| 基于集成RBF神经网络的小类别手写体汉字识别系统 |
| |
| 引用本文： | 居琰,汪同庆,刘建胜,王贵新,彭健.基于集成RBF神经网络的小类别手写体汉字识别系统[J].计算机工程与应用,2002,38(23):100-102,158. |
| |
| 作者姓名： | 居琰 汪同庆 刘建胜 王贵新 彭健 |
| |
| 作者单位： | 重庆大学光电工程学院,重庆,400044 |
| |
| 摘 要： | 该文介绍了RBF神经网络的模型,讨论了RBF网络分类器的机理和特点,提出了一种集成RBF神经网络并应用于小类别手写体汉字识别系统的设计,采用了组合重心分解网格特征方法来提取汉字特征,设计了遗传进化隐层节点自生成算法用于RBF的训练。实验表明该小类别手写体汉字识别系统有很高的识别率,具有一定的实用推广价值。
|
| 关 键 词： | 模式识别 RBF神经网络 集成神经网络 手写体汉字识别 |
| 文章编号： | 1002-8331-(2002)23-0100-03 |
Small Set Handwritten Chinese Character Recognition Based on Integration RBF Neural Network |
| |
| Ju Yan Wang Tongqing Liu Jiansheng Wang Guixin Peng Jian.Small Set Handwritten Chinese Character Recognition Based on Integration RBF Neural Network[J].Computer Engineering and Applications,2002,38(23):100-102,158. |
| |
| Authors: | Ju Yan Wang Tongqing Liu Jiansheng Wang Guixin Peng Jian |
| |
| Abstract: | This paper presents Radial Basis Function Neural Network(RBFNN),and discusses its classifying mechanism and characteristic.An integration RBFNN is used for small set handwritten Chinese character recognition.This paper pro-pose a new feature ex traction named combined barycenter mesh decomposed algorithm.Genetic evolutionary self-generat-ing of hidden units algorithm is contrived to train RBF network.Experiment of this small set handwritten Chinese char-acter recognition system possessing a high level of recognition rate,shows that the System is practical. |
| |
| Keywords: | Pattern Recognition Radial Basis Function Neural Network Integration Neural Network Handwritten Chinese Character Recognition |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
