Research On Noise Suppression And Point Cloud Denoising In Multispectral LiDAR System

The multispectral lidar can acquire the spatial information and spectral information of the measured target simultaneously to form the multispectral lidar point cloud.Compared with single wavelength lidar point cloud,this kind of point cloud has a better visualization effect and higher ability of ground object recognition.However,due to the influence of scanning equipment,detection environment,and operators,there will be noise points in the point cloud.In order to obtain the multispectral lidar point cloud with a better visualization effect and higher precision,the noise in the point cloud must be removed.Thesis,the point cloud obtained by multispectral lidar is taken as the research object,and the denoising research is carried out.The main contents are as follows:(1)According to the principle of multispectral lidar,the data structure of multispectral lidar point cloud is proposed.Compared with the existing singlewavelength lidar point cloud,this data structure has spatial information and contains the color information of the measured object.(2)According to the noise generation mechanism of the multispectral lidar system,some measures to suppress the system's noise are proposed.The system design,the scanning noise,detector noise,data acquisition noise,and synchronous control noise are suppressed.(3)Aiming at the shortcomings of the existing point cloud denoising algorithms,multispectral lidar point cloud denoising algorithm based on color clustering is proposed,the problem that the existing point cloud denoising algorithms were complex to effectively remove the mixed noise points is solved,and the noise points are removed by combining spatial information and spectral information.(4)Simulation data and measured data are used to compare the denoising effect of the multispectral lidar point cloud denoising algorithm based on color clustering with the existing denoising algorithm and the denoising performance of the algorithm is evaluated through quantitative calculation of the denoising accuracy.The experimental results show that the proposed algorithm has higher denoising accuracy,with an accuracy of more than 95%.Thesis,a series of technical solutions for suppressing system noise and a feasible denoising algorithm for a multispectral lidar point cloud is proposed,which will provide technical support for the application and development of multispectral lidar.