ONLINE CHINESE CHARACTER-RECOGNITION WITH EFFECTIVE CANDIDATE RADICALAND CANDIDATE CHARACTER SELECTIONS

A useful property of Chinese characters is that most of them possess a radical with a meaning. Our system is motivated by this characteristi c found in Chinese characters which leads us to radical-based candidat e selection approaches, performed before similarity measurement. The s earching ranges of the radicals and the characters can be basically de cided by the number of strokes in the input script. From the recognize d possible radicals, the candidate characters are limited to those hav ing such radicals and those having no radical. Since some of the candi date radicals may have very high matching costs, we can reduce the rec ognition time by discarding those unlikely candidate radicals. Further more, to speed up the inspection of radicals, we developed a radical e xtraction algorithm to narrow down the searching scope of the referenc e templates. The radical extraction algorithm was further improved by eliminating false extracted radicals. With these mechanisms, the numbe r of candidate radicals screened out is 286 out of 726 radical templat es and the number of candidate characters to be detailed matched is 12 3 out of 5401 Chinese characters. Through these efforts, the recogniti on rate improves to be 96.35% for the first rank and 98.96% for the fi rst 10 result candidates with the speed of 0.427 seconds on average pe r character on a PC using Intel 386-33 CPU. Copyright (C) 1996 Pattern Recognition Society.