| 基于VMD对SAR海洋内波参数的自动反演 |
| |
| 引用本文： | 张洪生,郑应刚,王有强.基于VMD对SAR海洋内波参数的自动反演[J].海洋工程,2021,39(3):1-10. |
| |
| 作者姓名： | 张洪生 郑应刚 王有强 |
| |
| 作者单位： | 上海海事大学 海洋科学与工程学院,上海 201306 |
| |
| 基金项目： | 国家自然科学基金项目（51679132，51079082）；上海市地方高校基地能力建设项目（17040501600） |
| |
| 摘 要： | 经验模态分解(empirical mode decomposition,简称EMD)是反演海洋内波参数的有效方法之一,但由于EMD存在模态混叠等问题,对海洋内波进行参数反演时会产生一定误差。相比较于EMD,变分模态分解(variational mode decomposition,简称VMD)能够有效地抑制模态混叠现象。为了更好地对海洋内波进行参数反演,提出了一种基于VMD对合成孔径雷达(synthetic aperture radar,简称SAR)遥感图像中的内波参数进行自动反演的方法。该方法先对SAR图像进行Canny处理,获取图像中的内波条纹信息,再根据内波传播方向自动选取灰度剖面;然后利用集合经验模态分解(ensemble empirical mode decomposition,简称EEMD)信号特征自适应分解模态函数的特点,再将分解得到的有效模态数作为VMD中参数K的参考值;最后利用VMD分解后的数据进行内波参数反演。试验结果表明:通过对Canny预处理后的条纹信息进行灰度剖面自动选取,解决了人为选取剖面所可能导致的误差;通过对剖面信号进行VMD处理不仅解决了EMD模态混叠的问题,成功地反演出内波的前导波振幅,而且所反演的结果与EEMD反演参数以及实测资料数据吻合得很好。
|
| 关 键 词： | 海洋内波 SAR Canny 参数反演 经验模态分解 变分模态分解 集合经验模态分解 |
| 收稿时间： | 2020/6/19 0:00:00 |
Automatically extracting parameters of oceanic internal wave from SAR image based on variational mode decomposition |
| |
| ZHANG Hongsheng,ZHENG Yinggang,WANG Youqiang.Automatically extracting parameters of oceanic internal wave from SAR image based on variational mode decomposition[J].Ocean Engineering,2021,39(3):1-10. |
| |
| Authors: | ZHANG Hongsheng ZHENG Yinggang WANG Youqiang |
| |
| Institution: | College of Ocean Science and Engineering, Shanghai Maritime University, Shanghai 201306, China |
| |
| Abstract: | Empirical mode decomposition (EMD) is one of the effective methods for inversion of oceanic internal waves parameters. However, when extracting the parameters of oceanic internal waves with EMD, there are some errors due to the phenomenon of mode mixing. Compared to EMD, variational mode decomposition (VMD) can effectively suppress the phenomenon of modal aliasing. In order to extract the parameters of oceanic internal waves more effectively, an automatic inversion method based on VMD for the internal waves parameters in synthetic aperture radar (SAR) remote sensing images was proposed. The method, as the first step, gets the information of the internal wave stripes through the edge detection algorithm of Canny operator in SAR image, then selects the grayscale profiles automatically according to the direction of internal waves propagation; as the second step, it get the effective modal number with ensemble empirical mode decomposition (EEMD) signal characteristics of adaptive decomposition of modal function, then takes the effective modal number as the reference parameter K in VMD; finally it extracts the parameters of oceanic internal waves using the decomposed data with VMD. The experimental results show that: the gray stripes can automatically be obtained after the pre-treatment of the Canny operator, and resolve the errors that may be caused by the artificial selection of profiles; processing the profile signals with VMD not only solves the problem of EMD modal aliasing, but also inverses the amplitude of the leading wave of the internal waves successfully, and the inversion results are in good agreement with the EEMD inversion parameters and the measured data. |
| |
| Keywords: | oceanic internal waves SAR Canny parameter inversion VMD EEMD EMD |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《海洋工程》浏览原始摘要信息 |
| 点击此处可从《海洋工程》下载免费的PDF全文 |
|
