| 基于RBF人工神经网络的空间插值 |
| |
| 引用本文： | 杨海荣. 基于RBF人工神经网络的空间插值[J]. 长沙交通学院学报, 2006, 22(1): 68-71 |
| |
| 作者姓名： | 杨海荣 |
| |
| 作者单位： | 长沙理工大学,公路工程学院,湖南,长沙,410076 |
| |
| 摘 要： | 根据RBF网络能以任意精度逼近任意函数这一特性,将RBF网络应用于空间插值,认定地表空间坐标的空间分布可以用一复杂的非线性函数模拟。该函数是由多种因素综合作用的结果,如果以各因素为输入、对应空间坐标值为期望输出,对网络进行训练可对地理要素的空间分布进行模拟。试验表明,神经网络应用于空间插值是可行的。
|
| 关 键 词： | 空间插值 神经网络 RBF |
| 文章编号： | 1000-9779(2006)01-0068-04 |
| 收稿时间： | 2005-09-14 |
| 修稿时间： | 2005-09-14 |
A Space Interpolation Method Based on RBF Artificial Neural Network |
| |
| YANG Hai-rong. A Space Interpolation Method Based on RBF Artificial Neural Network[J]. Journal of Changsha Communications University, 2006, 22(1): 68-71 |
| |
| Authors: | YANG Hai-rong |
| |
| Affiliation: | College of Highway Engineering, Changsha University of Science and Technologe, Changsha 410076, China |
| |
| Abstract: | It has been found that RBF neural network can represent all functions at any accuracy.From this point of view,RBF neural network to space interpolation is introduced.It is assumed that the spatial (distribution) of geographic coordinate can be fitted as a complex non-linear function which consists of many factors.Using these factors as input and spatial coordinate as expectant output,it is feasible to simulate the spatial distribution of geographic elements by training network.Finally,a space interpolation case is(simulated),and the result shows that it is feasible to introduce RBF to space interpolation. |
| |
| Keywords: | space interpolation neural network RBF |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
