ISOLATED HANDWRITTEN PERSIAN CHARACTER RECOGNITION USING MULTIPLE HIDDEN MARKOV MODELS

in this paper a new method for off - line recognition of isolated handwritten persian characters based on continuous density hidden makov models CDHMMs is proposed to remove the noise and equalize the character size some preprocessing operations such as median and morphological filters as well as normalization and centering procedures are applies. four types of feature vectors based on regional projection contour transformation RPCT are extracted for 32 persian characters.

کلیدواژه ها:

optical character recognition ، hidden markov models ، regioinal projection contiour transformation ، combination multiple experts