5 month ago
Andy Baio : Hoefler & Frere-Jones' Estupido Espezial, a joke version of OCR-A with swashes - even their joke typefaces get them paid [via]
nelson : Decorative OCR font - Funny story of typographer love
# copy
13 month ago
joshua : Twibright Optar - store data with a laser printer and paper
Rod Begbie : Twibright Optar - Codec allowing you to reliably back your data up to dead-tree. You can print about 200kb to a sheet of A4, then scan it in later to retrieve your data. [via] #
# copy
13 month ago
Simon Willison : tesseract-ocr - tesseract-ocr. Open source OCR, sponsored by Google. I just sat in on a talk on this at OSCON and the complexity of the problem is pretty incredible.
deusx : tesseract-ocr - Google Code - "one of the most accurate open source OCR engines available"
# copy
15 month ago
nelson : Breaking captchas - Some random guy in Ukraine attacks captchas and publishes results
# copy
15 month ago
Andy Baio : reCaptcha - using the effort answering captchas to help digitize scanned books
Rod Begbie : What is reCAPTCHA? - This I like a lot -- CAPTCHAs partially made up of text that couldn't be OCR'd successfully (so bots are unlikely to crack it), the solving of which in turn helps the Internet Archive digitize books. [via] #
Matthew M. Boedicker : captchas that have the useful side effect of digitizing books - (via lifehacker) [via]
kaninka.net : reCAPTCHA - Digitizing Books One Word at a Time
# copy25 month ago
Rod Begbie : FuzzyOcrPlugin - Spamassassin Wiki - SpamAssassin plugin which OCRs the images attached to image-only mails to work out if they contain text like V14GRA. I imagine this is massively CPU-intense, but it might be worth looking at. #
# copy
36 month ago
kayodeok : Camera phones will be high-precision scanners - The software, developed by NEC and the Nara Institute of Science and Technology (NAIST) in Japan, goes further than existing cellphone camera technology by allowing entire documents to be scanned simply by sweeping the phone across the page
# copy
37 month ago
cameron : PWNtcha - captcha decoder - A good rundown of various captcha strengths and weaknesses
# copy
