| 最小二乘支持向量机在海洋测深异常值探测中的应用 |
| |
| 引用本文： | 黄贤源, 翟国君, 隋立芬, 黄谟涛. 最小二乘支持向量机在海洋测深异常值探测中的应用[J]. 武汉大学学报 ( 信息科学版), 2010, 35(10): 1188-1191. |
| |
| 作者姓名： | 黄贤源 翟国君 隋立芬 黄谟涛 |
| |
| 作者单位： | 1信息工程大学测绘学院,郑州市陇海中路66号450052;2天津海洋测绘研究所,天津市友谊路40号,300061 |
| |
| 基金项目： | 国家863计划资助项目(2007AA12Z326)；国家自然科学基金资助项目(40971306,40974010) |
| |
| 摘 要： | 提出了利用最小二乘支持向量机(LS-SVM)构造海底趋势面,利用该趋势面对海洋测深异常值进行剔除的方法,并与趋势面滤波进行了分析和比较,用定理证明趋势面滤波只是LS-SVM取特定参数时的解。实测算例表明,通过调整LS-SVM的参数,使其构造的趋势面更合理,从而有效地剔除测深异常值。
|
| 关 键 词： | 最小二乘支持向量机(LS-SVM) 趋势面滤波 权重系数 测深异常值 |
| 收稿时间： | 2010-07-22 |
| 修稿时间： | 2013-07-09 |
Application of Least Square Support Vector Machine to Detecting Outliers of Multi-beam Data |
| |
| HUANG Xianyuan, ZHAI Guojun, SUI Lifen, HUANG Motao. Application of Least Square Support Vector Machine to Detecting Outliers of Multi-beam Data[J]. Geomatics and Information Science of Wuhan University, 2010, 35(10): 1188-1191. |
| |
| Authors: | HUANG Xianyuan ZHAI Guojun SUI Lifen HUANG Motao |
| |
| Affiliation: | 1Institute of Surveying and Mapping,Information Engineering University,66 Middle Longhai Road,Zhengzhou 450052,China;2Naval Institute of Hydrographic Surveying and Charting,40 Youyi Road,Tianjin 300061,China |
| |
| Abstract: | In order to solve the problem of trend surface conformation,a new method of constructing trend surface by LS-SVM is presented,and then outliers of Multi-beam data could be eliminated by the trend surface.In order to illuminate the correctness and rationality so a contrast between this method and the approach of trend surface filter.The theorem proves that the trend surface filter is the especial result of LS-SVM.The example shows that in the process of constructing trend surface by LS-SVM,the weight parameters could be adjusted,so the trend surface have the property of popular and steady,the outliers of Multi-beam data could be eliminated effectively. |
| |
| Keywords: | LS-SVM trend surface filter weight parameters outliers |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《武汉大学学报(信息科学版)》浏览原始摘要信息 |
|
点击此处可从《武汉大学学报(信息科学版)》下载全文 |
|
