| T—F互补OCON子网组合式神经网络模型的应用研究 |
| |
| 引用本文： | 徐静波 周美华. T—F互补OCON子网组合式神经网络模型的应用研究[J]. 上海工程技术大学学报, 1997, 11(2): 1-4 |
| |
| 作者姓名： | 徐静波 周美华 |
| |
| 作者单位： | 上海工程技术大学计算机及电子电气工程系!上海200335(徐静波)，中国纺织大学!上海200051(周美华) |
| |
| 摘 要： | 采用阴阳互补（T-F互补）网对结构于OCON子网之中，使OCON组合网在学习速度和识别率方面比常规BP网有很大提高。以手写体数字的识别为例，分析了该网络的应用特点，并给出实验结果。
|
| 关 键 词： | 神经网络 模式识别 邮政编码识别 算法 互补网对 |
A STUDY OF THE APPLICATION OF T-F COMPLEMENTARY OCON COMBINATORY NEURAL SUBNETS |
| |
| Xu Jingbo, Zhou Meihua. A STUDY OF THE APPLICATION OF T-F COMPLEMENTARY OCON COMBINATORY NEURAL SUBNETS[J]. Journal of Shanghai University of Engineering Science, 1997, 11(2): 1-4 |
| |
| Authors: | Xu Jingbo Zhou Meihua |
| |
| Affiliation: | Xu Jingbo; Zhou Meihua |
| |
| Abstract: | The T-F complernentary pair is applied to OCON subnet. The performance of OCON combiratory network is much better than that of general BP network on learning speed and recognition rate. The application feature of this network is analysed, based on the example of handwriting figure recognition. The result of the experiment is given. |
| |
| Keywords: | Neural networks Pattern recognition Post code recognition Algorithms Complementary network pair |
| 本文献已被 CNKI 维普 等数据库收录！ |
