Massive Vehicle Point Cloud Data Organization And Rapid Visualization Technology Research

In recent years, with the development of Geo-spatial information industry,3D GIS is more and more popular, which is the expansion of two-dimension GIS in the Z space. It is more intuitive to express the real world and the surface of the morphology, and can locate objects accurately with the information of the surface and contour of the entity, it also provides much rich and useful information by multi-angle rotating three-dimension surface features. With all these advantages,3D GIS will certainly be an important direction for the development of GIS. However, no matter what the three-dimensional data source is, large-scale data has become an important bottleneck of its development. How to manage and visualize the large3D data effectively has become the key problem. In this paper, we propose novel methods to manage and visualize the3D laser point cloud data.First, a method based on octree and pyramid structure of data management is proposed, which includes the building of the pyramid structure, the structure of the octree and the index based on the grid of the latitude and longitude of the earth body. Through the massive data pre-processing of point cloud data tiles of cut tissue, and then create the index file in order to achieve efficient management.Second, in order to meet the requirement of fast visualization, we describe the technology of retrieval and scheduling of the data, including quick retrieval the best point cloud data layer and the data tiles, and locating the best viewing area.Finally, we give our solutions to these problems, sum up the research results and put forward the prospect of future work.We develop a mass point cioud data rapid visualization system using the C++programming language, based on the platform of Visual C++and OpenGL3D graphics library. With the data obtained by SSW, the experiment shows the practical utility of our methods.