Anyone know of any OCR software that -actually- works? I've tried gocr and ocrad and neither one produces -any- readable output. One more quick question, does anyone know what the file .kturnedon refers to. Mine only has contents false. Thanks, Ken -- Ken Schneider UNIX since 1989, linux since 1994, SuSE since 1998
On Wednesday 01 February 2006 20:58, Ken Schneider wrote:
Anyone know of any OCR software that -actually- works? I've tried gocr and ocrad and neither one produces -any- readable output.
I've used ocrad with a reasonable amount of success - you need to feed it the highest resolution monochrome image you can, but the errors it makes are systematic so you can generally search and replace to sort out the bulk of the text.
One more quick question, does anyone know what the file .kturnedon refers to. Mine only has contents false.
Thanks,
Ken
-- Ken Schneider UNIX since 1989, linux since 1994, SuSE since 1998
-- "The man who strikes first admits that his ideas have given out." (Chinese Proverb)
On Wed, 2006-02-01 at 21:26 +0000, Dylan wrote:
On Wednesday 01 February 2006 20:58, Ken Schneider wrote:
Anyone know of any OCR software that -actually- works? I've tried gocr and ocrad and neither one produces -any- readable output.
I've used ocrad with a reasonable amount of success - you need to feed it the highest resolution monochrome image you can, but the errors it makes are systematic so you can generally search and replace to sort out the bulk of the text.
Tried at the highest resolution and had about a 20% success rate which is pretty lousy. I could type the document faster and I am not a typist. Maybe there is a windows solution I can run in wine or vmware. -- Ken Schneider UNIX since 1989, linux since 1994, SuSE since 1998
There's also Abbyy's FineReader for Windows which has a better repuation than the most. They also have a corporate edition which can be accessed through the Net, provided of course you have one Windows PC at the other end.
For scanning in linux there is one web site - www.vividata.com You may like to see ----- Original Message ----- From: "Ken Schneider" <suse-list@bout-tyme.net> To: <suse-linux-e@suse.com> Sent: Thursday, February 02, 2006 4:46 AM Subject: Re: [SLE] OCR software
On Wed, 2006-02-01 at 21:26 +0000, Dylan wrote:
On Wednesday 01 February 2006 20:58, Ken Schneider wrote:
Anyone know of any OCR software that -actually- works? I've tried gocr and ocrad and neither one produces -any- readable output.
I've used ocrad with a reasonable amount of success - you need to feed it the highest resolution monochrome image you can, but the errors it makes are systematic so you can generally search and replace to sort out the bulk of the text.
Tried at the highest resolution and had about a 20% success rate which is pretty lousy. I could type the document faster and I am not a typist. Maybe there is a windows solution I can run in wine or vmware.
-- Ken Schneider UNIX since 1989, linux since 1994, SuSE since 1998
-- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com
In a previous message, Ken Schneider <suse-list@bout-tyme.net> wrote:
Anyone know of any OCR software that -actually- works? I've tried gocr and ocrad and neither one produces -any- readable output.
I've used Kooka to drive both gocr and ocrad with fair success - the output required a few changes per sentence but not much more than that. As with most OCR software, they need the best scan you can provide (Kooka helps a lot with that), a good dictionary to work against (again, Kooka does that) and an original using a readable font. I have to say that driving either gocr or ocrad alone was a thankless task for me. If your original is good quality, try Kooka. Read the help files and plug in the suggested settings, and see how you go. John -- John Pettigrew Headstrong Games john@headstrong-games.co.uk Fun : Strategy : Price http://www.headstrong-games.co.uk/ Board games that won't break the bank Fields of Valour: 2 Norse clans battle on one of 3 different boards
Anyone know of any OCR software that -actually- works? I've tried gocr and ocrad and neither one produces -any- readable output.
One more quick question, does anyone know what the file .kturnedon refers to. Mine only has contents false.
