Optical chinese character recognition with a hidden Markov model classifier—a novel approach
B.-S. Jeng, M. W. Chang, Shujin Sun, Ching-Long Shih, Tian-xiu Wu · Electronics Letters · 1990
An optical Chinese character recognition system using a hidden Markov model (HMM) classifier has been developed to solve the input problem of Chinese characters. The recognition rate for 5,401 multi-font characters reaches 97% on average for inside tests and 91.5% on average for outside tests.