<br><br><div class="gmail_quote">On Fri, Feb 15, 2008 at 1:04 PM, Troy Arnold <<a href="mailto:troy@zenux.net">troy@zenux.net</a>> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div class="Ih2E3d">On Fri, Feb 15, 2008 at 10:41:52AM -0500, Jack Smith wrote:<br>
> Does anyone know of any good OCR software? Good enough to read smudged<br>
> copies? That doesn't cost an arm and a leg?<br>
<br>
</div>The best free stuff was pretty clearly tesserect last time I played around<br>
with it. <a href="http://code.google.com/p/tesseract-ocr/" target="_blank">http://code.google.com/p/tesseract-ocr/</a><br>
<br>
Getting good output depends a lot on the pre-processing you do to your<br>
images. Recently linux journal ran a very good article on OCR ... it<br>
may be online.<br>
</blockquote></div><br>Tesseract does a beautiful job on clean copy and the article does a pretty good job of showing how to clean up the source. I guess my source is just so bad it's faster to type it in. Unless there's something better on dirty copy out there?<br clear="all">
<br>-- <br>Jack Smith<br><br>English doesn't borrow from other languages -- English follows other languages down dark alleys and takes what it wants.