| 联合BEMD与LS-SVM的多波束异常数据探测方法 |
| |
| 引用本文： | 赵祥鸿,暴景阳,欧阳永忠,刘聚.联合BEMD与LS-SVM的多波束异常数据探测方法[J].海洋测绘,2017(4):43-46. |
| |
| 作者姓名： | 赵祥鸿 暴景阳 欧阳永忠 刘聚 |
| |
| 作者单位： | ;1.海军大连舰艇学院海洋测绘系;2.海军海洋测绘研究所 |
| |
| 基金项目： | 国家自然科学基金(41474012,41374018) |
| |
| 摘 要： | 针对多波束水深数据非线性、非平稳性的特点,将二维经验模态(BEMD)和最小二乘支持向量机(LSSVM)引入到多波束水深数据异常值探测中,构建BEMD和LSSVM混合模型。首先,利用二维经验模态将多波束水深数据分解为不同频率的若干个本征模态函数;然后,考虑到异常数据处在高频部分,利用最小二乘支持向量机探测高频本征模态函数(IMF1、IMF2)中包含的异常值;最后,综合两组异常值判定原始水深数据中异常数据。通过实验证明方法的可行性且相比单一模型取得更好的异常值探测效果。
|
| 关 键 词： | 多波束测深系统 异常值 本征模态函数 二维经验模态 最小二乘支持向量机 |
Detecting Outliers of Multibeam Sounding Data Based on BEMD and LS-SVM |
| |
| ZHAO Xianghong,BAO Jingyang,OUYANG Yongzhong,LIU Ju.Detecting Outliers of Multibeam Sounding Data Based on BEMD and LS-SVM[J].Hydrographic Surveying and Charting,2017(4):43-46. |
| |
| Authors: | ZHAO Xianghong BAO Jingyang OUYANG Yongzhong LIU Ju |
| |
| Institution: | Department of Hydrography and Cartography,Dalian Naval Academy,Dalian 116018 ,China;Naval Institute of Hydrographic Surveying and Charting,Tianjin 300061 ,China |
| |
| Abstract: | A method of detecting outlier of multibeam sounding based on BEMD and LSSVM is proposed for the nonlinear characteristics of multi-beam depth data.First,decompose the data into a number of functions in different frequencies using the method of BEMD;then,detect the outliers in IMF1 and IMF2 with the method of LSSVM;last,the outliers of original depth data was detected by the two outliers before.The feasibility of the method is proved by experiment,and better than the single model in defecting outliers. |
| |
| Keywords: | |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《海洋测绘》浏览原始摘要信息 |
| 点击此处可从《海洋测绘》下载免费的PDF全文 |
|
