Title :
OCR Correction via Human Computational Game
Author :
Jovian, Lin Tingji ; Amprimo, Olivier
Author_Institution :
Nat. Univ. of Singapore, Singapore, Singapore
Abstract :
We present TypeAttack, a Facebook game that improves the efficiency of digitization, which is the process of converting analog information into digital information. In this game, players compete to type out texts from images of scanned documents in the quickest time possible, with the utmost accuracy. These historical documents are thus manually digitized as a bi-product of game play. Harnessing the perceptual abilities of humans as well as their desire to be entertained, TypeAttack enhances the digitization of old documents, especially those that cannot be transcribed accurately by Optical Character Recognition (OCR) programs.
Keywords :
computer games; optical character recognition; Facebook game; OCR correction; TypeAttack; analog information; digital information; historical documents; human computational game; optical character recognition; scanned documents; Accuracy; Facebook; Games; Humans; Libraries; Optical character recognition software; XML;
Conference_Titel :
System Sciences (HICSS), 2011 44th Hawaii International Conference on
Conference_Location :
Kauai, HI
Print_ISBN :
978-1-4244-9618-1
DOI :
10.1109/HICSS.2011.319