Research On Compress Method For LiDAR Point Cloud Data Of Broadleaf Tree

As a new technology of surveying and mapping, the terrestrial LiDAR tact, and other characteristics. It has been used in many fields, for example, digital city, protecting of ancient architectural structures, measurement engineering. 3D laser scanner can quickly scan the object and obtain 3d coordinate from the surface of object, and the point cloud data of scanning object quantity is very enormous. Such a large number of point cloud data is very not conducive to the next step in the transmission and the computer storage, and also set up obstacles for the development of the follow-up work. So full compression processing point cloud data has become the necessary processing steps and highlights the important status. We selected a kind of broadleaf tree(magnolia tree) in the campus of University of Electronic Science and Technology of China(UESTC) for the study. With the point cloud data obtained by terrestrial LiDAR, we introduced a point cloud degree compress method in this thesis. The main work and conclusions are as follows:(1) Data acquisition and pre-processingWe researched on the method of experimental data collection outdoor using the Leica Scan Station C10 laser scanning system based on the basic theory of three-dimensional laser scanning and acquired the experimental data of single tree and more trees in the area. Combined with the performance characteristics of the scanner, we also researched the registration method of multiple point cloud data from different azimuthally positions and the de-noising method.(2) Compression method evaluation of broadleaf tree point cloud dataRandom sampling compress method, bounding box barycenter sampling compress method and common vertex compress method was studied and implemented, and compression experiments were done with broadleaf tree(magnolia tree) point cloud data. Experiment of the three compression methods were completed in different compression parameter, the experimental results are compared with the method of image display and 3D modeling. We make full use of the visual effect to carry on the compression experiment evaluation. At last, the diagram on compression ratio and compression parameter was established and contrastive studied.(3) Putting forward the point cloud degree compress methodThe vertex connectivity of graph theory plays an important role in the connectivity of graph. Basing on the theory of the vertex connectivity, we will analyze the enormous amount of point cloud data embedded in the controllable finite graphics. The vertex connectivity is the important parameter in the application of point cloud data processing. Series data compression experiments of point cloud degree compress method were done with broadleaf tree(magnolia tree) point cloud data. And through the comparative analysis of random sampling compress method and bounding box barycenter sampling compress method and common vertex compress method, verify the point cloud degree compress method has the advantage of accuracy and simplify in the compression of scattered point cloud data.(4) Programming of a point cloud data compressionPoint cloud degree compress method, random sampling compress method, bounding box barycenter sampling compress method, common vertex compress method was realized on the.net platform by C++ or MATLAB coding.