A New Synthesizing Method for Handwriting Korean Scripts
Do-Hoon Lee, Hwan-Gue Cho · International Journal of Pattern Recognition and Artificial Intelligence · 1998
So far there have been models for generating handwriting scripts of western languages. Since the basic structure of oriental languages such as Chinese, Japanese and Korean, is quite different from that of alphabet-based western languages, western handwriting models cannot be applied directly to oriental languages except for a few models. In this paper, we propose a new generation model for generating Korean handwriting scripts. For this we choose 11 characteristic parameters which determine the shape of handwriting scripts. These parameters were selected experimentally by investigating ample Korean scripts. Furthermore we propose the Beta-velocity model to simulate the cursive stroke movement by manipulating curvilinear and cursive connection for consecutive strokes within a letter or a word. In order to generate these oriental cursive strokes, we adopted Beta distribution curve basis which enables one to make easy asymmetric velocity profile. By combining all these features, we synthesize various handwriting Korean scripts by controlling 11 characteristic parameters. In the final section of this paper, we give experimental results. These easy-to-synthesize Korean scripts can be used in evaluating off-line character recognition system and a word processor by supporting handwriting-style fonts.