Message-Id: <199701240214.UAA20818@library.wustl.edu> Date: Thu, 23 Jan 1997 21:04:15 -0500 From: Stacey Kimmel <mailto:stacey_kimmel@LIBRARY.LIB.NCSU.EDU> Subject: Index card inquiry- respons To: mailto:IMAGELIB@LISTSERV.ARIZONA.EDU
Hi,A few days ago I posted an inquiry about methods for scanning index cards. Below are a list of sources and listserv responses on this topic. THANKS to all of those who took the time to offer advice/suggestions--they were a great help.
Sources:
"Digitization and the creation of virtual libraries" in _Information Technology and Libraries_ March 1995, pp. 38-40 for an account of how Princeton scanned index cards. -------------------------------------------------------------------------- URLs
Library of Va Digital Card Index http://image.vtls.com/collections/ Princeton U Card Catalog http://imagecat1.princeton.edu/ecc
--------------------------------------------------------------------------
Listserv responses
Response 1 The Library is initiating work to scan our file of visible periodical receipts. This is a 4"x6" card file with a number of "flimsey" overlays that report various receipt/deposit types. I mention it to you because we have done a fair amount of investigation and planning since the file incorporates a real "mish-mash" of sizes, paper weights, colors, print and hand annotations in both pen and pencil, and verso-rechto problems. Our automation people have identified a "BanTec" Scanner for our use with accompanying software (also from BanTec (or BanTech)) This is not an endorsement of a particular product, we have found this to be a good solution for our particular problems and environment, your needs and requirements might be quite different and justify other equipment) The scanner will accommodate the different weights, colors, verso-rechto, etc. with minimal operator intervention. I don't have technical information on the device but I could probably get specs for you if you are interested. I do have access to some of our preliminary documentation which you might be of interest, or might be too far afield to be of much use. Your file would SEEM to be better organized than ours, i.e. with data in consistent places and perhaps more available to data scanning and OCR than we anticipate ours to be (we don't expect to do any OCR, but it could be a consideration). We will also be providing barcoded links to separate each document (card set) and to facilitate subsequent retrieval from our mainframe based applications. We anticipate start-up of our pilot project within the next month to six weeks.
If I can answer further questions or do a bit of leg work for you here please do not hesitate to ask.
Dexter L Fox ++++++++++++++++++++++++++++++++++ + + ++++++ ++ + Automation Planning & Liaison Office +++++++++ + + +++ + +++++ +++ + LM-532, STOP 4010 +++++++++++++++++++++++++++ ++ + +++ + +++++ +++ + The Library of Congress ++++++++++++++++++++ +++ + +++ + +++++ +++ + Washington, DC 20540 ++++++++++++++++++++++ ++++ + +++ + +++++ +++ + TEL: 202/707-2550 ++++++++++++++++++++++++ + + +++++ +++ + FAX: 202/707-6269 +++++++++++++++++++++++ ++++++ + ++++++ +++++ +++ + INET: mailto:fox@mail.loc.gov +++++++++++++++++ +++++++ + ++++++ +++++ +++ + INET: mailto:dfox@loc.gov ++++++++++++++++++++ ++++++++ + ++++++ + ++ ++
------------------------------------------------------------------------
Response 2 Any good, "brand name", flatbed scanner capable of 1200dpi optical resolution should work fine for the project you described, and should also work fine for any future image scanning project. I might suggest purchasing a transparency adaptor for the scanner so you could scan 4 X 5 or larger negatives, or transparencies. To scan a slide collection, a good 35mm scanner is a must - $6,000-$10,000. Scanning just a few thousand cards or pages of text is okay with a flatbed, however, if the count grows to a few ten thousand, a document scanner, such as the Xerox document scanner ($20,000) would be necessary - 30 pages/minute, scanning both sides automatically. Omnipage is a fine OCR program, and I recommend it highly. The most labor intensive part of this phase is spell checking and proofing - do not trust any OCR program to convert text accurately. I assume the Kofax board you asked about is an graphics accelerator board. Unless your workload exceeds five to ten thousand items, it is not really necessary, nice, but not needed. Our library has two projects ongoing now, one is image collections, sample is www.libs.uga.edu/darchive/darchive.html We do about 7,000 images a year. Each image is converted to film before scanning for the preservation part of this project. Our text scanning project can be seen in www.galileo.peachnet.edu, choose public databases, then state government publications. We scan about 100,000 pages per year in this project
Bob Henneberger INTERNET:mailto:bhennebe@uga.cc.uga.edu Computer Services WWW:scarlett.libs.uga.edu Preservation services Compu$erv:73727,2615 University of Georgia Libaries Voice: 706-542-2131 Athens, GA 30602 FAX: 706-542-4144
------------------------------------------------------------------------
Response 3
I recall Princeton University library getting involved in a major project with their old catalog cards. YOu might try contacting someone there.
+-+-+-+-+-+-+-+-+-+-+-+-+-+ Peter Nelson 5 College Project Archivist c/o Mount Holyoke College Archives (413) 538-3020
------------------------------------------------------------------------ Response 4 i can only give you some quick shoot from the hip remarks at the moment. --a $5-6,000 scanner (Fujitsu, Bell&Howell) would be appropriate, though potentially more scanner than is needed. >I think the cost is more around 4000.00 the quality of the feeder is
critical. the cost may have gone up. Don't scrimp.-- OmniPage Professional supports zone scanning and seems to have other desirable features for this type of project. How are you planning on getting the image into a database?
-- Kofax card recommended? this is what i use. seems good and widely used.
-- Im hoping the same scanner could conceivably handle both the image scanning & the OCR work. yes.
let me know what you wind up doing.
good luck steve
------------------------------------------------------------------------- Response 5
i scan display ads for an online newspaper
i use cuneiform with an agfa scanner (shareware)
if the cards are short me thinks a really fast secretary, would be your best bet. scanning cards individually requires sseveral steps the slowest of which is the the actual scan
if you photocopy the cards prior to scanning, enlarge them to a font size of 10 or 12
Susan ######################################################################################################################## ##### Susan "Instinct must be thwarted, mailto:sforman@one.net just as one prunes the branches http://w3.one.net/~sforman of a, tree so that it will http://www.nwdc.com/~exotic grow better." Henry Matisse ######################################################################################################################## #####
---------------------- Stacey Kimmel Collection Manager for Education, Psychology, and Management NCSU Libraries Box 7111 North Carolina State University Raleigh, NC 27695-7111 919.515.1624 fax: 919.515.7854 Internet: mailto:stacey_kimmel@ncsu.edu ------------