激光雷达回波信号多尺度增强方法研究  

Research on multi-scale enhancement methods for LiDAR echo signals

作　　者：王芳 宋斌 袁浩 WANG Fang;SONG Bin;YUAN Hao(School of Computer and Software Engineering,Sias University,Xinzheng 451150,China;Henan Intelligent Manufacturing Digital Twin Engineering Research Center,Xinzheng 451150,China;School of Information Engineering,Henan University of Science and Technology,Luoyang 471003 China)

机构地区：[1]郑州西亚斯学院计算机与软件工程学院,河南新郑451150 [2]河南省智能制造数字孪生工程研究中心,河南新郑451150 [3]河南科技大学信息工程学院,河南洛阳471003

出　　处：《激光与红外》2025年第1期60-66,共7页Laser & Infrared

基　　金：河南省2021年省级民办教育发展专项资金资助项目(No.豫财教[2021]16号软件工程专业);2023年河南省科技厅科技攻关项目(No.232102210137)资助。

摘　　要：由于目标物体的形状、大小、距离等因素的影响,激光雷达回波信号中的特征信息会分布在不同尺度上,且信号经历多径效应后,导致信号相位扭曲和叠加,使单纯依靠小波阈值对整体回波信号进行增强的思路变得过于简单,并且信号的非线性也影响信号增强的效果。提出一种基于CEEMDAN联合小波阈值算法的激光雷达回波信号多尺度增强方法。该方法首先利用形态滤波器去除激光雷达回波信号中含有脉冲、间歇等间断事件;再基于上述处理结果利用CEEMDAN算法对激光雷达回波信号实施有效的尺度分解,以有效地将复杂非线性信号分解成若干个IMF分量,提高信号处理的准确性;利用相关系数指标,获取回波信号中噪声以及信号之间的IMF临界点,使用小波阈值算法对其主导IMF分量实施去噪处理,根据处理结果实现信号的多尺度增强。实验结果表明,利用该方法开展回波信号增强时,信号去噪效果好、增强性能高。Due to the shape,size,distance and other factors of the target object,the feature information in the LiDAR echo signal is distributed at different scales,and the signal undergoes multipath effects,which leads to the signal phase distortion and superposition,making the idea of relying solely on wavelet thresholding to enhance the overall echo signal too simple.Moreover,the nonlinearity of the signal also affects the effectiveness of signal enhancement.In this paper,a multi-scale enhancement method for LiDAR echo signals based on the CEEMDAN joint wavelet threshold algorithm is proposed.Firstly,a morphological filter is used to remove intermittent events such as pulses and intermittency from the LiDAR echo signal.Then,based on the above processing results,the CEEMDAN algorithm is used to implement effective scale decomposition of the LiDAR echo signal to effectively decompose the complex nonlinear signal into several IMF components,and to improve the accuracy of signal processing.Finally,the correlation coefficient index is used to obtain the IMF thresholds between the noise and the signals in the echo signal,and the wavelet thresholding algorithm is used to denoise the dominant IMF components,and the multi-scale enhancement of the signal is realized according to the processing results.The experimental results show that the signal denoising effect is good and the enhancement performance is high when the echo signal enhancement is carried out using this method.

关 键 词：CEEMDAN分解 小波阈值 激光雷达通信 回波信号 多尺度 增强算法设计 

分 类 号：TN958[电子电信—信号与信息处理] TN911.7[电子电信—信息与通信工程]