Hierarchical Chinese Character Database Based on Global Affine Transformation

This paper uses optimal global affine transformation (GAT) to construct hierarchical Chinese character database. GAT can automatically match the components in database and the corresponding components in characters. Therefore, GAT can avoid large labor that is needed for selecting points by hand. In order to improve the computing speed, specific expressions of the parameters are achieved and preprocessings are applied to characters´ and components´ images in this paper. Experimental results show that computing speed has been improved twenty times with good simulated results.