作者: Wu Qinqin , Qiang Shengzhi , Wang Yuanqing , Ren Shuping
DOI: 10.1364/AO.58.007943
关键词:
摘要: The light detection and ranging (LIDAR) full-waveform echo decomposition method based on empirical mode (EMD) the local-Levenberg–Marquard (LM) algorithm is proposed in this paper. can decompose into a series of components, each which be assumed as essentially Gaussian. original decomposed intrinsic functions (IMFs) final residual by using EMD first. Then, average period (Tm¯) corresponding energy densities (EDs) all IMFs are calculated. A suitable IMF selected relationship between EDs white-noise theoretical spread lines 99% confidence-limit level. components detected according to positions maxima IMF. initial parameters estimated local-LM fitting. fitted global-LM Compared traditional (zero-crossing) ZC method, has strong anti-noise performance. It precisely detect estimate components. verified synthetic data; coding LIDAR recorded Land, Vegetation, Ice Sensor data.