On-line Thai handwritten character recognition using hidden Markov model and fuzzy logic

This paper presents a new on-line recognition of Thai handwritten characters. Active researches in Thai handwritten character recognition are converged into two distinct methods, HMM and fuzzy logic classifier. The former showed poor recognition rate due to Thai fuzzy characters. The shortcoming of the latter is on difficulties in establishing the set of rules to cover a whole handwriting styles. Our method is proposed to exploit the better of two worlds (HMM and distinctive feature based fuzzy classifier). The experimental result was shown an average recognition rate is improved from 89.1%(using HMM) to 91.2 using our proposed method.