摘要
经验模态分解(EMD)是希尔伯特-黄变换(HHT)过程的重要部分,其核心技术难点为端点延拓.本文基于目前各种EMD端点处理方法分析基础上,紧紧围绕如何防止端点发散这么一个关键点,开发了一种高效简便的EMD固定端点延拓新方法.利用该新方法,对线性仿真信号、非线性仿真信号和合成孔径雷达(SAR)遥感图像真实信号进行了经验模态分解,分解结果表明,利用EMD固定端点延拓新方法可以把每次分解的结果误差限制在较小的范围,对信号不会产生较大的污染,而且得到的分解结果能有效反映信号的真实调制,在实际SAR遥感图像应用过程中在具有很好的有效性和可行性.
Empirical Mode Decomposition(EMD)plays the most important role in the Hilbert-Huang Transform(HHT),and the most technical difficulty during the mode decomposition is the boundary process.Based on the analysis of all kinds of EMD boundary process methods,this paper has developed one new efficient and simple method of EMD boundary process method,which is closely around the key point question of preventing the boundary signal from diverging.Using this new efficient and simple method of EMD boundary process method,the mode decompositions of linear simulation signal,nonlinear simulation signal and real signal of Synthetic Aperture Radar(SAR)remote sensing image are carried out.The decomposition results show that,for the linear signal decomposition processing,the decomposition results are better than the extremum continuation method,and for the nonlinear signal decomposition processing,the decomposition results are better than the other previous research results.And also as for the asymmetric signals,the resolution accuracy of the decomposition results is better than that of the mirror closed continuation method,which is equivalent to that of extremum extension analysis method.And for the symmetric signals,the resolution accuracy of the decomposition results is better than the extremum extension analysis method,which is equivalent to that of mirror closed extension analysis method.As for the decomposition processing of the real SAR oceanic internal wave signal,due to the fact that the actual signal of SAR oceanic internal wave has the bad symmetry and the influence of noise,the boundary process method of extremum extension analysis method can't really reflect the signal of oceanic internal wave,which leads to the limitation of characteristic extension for mirror closed extension method and extremum extension analysis method.Using this new efficient and simple method of EMD boundary process method gave in this paper,we not only can limit the error to a small range during each iteration process,but also can reduce the influence of noise and reflect the true modulation of the real signal,which shows this new efficient and simple boundary process method is very effective and feasible for processing SAR remote sensing image.
作者
范开国
梁建军
韦道明
王亚锋
方芳
鱼蒙
FAN KaiGuo;LIANG JianJun;WEI DaoMing;WANG YaFeng;FANG Fang;YU Meng(Army 32021,PLA,Beijing 100094,China;Key Laboratory of Digital Earth Science,Research Institute of Aerospace Information,Chinese Academy of Sciences,Beijing 100094,China;State Key Laboratory of Satellite Ocean Environment Dynamics,Second Institute of Oceanography,Ministry of Natural Resources,Hangzhou 310012,China)
出处
《地球物理学进展》
CSCD
北大核心
2021年第2期458-463,共6页
Progress in Geophysics
基金
国家自然科学基金(41876201)资助.
关键词
希尔伯特-黄变换
经验模态分解
端点延拓
合成孔径雷达
海洋内波
Empirical Mode Decomposition(EMD)
Hilbert-Huang Transform(HHT)
Boundary process method
Synthetic Aperture Radar(SAR)
Oceanic internal wave
