RECMGMT-L Archives

Records Management

RECMGMT-L@LISTSERV.IGGURU.US

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Larry Medina <[log in to unmask]>
Reply To:
Records Management Program <[log in to unmask]>
Date:
Thu, 17 Aug 2006 08:13:39 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (66 lines)
On 8/16/06, Roach, Bill J. <[log in to unmask]> wrote:
>
> >>Digital Imaging Guidelines available online<<
>
> Just a quick question.  Most of the guidelines I have read suggest
> scanning and storing documents at 300 dpi.  Just a thought, but I find
> this a bit excessive unless there is a good chance that the documents
> will be OCR'd in the future.
>
> Any discussion?


As you stated and John A agreed, the quality of the source documents is the
key to establishing a threshold for scanning.  If you have all clean, like
size type, near-white to white paper, uncreased, stapled, folded source
documents, you can likely survive with 200dpi, and maybe would even want to
consider doing some serious testing of 150dpi as an acceptable resolution
for capture... but if this is what your population of source documents are,
the materials are quite possibly so recent that you already have them
available in electronic form (?) somewhere.

Most conversion/imaging/scanning projects involve backfile conversion of
mixed materials, some new, some not-so-new, and can involve a number of
different substrates from vellum to NCR to bond to onionskin.  They have
type of various sizes and qualities, maybe even some handwriting.  They have
holes punched in them, staples, folded corners, may have interspersed
colored sheets, two sided and odd sized sheets.  In some cases, you have
originals that are in such bad condition, you have to consider making an
intermediate copy to scan from to get a decent image.

The best bet when beginning a project of this type is to take a close look
across the "universe of documents" you will be scanning and find what
represents a good cross section of the source documents and then running a
series of test scans at varying levels of resolution, from 150dpi to (gasp!)
600dpi.  Gather relevant data about the run time for capture and processing,
resultant scan file sizes, and then do some image quality testing.

The testing should be done on a number of different monitors, because images
portray themselves differently depending on the hardware being used to view
them, so your images need to be acceptable on the worst of these, or you
need to consider upgrading hardware for any potential users of the images.
You should also see how long it takes the various images to display, and how
long it takes to refresh the screen when you load a second image. You should
also check the quality of a printed copy of the scanned images on the
various types of printers and MFDs you may be using.  Lastly, you should
attempt to OCR the various samples and see what the success rate is and how
large the resultant files are.

The storage size of the files, time it takes to transmit and load the
images, and speed of capture are factors that need consideration when
undertaking a project of this type, but the goal should be to have images
that are equal to, or better than, the source documents that can be used by
the individuals needing access to the information contained in them.   If
the end user has to spend excessive time enlarging the image, zooming in and
panning around to be able to clearly read the content on a display, and gets
prints that are unreadable, the whole conversion process was for nothing.

Larry
-- 
Larry Medina
Danville, CA
RIM Professional since 1972

List archives at http://lists.ufl.edu/archives/recmgmt-l.html
Contact [log in to unmask] for assistance

ATOM RSS1 RSS2