Complete.Org: Mailing Lists: Archives: discussion: March 2003:
[aclug-L] Re: OCR Success
Home

[aclug-L] Re: OCR Success

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: <discussion@xxxxxxxxx>
Subject: [aclug-L] Re: OCR Success
From: "John Alexander" <wicjra0@xxxxxxxxxxx>
Date: Sat, 22 Mar 2003 13:02:41 -0600
Reply-to: discussion@xxxxxxxxx

Probably a little overkill, but does you scanner also allow you to make
enlarged copies, or increase the focal size of the scanned image? If so,
bump it up a few percent, and see if it gets any better.

ja 

-----Original Message-----
From: discussion-bounce@xxxxxxxxx [mailto:discussion-bounce@xxxxxxxxx]
On Behalf Of bruce
Sent: Wednesday, March 19, 2003 11:08 PM
To: discussion@xxxxxxxxx
Subject: [aclug-L] OCR Success


I needed and ocr program several months ago and tried 'clara'.  It was 
very disappointing, being unable to identify five percent of the 
characters in a scanned document, even after several hours trying to 
train it.

Today I found 'gocr' and played with it for several hours.  With no 
training, it can identify probably 99% of a 12 pt. ariel font document 
- not perfect, but a definite time saver.  Gocr is not too good on 
smaller text and my scanner only goes to 300x300 dpi.  In a few 
paragraphs from Linux Mag it didn't identify a single lower case 't'.  
There were about 40 of them and most were identified as being the 
symbol for the Euro.  A few times through it with sed s/EUR/t/ made it 
readable, but just barely.  A scanned newspaper is not much better.

So I got  an ocr program and learned a little sed.  Love my linux more 
every day.
bruce
-- This is the discussion@xxxxxxxxx list.  To unsubscribe, visit
http://www.complete.org/cgi-bin/listargate-aclug.cgi

-- This is the discussion@xxxxxxxxx list.  To unsubscribe,
visit http://www.complete.org/cgi-bin/listargate-aclug.cgi


[Prev in Thread] Current Thread [Next in Thread]