|
|||||
|
|
| 一种适用于模式识别的新型神经网络 | |
| 引用本文： | 乐清洪,高星海,郝俊,朱名管.一种适用于模式识别的新型神经网络[J].计算机工程,2004,30(17):17-18,35. |
| 作者姓名： | 乐清洪 高星海 郝俊 朱名管 |
| 作者单位： | 1. 航空飞行自动控制研究所,西安,710065;西北工业大学机电工程学院,西安,710072 2. 航空飞行自动控制研究所,西安,710065 3. 西北工业大学机电工程学院,西安,710072 |
| 摘 要： | 提出了一种适用于模式识别的新型神经网络模型——局部有监督特征映射网络，描述了该网络的拓扑结构和学习算法，研究了网络的基本性能，最后将其应用到了质量控制图的模式识别中。理论研究和仿真实验表明，该网络结构简单、算法简洁，收敛速度快、识别精度高，适用于需要大样本训练、随机干扰严重的复杂模式的分类与识别。 |
| 关 键 词： | 人工神经网络 模式识别 控制图 |
| 文章编号： | 1000-3428(2004)17-0017-02 |
A New Neural Network Adaptable to Pattern Recognition |
|
| LE Qinghong,GAO Xinghai,HAO Jun,ZHU Mingquan.A New Neural Network Adaptable to Pattern Recognition[J].Computer Engineering,2004,30(17):17-18,35. | |
| Authors: | LE Qinghong GAO Xinghai HAO Jun ZHU Mingquan |
| Affiliation: | LE Qinghong1,2,GAO Xinghai1,HAO Jun1,ZHU Mingquan2 |
| Abstract: | A new neural network named regional supervised feature mapping (RSFM) network is proposed in this paper. The topology structure and training algorithm of this network are represented, and its basic performance is studied. Patten recognition for quality control charts based on this new network is employed at last. Theoretical reasoning and numerical simulation results show this model possesses many advantages, such as simple structure and training algorithm, quick training and good recognition performance. It is suitable for pattern classification and recognition, especially for the situation which needs large training samples and associate with serious random disturbance. |
| Keywords: | Artificial neural network Pattern recognition Control charts |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |