基于RBF神经网络的面状水体识别模型及其应用
文献类型：期刊论文
| 作者 | 胡静涛
|
| 刊名 | 水电能源科学
 |
| 出版日期 | 2014 |
| 卷号 | 32期号:1页码:59-62, 83 |
| 关键词 | 面状水体 识别模型 RBF神经网络 DEM |
| ISSN号 | 1000-7709 |
| 其他题名 | Surface Water Recognition Model Based on RBF Neural Network and Its Application |
| 产权排序 | 1 |
| 中文摘要 | 针对面状水体识别过程中面状水体数据特征不宜提取、伪洼地易与面状水体混淆的问题,通过分析面状水体的面积、深度和潜在出水口等基本DEM数据特征,构建了面状水体识别模型,并将面状水体的三个数据特征和面状水体识别模型的计算结果作为输入输出神经元,利用RBF神经网络对建立的面状水体识别模型进行了仿真验证。从全国1∶250 000 DEM数据中选取150组洼地数据作为样本数据,采用减聚类算法对RBF神经网络进行训练,训练时样本的最小平均相对误差为2.75%,仿真的准确率为98%,表明面状水体识别模型可解决面状水体和伪洼地难以区分的问题,并提高了面状水体识别的准确率。 |
| 英文摘要 | Aiming at the problems of extraction of surface water data characteristics and easily confusing between pseudo-depression and surface water，the surface water recognition model was proposed based on the DEM data features that include area， depth and potential outlet． Three data features of surface water and computational result of recognition model were taken as input and output of neuron． And ＲBF neural network model was established to verify the surface water recognition model． 150 groups of depressions extracted from 1: 250 000 DEM data were taken as sample data． ＲBF neural network model was trained by using subtractive clustering method． The minimum average relative error was 2. 75% in the training of the samples． The recognition accuracy of the model was 98%． As the results，the proposed model solves the problem of difficulty distinguishing between surface water and pseudo-depression，and it improves the recognition accuracy rate． |
| 资助信息 | 国家科技重大专项(2009ZX07528-004) |
| 语种 | 中文 |
| 公开日期 | 2014-11-03 |
| 源URL | [http://ir.sia.ac.cn/handle/173321/15190]  |
| 专题 | 沈阳自动化研究所_信息服务与智能控制技术研究室 |
| 推荐引用方式 GB/T 7714 | 胡静涛. 基于RBF神经网络的面状水体识别模型及其应用[J]. 水电能源科学,2014,32(1):59-62, 83. |
| APA | 胡静涛.(2014).基于RBF神经网络的面状水体识别模型及其应用.水电能源科学,32(1),59-62, 83. |
| MLA | 胡静涛."基于RBF神经网络的面状水体识别模型及其应用".水电能源科学 32.1(2014):59-62, 83. |
入库方式： OAI收割
来源：沈阳自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。