Research On Preprocessing Algorithms Based On Point Cloud Data

As a high-tech in the 21 st century,3D laser scanning technology has the advantages of non-contact,high precision,fast data collection and full collection of information.It plays a huge role in reverse engineering,cultural relic protection and some rapid emergency industries.It also brings another big change to the surveying and mapping industry.However,with the arrival of the "Internet +" era,the improvement of data volume puts forward new requirements for data processing speed.The traditional working mode of surveying and mapping "Perform measurement work first,then perform data processing" is inefficient.Some cumbersome point cloud processing software can perform point cloud processing,but it requires higher professional knowledge of operators,and the accuracy cannot be guaranteed.How to realize fast processing of massive point cloud data under the premise of ensuring certain accuracy,so that the public can see the digital results and realize real-time surveying and mapping work has become a major problem to be solved urgently.In this paper,aiming at this problem,the working principle,workflow and parameter configuration of different equipment of 3D laser scanning technology are studied and studied in depth.Based on ground-based three-dimensional laser scanner and Visual Studio 2015 as the programming platform,point cloud denoising,point cloud registration and model reconstruction are realized.Firstly,the study and research of point cloud data format are carried out,and the in depth study of its solution for different point cloud data formats.Aiming at the scattered point cloud data,combining with several examples of point cloud,a series of research and study are carried out on the denoising of point cloud data,registration of point cloud,reconstruction of point cloud model and texture mapping of model.In the process,this paper solves the problem of time-consuming extraction of feature points of large-point cloud data by sampling the point cloud before feature point extraction.And model it in Geomagic Studio software and display its precision.The results show that,the processing efficiency of point cloud data and the accuracy of point cloud after processing basically reach the expected target similar to professional software.