Generation of Diversiform Characters Using a Computational Handwriting Model and a Genetic Algorithm

In pattern recognition, a diversification of characters is necessary for learning, which is like neural network learning. The artificial diversification of characters has been suggested as one means of collecting a variety of characters. Accordingly, we show that a computational handwriting model can be applied to the diversification of characters. It is thought that characters diversified by the model can be used as a database of character images for learning.Wada & Kawato's handwriting model [11] is based on an optimal principle and the feature space of the characters includes sets of via-points extracted from actual handwritten characters. Therefore, if the via-point information is changed, diversiform characters can be generated using the handwriting model. In this paper, we propose a method for generating a diversification of characters by changing via-point information based on a genetic algorithm.