free OCR program?

roman modic · External Since: Sep 06, 2005 Posts: 63

Hello!

"Jean" <Jean DeleteThis @spam.not> wrote in message news:e2%23z6WF2HHA.748@TK2MSFTNGP04.phx.gbl...
> Anyone know of a decent basic free OCR program that works on Win ME?
>

HP / UNVL Tesseract OCR
http://en.wikipedia.org/wiki/Tesseract_%28software%29
was open-sourced and it is now located on Google Code:
where development continues ...

Here are the Windows binaries, but I don't know if they work in Windows ME/98:
http://code.google.com/p/tesseract-ocr/downloads/list
(there are two win32 downloads - one is built with VC++6 and
the second with VC++Express)

BTW, it is possible to train it for special characters (with some limitations)
http://code.google.com/p/tesseract-ocr/wiki/TrainingTesseract

HTH, Roman

roman modic · External Since: Sep 06, 2005 Posts: 63

Hello!

"Heather" <figgyd.RemoveThis@nospam.invalid> wrote in message news:eOnRMpJ2HHA.3400@TK2MSFTNGP03.phx.gbl...
> Thanks......not too sure I am any further ahead. I will stick with my photo proggies and leave the complicated stuff to you
> guys!!
>
http://en.wikipedia.org/wiki/Optical_character_recognition

HTH, Roman

Jean · External Since: Feb 07, 2007 Posts: 17

Thank you!!! I'll check this out.

Jean

"roman modic" <modicr RemoveThis @astral-it.com> wrote in message
news:#gUg1mU2HHA.4880@TK2MSFTNGP03.phx.gbl...
| Hello!
|
| "Jean" <Jean RemoveThis @spam.not> wrote in message
news:e2%23z6WF2HHA.748@TK2MSFTNGP04.phx.gbl...
| > Anyone know of a decent basic free OCR program that works on Win ME?
| >
|
| HP / UNVL Tesseract OCR
| http://en.wikipedia.org/wiki/Tesseract_%28software%29
| was open-sourced and it is now located on Google Code:
| where development continues ...
|
| Here are the Windows binaries, but I don't know if they work in Windows
ME/98:
| http://code.google.com/p/tesseract-ocr/downloads/list
| (there are two win32 downloads - one is built with VC++6 and
| the second with VC++Express)
|
| BTW, it is possible to train it for special characters (with some
limitations)
| http://code.google.com/p/tesseract-ocr/wiki/TrainingTesseract
|
|
| HTH, Roman
|
|

N. Miller · External Since: Jul 31, 2006 Posts: 820

This message is not archived

Jean · External Since: Feb 07, 2007 Posts: 17

It didn't occur to me that some OCR (OPTICAL CHARACTER RECOGNITION, folks)
programs could input saved images. That ability could actually help me out
a bunch since there are others in my home owners association who can also
help scan the documents.

Our association is a "not for profit corporation". Thanks for the info
about your services.

