Printed Chinese character recognition via the cooperative block neural networks
Ning Xu, Xiaoqing Ding · 2003
In this paper, the authors propose a new architecture of artificial neural networks, which is called the cooperative block neural networks (CBNN), and use it for the recognition of multi-font printed Chinese characters. Preliminary simulation results have proven that in tackling the difficulties associated with sophisticated topological structures of characters and the large number of Chinese characters to be recognized, the CBNN system is successful.>