>>Is OCR to the point where it is very reliable....or, is it still as I
used to hear more of a problem than a help?<<

We have had outstanding results for several years with OCR.  We use it
not only for forms processing but also for document re-creation when a
paper copy of legacy documents is the only source available.  

I have also used it extensively for litigation support with the first
major application (Excalibur Technologies PixTex/EFS) being deployed in
1994.  At that time we OCR'ed and searched several million pages of
business documents to support a very large contract dispute.  We used
text search at our support site in ND to support a dozen or more
concurrent depositions around the country.  We could find responsive
information while the line of questioning was still in progress, often
contradicting what the individual being deposed had just stated.
Success was measured by receipt of a $384 million dollar settlement.
Based on that, I would say OCR has been of value for a long time.

There are some things to remember.  Just like Baskin Robbins, there are
different flavors of OCR.  Different engines work better for different
things.  Some were developed to read numbers, even specific fonts, or
alpha or mixed case alpha.  Selecting the right engine will greatly
improve your success. In addition, we use lots of validations,
comparisons and calculations to further improve our OCR success.  

The bottom line is we have found it to be very powerful in a wide range
of applications.

Bill R

Bill Roach, CRM
Enterprise EDMS Coordinator
State of North Dakota
ITD/Records Management
701-328-3589
 

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