[bksvol-discuss] Re: OCR errors

  • From: talmage@xxxxxxxxxx
  • To: bksvol-discuss@xxxxxxxxxxxxx
  • Date: Wed, 09 Mar 2005 15:49:00 -0500

Hi Pratik,

I have to disagree on this one, at least as far as Openbook users go. I don't know if it is still the same with the current version, but version 6, and perhaps others, have an entry in their correction dictionary for com to corn. While I still may come across occasional m to rn, the occurances are no longer universal.
Another one that Openbook has in its dictionary is tom to torn, which can give you a great many trendy names if any of the characters happened to be named Tom in the book.


Dave

At 03:02 PM 3/9/2005, you wrote:
Dave,

I suspect the Mig to MiGo change may be due to some corrections file error.

The change from com to corn is not due to a correction file error but an OCR
error.  This is where the letter M is broken down into the fragment rn.
While common words such as cornputer may be individually placed into
correction files for processing,  Not all instances are as easily replaced.

Pratik

Pratik Patel
Interim Director
Office of Special Services
Queens College
Director
CUNY Assistive Technology Services
The City University of New York
     ppatel@xxxxxx

-----Original Message-----
From: bksvol-discuss-bounce@xxxxxxxxxxxxx
[mailto:bksvol-discuss-bounce@xxxxxxxxxxxxx] On Behalf Of talmage@xxxxxxxxxx
Sent: Wednesday, March 09, 2005 12:20 PM
To: bksvol-discuss@xxxxxxxxxxxxx
Subject: [bksvol-discuss] OCR errors

One that I can't fathom, that I've seen in a number of the military
oriented books, is when a MIG becomes a MiGo.  Typically, it will be
something like a MIG-29 becoming a MiGo-29.
While I'm on the subject of military books, I guess it is once again time
for my quarterly request for people to either delete the [com becomes corn]
correction in their recognition dictionary, or do a global replace before
submitting, cornmander, and cornmand are not words folks.

Dave

At 11:49 AM 3/9/2005, you wrote:
>Hehehe
>If you think r and t can't be confused just go to the web address in my
>signature and when you get there look for the scannos link and hit enter
>on it.  You will see just how many of those mistakes were made by that
>particular confusion.
>
>I used to be able to read print, so I know what a lot of letters look
>like, too.  I just don't know much about unusual fonts.  Sometimes you
>really have to use your imagination to fathom how things can be
>misrecognized in the way they were.
>
>Sarah Van Oosterwijck
>http://home.earthlink.net/~netentity


Other related posts: