Tesseract开发者对Tesseract讲解PPT

【实例简介】
Tesseract开发者的对Tesseract讲解PPT，包含OCR的历史，Tesseract的原理等
A Brief History of OCR Google hat is optical character Recognition? Mv invention relates to statistical machines of the type in which successive comparisons are made hetween a character and a charac- OCR My invention relates to statistical machines of the type in which successive comparisons are made between a character and a charac A Brief History of OCR Google OCR predates electronic computers. 2 a器 6 32 占5- 女7人小4 US Patent 1915993, Filed Apr 27, 1931 A Brief History of OCR Google 1929-Digit recognition machine 1953- Alphanumeric recognition machine 1965-US Mail sorting 1965-British banking system 1976-Kurzweil reading machine 1985-Hardware-assisted pc software 1988- Software-only PC software 1994-2000- Industry consolidation Tesseract Background Google ·Deve| oped on HP∪ X at hp between1985 and 1994 to run in a desktop scanner Came neck and neck with Caere and xIs in the 1995 UnlV test seehttp://www.isri.unlv.edu/downloads/at-1995.pdf) Never used in an hP product Open sourced in 2005. Now on http://code.googlecom/p/tesseract-ocr Highly portabl e Tesseract OCR Architecture Google Input: Gray or Color Image [ Region Polygons Adaptive Thresholding inary Image Character Find Text Outlines Connected Lines and Component Words Analysis Character Outlines Organized Into words Recognize Recognize Word Word Pass Pass 2 Adaptive Thresholding is Essential Google the Daily Value"re places the old“%US RDA. Foods are still evalu Some examples of how difficult it can be to make a binary image ated on how much the nutri Taken from the UNLV Magazine set ents in one serving contribute to the amount recommended http://www.isri.unlv.edu/isri/ocrtk) for each day In other words 300 milligrams (mg) of so dium accounts for 13% of the 2, 400 mg (or less) rec ommended daily. The i0o Daily value is for a 2.000 calorie intake if vou eat adjustments. MERMAID more or less than that. you have to make adjustments Disney, Secret Games 2 VINTAGE URTING E The Escort WITH ARTSIER video renters having Chained eat 25 DAVID HOLZMANS DIARY L.M. Kit Car-: run through Howards End(which drops son(1967. Foz Lorber unrated $79.95)Years before he went Hol: off the top 10 this week), The Crying Amityville: A New lywood with The Big Easy and: Game finally joins the chart at No7 in Great Balls of Fire/, director Jim: its third week in Generation McBride scored with this droll put. i release. Apparently, on, a fake underground movie that: laserdisc buyers were awaiting the Oscar Lady Dragon 2 shoots snarky darts through film.: nominee far more eagerly than tape student pretensions. David Holz: renters: The laser version debuts at a sur. Relentless 3 man (Carson) is a sincere young i twerp who, seeking Big Truth, prising No. 2. films his life in what was the the WhatI: fashionable grainy B&W verite. Meanwhile, the upward mobility of Columbia TriStar What he finds is a swift corrective Voyager's laserdisc of Bram Stoker's Dracula may soon to all those who think movies are be halted. Two just-released inexpensive, wide-screen UNATTRIBUTED more important than reality. Raw, ragged and right Diar is hoth an versions should cut into the sales of the current pricey Baselines are rarely perfectly straight Google Text Line Finding -skew independent published at ICDAR, 95 Montreal http:/scholar.google.com/scholar?g=skew+detection+smith) Baselines are approximated by quadratic splines to account for skew and curl Meanline, ascender and descender lines are a constant displacement from baseline Critical value is the x-height Volume 69 puges 872-879 Spaces between words are tricky too Google Italics, digits, punctuation all create special-case font-dependent spacing Fully justified text in narrow columns can have vastly varying spacing on different lines of 9. 5 annually while the Fed erated junk fund returned 11g% fear of financial collapse, 【实例截图】
【核心代码】