Thanks,
Ken
-- Ken Schneider UNIX since 1989, linux since 1994, SuSE since 1998 Ken I've been using xsane and kooka. Bot cover to OCR but I then need to go
On Wednesday 01 February 2006 12:58, Ken Schneider wrote: thru and correct the characters that they miss read. If I'm not mistaken xsane uses gOCR. Please post if you find some OCR software that works good. -- Russ
At 03:58 PM 2/1/2006 -0500, Ken Schneider wrote:
Anyone know of any OCR software that -actually- works? I've tried gocr and ocrad and neither one produces -any- readable output.
The program from what used to be called Caere, Omnipage, is wonderful. It runs in MS Windows; maybe it can be coaxed to run in one of the clone things that Linux has. It's not free--the basic version costs about $100, but it's worth it to have something that just _works_! It does columns, it is not fazed by scans that are a little off the absolute vertical, all the cheapies that come with scanners should be thrown in the trash can. This is the program you need. If you don't use Windows for anything else, this is a reason to do so. (I don't have any financial interest in the company.) --doug -- No virus found in this outgoing message. Checked by AVG Anti-Virus. Version: 7.1.375 / Virus Database: 267.14.25/247 - Release Date: 1/31/2006
On Wednesday 01 February 2006 04:59 pm, Doug McGarrett wrote:
At 03:58 PM 2/1/2006 -0500, Ken Schneider wrote:
Anyone know of any OCR software that -actually- works? I've tried gocr and ocrad and neither one produces -any- readable output.
The program from what used to be called Caere, Omnipage, is wonderful.
I used to support hundreds of scan stations running either a custom VP app I wrote with the Calera engine or OmniPage Pro from then Calera a.k.a. Caere a.k.a. Nuance. (Check out the website, looks as if they own both Caere and Dragon now. Plus there's a hottie wearing an iPod.)
It runs in MS Windows; maybe it can be coaxed to run in one of the clone things that Linux has.
I doubt it. Probabaly way too much in terms of Win32 integration. I just looked on codeweavers and saw no OCR listed. Worth a shot, though.
-- No virus found in this outgoing message. Checked by AVG Anti-Virus. Version: 7.1.375 / Virus Database: 267.14.25/247 - Release Date: 1/31/2006
What's a Anti-Virus? Oh, that's right, something you run on a Windows system. :P -- kai www.perfectreign.com linux - genuine windows replacement part
Ken Schneider wrote:
Anyone know of any OCR software that -actually- works? I've tried gocr and ocrad and neither one produces -any- readable output.
One more quick question, does anyone know what the file .kturnedon refers to. Mine only has contents false.
Thanks,
Ken
I really like PaperPort. Windows program. Same vendor (after mergers) as Omnipage, so I think the OCR engine is the same. I really like the way the program works, is layed out, functions. I'm still on vesion 9, 10 is out now. Does ocr pretty well, much much better that the first version I purchased, about 6 or so. Very smart from a document management/filing/storage point of view. I'm not sure of the difference between Omnipage and Paperport though they came from different companies originally. Paperport does export to pdf format, which is pretty cool for a $99 windows code chunck. Company was called ScanSoft, now called Nuance, www.nuance.com. I was thinking of adding PaperPort to the Novell list of required/requested linux must have programs. No idea if it works with wine. Jim
On Wed, Feb 01, 2006 at 03:58:38PM -0500, Ken Schneider wrote:
Anyone know of any OCR software that -actually- works? I've tried gocr and ocrad and neither one produces -any- readable output.
I've had fairly good results with claraocr. It's not the most stable of all programs (so save often), but as it is a learning OCR, results do get better and better the more you teach it. Development seems to have stalled, though. You can find it at http://www.geocities.com/claraocr/ Cheerio, Thomas
participants (9)
-
Doug McGarrett -
Dylan -
Jim Flanagan -
John Pettigrew -
kai -
Ken Schneider -
nathan -
russbucket -
T. Ribbrock