Database Construction And Management Based On 3D Laser Scanning Data

With the development of Information Technology and the advent of Digital Earth and Virtual City, the demand for three-dimensional spatial information has become more and more urgent and the total amount of cumulate three-dimensional spatial data has being increasing. Because of its characteristics of high precision, fully automatic, optimization notion and three-dimensional,3D scan data, different from the traditional data acquisition, gain highly attention and application.3D scan data can be used in the field of not only high precision engineering modeling and reconstitution, but also protection of old architecture. A problem need to be solved is how to manage and store these data efficiently. Hence, the study of3D scan data management and maintenance has become the focus of research in the field of processing of3D scan data.In this paper, start from point of database, analyse and compare of advantages and disadvantages of each database products, determine to use the Oracle database as platform, through the analysis of the characteristics of3D laser scan data and its data structure, design and implement the technique of efficient management and storage. The main contents in this paper as follows. First, analyse and expound the research status and characteristics of3D scan point cloud data at home and abroad. Expound the whole system and internal data structure and the advantage of managing massive data. Unique internal components, Spatial object type, has been proposed for organizing and managing spatial data. Second, in practical work, because of their characteristics of3D laser scan data, rate of data access of massive point cloud data will become very slow. To solve this problem, accord to mature internal indexing mechanism of Oracle, establish index in the table, as well as propose the scheme which use Spatial combines the traditional database indexing—B tree, Hash structure and spatial database indexing mechanism base on linear four fork tree to manage spatial data and improve the retrieval efficiency. Approximately, make the efficiency operation and management for massive point cloud data come true, than improve the performance of3D database management.