home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Power Programmierung
/
Power-Programmierung CD 2 (Tewi)(1994).iso
/
doc
/
hypertex
/
file31
< prev
next >
Wrap
Text File
|
1988-02-01
|
2KB
|
33 lines
How to OCR (optical character recognition) documents
====================================================
The first step in building a hypertext system <FILE30 BUILDING> is converting
the information into ASCII or graphic files. The machines that convert text
printed onto paper into an ASCII file are called optical character reader
(OCR) machines.
As for this service, we've paid OCR prices ranging from $1 to $4 per page.
Most of these prices apply only to OCR reading of typewritten texts. And
these prices also do not include proofing, which we've found to be the
major cost in preparing information for hypertext indexing.
We now have two system that read both typewritten and typeset text. The
cost for these OCR scanners and necessary software range from $4,000 to
$15,000. The differences in price reflect the error rate in reading
typeset text (95% to 98.5% accuracy), with the more expensive machines
being more accurate.
The cheaper machines simply scan the document, then display the text on a
screen. The more expensive machines have a learning mode in which you note the
scanning errors in one or more pages of text. The machine then rescans the
material, correcting those errors.
As a result of having access to these machines, if you have need for such
OCR service, I'd be happy to describe the hardware, software, or quote the
cost of rapidly converting your text on paper to ASCII files.
Neil Larson 1/16/88 FILE31
44 Rincon Rd., Kensington, CA 94707
Copyright MaxThink 1988 -- Call 415-428-0104 for permission to reprint