Design And Development Of The Laser Radar Point Cloud Processing And 3D Visualization Platform

Lidar has been widely used in the commercial field,and the research and development of lidar data acquisition and processing system has made great progress.The data obtained by laser radar scanning is a large amount of discrete data,and generally is called point cloud data.Point cloud data is a data set that describes the three-dimensional spatial information of real objects,with three-dimensional coordinates,color value and strength.The lidar data acquisition and processing system has a lot of noise and holes due to external accidental factors or its own acquisition accuracy & mechanical vibration when scanning objects,and the data density is quite large,so it is difficult to use the data directly to describe the three-dimensional information of the real object.Therefore,we need to design and develop a piece of software that processes and visualizes point cloud data.At present,the point cloud data processing and visualization software is basically matched with the data acquisition system and the price is relatively high.Besides,several pieces of independent software also hold some limitations to our needs.Thus,a piece of point cloud data processing and visualization software has been designed for the independent research and development lidar data acquisition and processing system in this paper.Based on PCL,the software combines with VS2010 program development platform to complete the data processing module of the software and designs the interface of the software through the user graphical interface development tool Qt.In this paper,firstly we studied the characteristics of point cloud data collected by lidar data acquisition and processing system and the storage structure of point cloud data.According to these characteristics,the processing algorithms of point cloud data are studied,including simplification,filtering,retrieval,normal estimation,curvature estimation,coordinate transformation,surface reconstruction and visualization of point cloud data.After that,according to the actual demand,the framework of the point cloud data processing and 3D visualization software,the basic process of point cloud data processing,the function module of software and the development platform of software are designed,and the function of each module of software is realized.The main functions of the software are point cloud data input and output,using VTK and PCL to achieve point cloud data visualization and user interaction,the use of kd-tree to achieve the nearest neighbor search,the use of least squares method to complete the normal estimate,to simplify the data by the down-sampled voxel grid method,a variety of filtering algorithms are implemented to realize the data filtering of different characteristics,including Statistical Outlier Removal,Radius Outlier Removal,Conditional Removal,Pass Filtering,Bilateral Filtering and Sliding Least Squares Filtering,the use of Greedy Projection Triangulation and Poisson surface reconstruction algorithm to achieve 3D surface reconstruction of point cloud data.Finally,several sets of test data are used to test and analyze the function,display effect and processing efficiency of the software and we combined the project undertaken by the laboratory to verify the software to the point cloud data processing and display functions.