Research On Geomorphologic Visualization Based Onlaser Ranging Technology

Visualization technique of 3D laser scanning is based on the data collected on the target surface by three-dimensional laser radar,and realizes the reconstruction technique of the target object in the computer through pretreatment,surface reconstruction and other processes.This technique is developing rapidly with the lidar technology,and has received attention and recognized by worldwide countries in the practical application of military vehicles,aircraft and so on.There are many research works on point cloud reconstruction,but the sparse point cloud reconstruction based on single-line lidar acquisition is deficient.In this paper,the pretreatment and visualization technique of scattered point clouds will be discussed by using sparse point clouds as experimental data after the structure analysis of single-line lidar.This paper is aimed at the laser radar point cloud simulation.The acquisition principle and structure of the developed single-line pulsed laser radar are analyzed,and the radar model is built and the three-dimensional scanning platform of single-line pulsed laser radar is built.The point cloud simulation method for single-line lidar 3D scanning mode is proposed and improved.The experimentally successful 3D point cloud experimental data of any model was obtained.For data preprocessing and visualization.For the problem that the ground random noise amplitude in the collected geomorphic data has too large influence on the modeling effect,the idea of separating the ground alone is adopted to realize the processing method of leveling the ground data without affecting the target object.In the point cloud process of separating trees,an identification method based on detecting point cloud normal vectors and querying the number of ground points in the statistical range is found.For the problem of no display or reconstruction error in single-line outlier feature point recognition and visualization,the method of feature extraction and neighboring point statistics is proposed to find single-row feature points,and the random point method is added to achieve the reconstruction in the vicinity of single-line feature points.The purpose of displaying a single line feature.In the process of reconstruction,the direct surface fitting between different targets will affect each other.The idea of surface fitting by fitting the parameters after interpolation is used to solve the problem of mutual influence in the overall interpolation and successfully separate the ground target.With the ground.Finally,the entire landscape visualization system is integrated to produce a system that includes landform visualization and point cloud simulation.In the system,C# is used to call MATLAB three-dimensional point cloud map to realize the three-dimensional visualization of the system software.It has been proved that it can run stably.