[EMAIL PROTECTED] wrote:

>amen to that. If you're ever bored David, there's still plenty of
>work over at gnumed. ;-)
>  
>
Thanks Ian. The good thing about pitifil for me was that it was a
project that I could get my head around and it did not matter if I broke
anything. (It also scratched a particular itch I had.)

It took a while to make my first faltering steps in python but once you
get the swing of it you can get productive pretty rapidly.

>>On 20/11/2005, at 4:50 PM, David Guest wrote:
>>    
>>
>>>Medical Director 2 currently locks up with big pit files (? ~ 60k). To
>>>clear the lock you will need to crash out and then go into
>>>C:\MDW2\Processing on the local machine and delete the offending pit
>>>file. Most scanned RTF files are less than 10k however.
>>>      
>>>
>I'm fascinated that your OCR is accurate enough to match patient names.
>What OCR package are you using?
>
Omnipage Pro version 15. I don't know if the recent upgrade made any
difference. As mentioned previously we scan to simple RTF Windows 97
version. That seemed to work pretty well.

I had a go at Kooka using the OCRAD and gocr OCR engines last week on my
debian machine. My output on some faxed TIFFs was very dismal. It was
not a fair comparison compared to clean scanned images on the Windows
box but open source OCR still has a fair way to go before catching the
commercials.

David

-- 
"UFW. Deb does linux." 

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

_______________________________________________
Gpcg_talk mailing list
[email protected]
http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk

Reply via email to