DIGIT-ARC read-only archive · snapshot: July 23, 2026

Google spreads its legs in open source OCR field.

5 posts · originally from https://geek.digit.in/community/threads/google-spreads-its-legs-in-open-source-ocr-field.37509/

eddie
eddie Original post #1
Sep 29, 2006 at 1:52 AM
Anyone who has ever tried his hand on any open source OCR application would know how pathetic and painful they are. There are simply not upto the mark. Thankfully Google is attempting to revive, a old HP's OCR application named, Tesseract and has released it as open source on sourceforge.

According to a reviewer from Linux.com, the tool showed an accuracy of >94% in its preliminary tests which imho is simply awesome. The tool is still a CLI only application but it shows so much promise that it will make any Linux user excited :)

Linux.com review
Google Code Blog post about the tool
Tesseract home on sourceforge
JGuru
JGuru #2
Oct 1, 2006 at 12:49 AM
That's good news. atlast Linux users will get to work with a good OCR software.
M
Oct 1, 2006 at 9:33 AM
And AFAICR, it's the first time google is releasing something under a open source license and specifically something that runs on linux. I hope this, too doesn't run on wine like google earth.
BTW, why are they using sourceforge for releasing it. They had started some similar project too na? This would be great opportunity to promote it.
L
Oct 1, 2006 at 6:57 PM
eddie said:
According to a reviewer from Linux.com, the tool showed an accuracy of >94% in its preliminary tests

!!!!!

(faints)
gary4gar
Oct 1, 2006 at 7:21 PM
well linux is the future os!!
after its user share increase no big companies can't just ignore it!
so its just one step ahead in a long trail