within a scanner to "calibrate" the process - when in fact the
software was never really designed to do that - and without an outside and
known reference it is highly unlikely that any such process would really
work or be accurate. Even things as obvious as color temperature - and even
the generation of meta-data associated with the scanning process itself is
pretty flimsy when you really crawl into it. For example - just because a
scanner from vendor X says the lamp is 6400 Kelvin does not mean that ALL of
them are, and there certainly can be variations between different lamps and
even the response over time and ambient temperature can change a lamps
performance over time. So how precisely is that color temperature measured
before / during a scan and at the point of imaging - meaning that sometimes
the lamps themselves expose the object through a platen glass which itself
will of course change color temperature.



Even issues of sensor calibration - can it be said that ALL sensors are
exactly linear with the same response curves? I know this is not the case by
just looking at information that the companies that make the imaging chips
provide.



Which leads me to wonder - it seems that a great deal of imaging is going on
with people using an over the counter consumer type scanner with a process
that is totally uncalibrated. Sure the software tells you it is "calibrated"
but that is nothing more then a simply white balance measured by a histogram
- but that is certainly not a calibrated system the way I think of one.



Am I totally wrong here? Have I missed a few crucial bits of information or
understanding - or even missed the point? Otherwise - it seems to me that
the efforts made heretofore in terms of preservation scanning are really
questionable and allot of work needs to be done in this area before
preservation quality scanning is really something that the field can move
forward with.



Please tell me how I have gone astray in my thinking.



jim



Jim Lindner

*       Email:  <mailto:[log in to unmask]> [log in to unmask]
*       Media Matters LLC.
*       Note New Address: 450 West 31st Street 4th FL
New York, N.Y. 10018
*       eFax (646) 349-4475
*       Mobile: (917) 945-2662
*       www.media-matters.net <http://www.media-matters.net/>

Media Matters LLC. is a technical consultancy specializing in archival audio
and video material. We provide advice, analysis, and products to media
archives that apply the beneficial advances in technology to collection
management.



  _____ 

From: Archives & Archivists [mailto:[log in to unmask]] On Behalf
Of [log in to unmask]
Sent: Wednesday, June 14, 2006 10:11 PM
To: [log in to unmask]
Subject: Re: Digitizing Glass Plates



or.... you can scan a hole bunch and then use something like irfan view to
do a batch convert.....




Subj:Re: Digitizing Glass Plates
Date:6/14/2006 6:39:59 PM US Mountain Standard Time
From:[log in to unmask]
To:[log in to unmask], [log in to unmask]
Sent from the Internet



I once had a vendor tell me he could digitize many, many photographs per
hour.  It took a fair amount of work to explain why a rotary scanner was not
appropriate for glass negatives and lantern slides.

You'll want to put the images on the scanner emulsion side down -- in
contact with the platen.  However, you may wind up with the image laterally
reversed.  (Maybe some scanners automatically reverse the image when doing a
transparency.)  Be sure to check your model.

-- Richard Pearce-Moses






----- Original Message -----
From: [log in to unmask]
Sent: Tuesday, June 13, 2006 7:59 PM
Subject: Re: Digitizing Glass Plates


Get a scanner that goes up to 4x5 negatives or traspanances.....   the
instructions that come with the scanner will be all you need.

Ed Sharpe archivist for SMECC





Good question Allaina; and could anyone tell me where I can find
instructions for digitizing glass lantern slides?

Thanks,
Karen



A posting from the Archives & Archivists LISTSERV List sponsored by the
Society of American Archivists, www.archivists.org. For the terms of
participation, please refer to
http://www.archivists.org/listservs/arch_listserv_terms.asp.
To subscribe or unsubscribe, send e-mail to [log in to unmask] In
body of message: SUB ARCHIVES firstname lastname *or*: UNSUB ARCHIVES To
post a message, send e-mail to [log in to unmask]
Or to do *anything* (and enjoy doing it!), use the web interface at
http://listserv.muohio.edu/archives/archives.html
Problems? Send e-mail to Robert F Schmidt <[log in to unmask]>






Thanks,

Ed Sharpe, Archivist for SMECC

See the Museum's Web Site at  www.smecc.org

We are always looking for items to add to the museum's display and ref.
library  - please advise if you have anything we can use.

Coury House / SMECC
5802 W. Palmaire Ave.                          Phone    623-435-1522
Glendale Az 85301  USA




CONFIDENZIALE: Questo messaggio e gli eventuali allegati sono confidenziali
e riservati. Se vi è stato recapitato per errore e non siete fra i
destinatari elencati, siete pregati di darne immediatamente avviso al
mittente. Le informazioni contenute non devono essere mostrate ad altri, né
utilizzate, memorizzate o copiate in qualsiasi forma.

CONFIDENTIAL: This  e-mail  and  any attachments are confidential and may
contain reserved information. If  you are not one of the named recipients,
please notify the sender immediately. Moreover, you should not disclose the
contents to any other persons, nor should the information contained be used
for any purpose or stored or copied in any form.

A posting from the Archives & Archivists LISTSERV List sponsored by the
Society of American Archivists, www.archivists.org. For the terms of
participation, please refer to
http://www.archivists.org/listservs/arch_listserv_terms.asp.

To subscribe or unsubscribe, send e-mail to [log in to unmask] In
body of message: SUB ARCHIVES firstname lastname *or*: UNSUB ARCHIVES To
post a message, send e-mail to [log in to unmask]

Or to do *anything* (and enjoy doing it!), use the web interface at
http://listserv.muohio.edu/archives/archives.html

Problems? Send e-mail to Robert F Schmidt <[log in to unmask]>


A posting from the Archives & Archivists LISTSERV List sponsored by the Society of American Archivists, www.archivists.org.
For the terms of participation, please refer to http://www.archivists.org/listservs/arch_listserv_terms.asp.

To subscribe or unsubscribe, send e-mail to [log in to unmask]
      In body of message:  SUB ARCHIVES firstname lastname
                    *or*:  UNSUB ARCHIVES
To post a message, send e-mail to [log in to unmask]

Or to do *anything* (and enjoy doing it!), use the web interface at
     http://listserv.muohio.edu/archives/archives.html

Problems?  Send e-mail to Robert F Schmidt <[log in to unmask]>

A posting from the Archives & Archivists LISTSERV List sponsored by the Society of American Archivists, www.archivists.org. For the terms of participation, please refer to http://www.archivists.org/listservs/arch_listserv_terms.asp.

To subscribe or unsubscribe, send e-mail to [log in to unmask] In body of message: SUB ARCHIVES firstname lastname *or*: UNSUB ARCHIVES To post a message, send e-mail to [log in to unmask]

Or to do *anything* (and enjoy doing it!), use the web interface at http://listserv.muohio.edu/archives/archives.html

Problems? Send e-mail to Robert F Schmidt <[log in to unmask]> A posting from the Archives & Archivists LISTSERV List sponsored by the Society of American Archivists, www.archivists.org. For the terms of participation, please refer to http://www.archivists.org/listservs/arch_listserv_terms.asp.

To subscribe or unsubscribe, send e-mail to [log in to unmask] In body of message: SUB ARCHIVES firstname lastname *or*: UNSUB ARCHIVES To post a message, send e-mail to [log in to unmask]

Or to do *anything* (and enjoy doing it!), use the web interface at http://listserv.muohio.edu/archives/archives.html

Problems? Send e-mail to Robert F Schmidt <[log in to unmask]>