Thanks,

Jean

"Gary S. Terhune" <none> wrote in message
news:#Rod7rQ2HHA.4004@TK2MSFTNGP05.phx.gbl...
| Well, as you've noted, nobody here has a decent recommendation for you,
| other than to try several and see what the results are. If I recall
| correctly, most of the paid versions have trial demos that you can at
least
| test for efficacy before purchasing. While the user interface may not
make
| it easy to find, you should be able to process stored images and not have
to
| scan each time you want to process a page. IOW, you can scan several
pages
| and store them, then try the same images on different OCR apps. Note that
| the best formats for images headed for OCR are loss-less types like BMP
or
| TIFF. JPEGs are lossy compressed and the edges of fonts get blurred. OK
for
| large type, not good for small type or where the paper quality of the
| document is less than pristine. Unless the software states its preferred
| resolution, you scan at the highest resolution your scanner is capable
of,
| but if the paper is dirty or aged, using a lower resolution can actually
| work better. You judge that based upon false returns of text where there
| isn't any, or mistaken returns that visual inspection suggests are due to
| smudging, dirt, etc. It's an acquired talent, even with the good
software.
| Underlining, italics and colors other than black are major tripping
points.
| Be sure to include those cases in your testing if they apply. (Scanning
in
| B&W sometimes helps with colors.)
|
| "...board of directors..." Is that a corporation, government agency, or a
| non-profit? If non-profit, I offer my services for a *very* reasonable
price
| (but you do the scanning.) If not, it would cost you a lot less to simply
| buy the ABBYY software. You won't get the database I usually produce,
| because I use Concordance (www.dataflight.com) -- it's expensive and
| requires that the *user* own a copy. Besides, it's designed for legal
usage,
| and while it would work for other purposes, it's overkill. If you take me
up
| on the offer, just know that I'm pretty busy at the moment with house
| repairs and a couple of other projects. Might be a month or two (or three
or
| four) before I could get to it. You can contact me at
| gryst_at_grystmill.com.
|
| --
| Gary S. Terhune
| MS-MVP Shell/User
| www.grystmill.com
|
| "Jean" <Jean DeleteThis @spam.not> wrote in message
| news:%23Ic7kfP2HHA.4712@TK2MSFTNGP04.phx.gbl...
| >I want to scan several years worth of minutes for board of directors
| > meetings. Obviously I could scan them in as images, but then I'd lose
the
| > ability to search through the documents for key words. My scanner (an
old
| > cheapo HP) comes with a primitive OCR program, but I was hoping to find
a
| > freebie that was less error prone.
| >
| >
| > "Gary S. Terhune" <none> wrote in message
| > news:#2sbLlJ2HHA.4680@TK2MSFTNGP03.phx.gbl...
| > | I thought you might have... Just wanted to cover the bases. Sorry
that I
| > | can't honestly recommend any but the high-priced ones. I once went
| > through a
| > | lot of testing, and even the ones that cost less than a hundred bucks
| > all
| > | failed in my book. Too many errors that needed correcting, and no
easy
| > way
| > | to do that correcting. What are your exact needs? To do a lot of
| > scanning? A
| > | little every once in a while? Scanning what kinds of documents?
| > |
| > | --
| > | Gary S. Terhune
| > | MS-MVP Shell/User
| > | www.grystmill.com
| > |
| > | "Jean" <Jean DeleteThis @spam.not> wrote in message
| > | news:%23CbA6QJ2HHA.1124@TK2MSFTNGP06.phx.gbl...
| > | > Being resourceful, I did a google search before I posted here. I
was
| > | > hoping
| > | > that someone in the newsgroup could make a recommendation.
| > | >
| > | >
| > | > "Gary S. Terhune" <none> wrote in message
| > | > news:u9RRQUH2HHA.5772@TK2MSFTNGP02.phx.gbl...
| > | > | AFAIC, it's hard to find a decent OCR program that you PAY for. I
| > doubt
| > | > I
| > | > | would consider any free OCR "decent", but then, I'm a
perfectionist
| > with
| > | > | high demands when it comes to OCR. I've done several projects
| > wherein
| > I
| > | > | batch-process hundreds or thousands of already scanned images and
| > then
| > | > put
| > | > | them into searchable databases. (Largest project, and the first I
| > did,
| > | > was
| > | > | ~650,000 pages.) I use ABBYY FineReader http://tinyurl.com/2pdrbo
| > | > |
| > | > | That said, Google is your friend. Here's a few. Why don't you
test
| > them
| > | > out
| > | > | and report back?
| > | > | http://www.google.com/search?hl=en&q=OCR+free
| > | > |
| > | > | --
| > | > | Gary S. Terhune
| > | > | MS-MVP Shell/User
| > | > | www.grystmill.com
| > | > |
| > | > | "Jean" <Jean DeleteThis @spam.not> wrote in message
| > | > | news:e2%23z6WF2HHA.748@TK2MSFTNGP04.phx.gbl...
| > | > | > Anyone know of a decent basic free OCR program that works on
Win
| > ME?
| > | > | >
| > | > | > Thanks,
| > | > | >
| > | > | > Jean
| > | > | >
| > | > |
| > | > |
| > | >
| > |
| > |
| >
|
|