Google Adds OCR To PDF and Images 76
Kilrah_il writes "Now you have the option to OCR every PDF and image you upload to Google Docs. 'When you upload files to Google Docs, you'll notice a new option that tells Google to convert the text from PDF and image files to Google Docs documents. ... I've tried to convert an excerpt from the book Rework and the result wasn't great. About 10% of the text has been incorrectly converted and the formatting hasn't been preserved.'"
Captcha correction? (Score:4, Interesting)
Could google provide some sort of opt-in service where our PDFs (one word at a time) could appear as a captcha. More or less what reCaptcha does, except with something a bit newer.
Where did all the ReCAPTCHA go? (Score:3, Interesting)
Google Captcha processor here I come!!!! (Score:3, Interesting)
Re:lolwut? (Score:4, Interesting)
Didn't fail at all on a PDF with typed text for me. Did you actually try it?
I bet they don't actually use OCR on a PDF with typed text as they can just extract it from the PDF, they probably use that on images inside PDFs though.
OCR efficiency (Score:1, Interesting)
> the result wasn't great. About 10% of the text has been incorrectly converted and the formatting hasn't been preserved.
Well, what is the state of the art of OCR today? I wouldn't call this a bad result either... And OTOH, if people were correctly trained in spelling, we would have made do without spell checkers and have invested in OCR technology instead, right? ;)
Re:Google Captcha processor here I come!!!! (Score:2, Interesting)
Re:Captcha correction? (Score:1, Interesting)
Google search had OCR before Google Docs (Score:2, Interesting)
Google's search engine started doing OCR on any scanned documents they found in late 2008. The results were horrible in some cases, but it didn't matter. The searchable OCR results made it possible to find things more easily and you could obviously refer to the original source if the OCR was too garbled.
lots of tough problems in OCR (Score:2, Interesting)
OCR consists of many steps; recognizing the individual characters is only one of them. You also need to separate text from images, group characters into lines and columns, separate floats, captions, and body text, etc. Many of those are tough problems even if someone hands you a PDF with all the characters. And if any one of them is wrong, the entire output may be wrong.
Recognizing individual characters is also harder than you may think because there is such a wide variety of fonts in use and because there are so many odd things that can happen. Even in perfectly rendered images (no dirt etc.), two characters may be bit-identical but mean something different in different fonts. Ligatures, underlines, unknown characters, etc. also make the problem quite a bit harder.
And even though 1% error would be low for just about any other machine learning or pattern recognition problem, that's a high OCR error and looks quite bad; people are much more sensitive to OCR errors than pattern recognition errors in other contexts. Furthermore, there are a lot of characters to be classified and you only get very little CPU time per character.
We've been developing an OCR system (ocropus.org) for a while now (see http://bit.ly/9Xputj [bit.ly] for status info). It's fairly easy to get excellent performance on a closed dataset with a well-defined character set. Getting acceptable performance on arbitrary documents and dealing with all the special cases (ligatures, foreign characters, color images, magazine layouts, unknown languages, Unicode issues, etc.) is tons of work.
Oh, and in case you're wondering, although Google has sponsored OCRopus (thanks!), OCRopus is a separate project from Google's internal OCR efforts.