Font Size: a A A

Modeling Methods And Application For Large-Scale LiDAR Data

Posted on:2011-09-06Degree:MasterType:Thesis
Country:ChinaCandidate:Y HuangFull Text:PDF
GTID:2218330338466835Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Light Detection and Ranging (LiDAR) has made it become much easier to obtain huge mass of terrain data. With intensive application, digital terrain model (DTM) need to be created generally for a large area. In some application areas, because of the limitation of memory and computation capacity, it is a big challenge, memory limit, to create TIN for mass LiDAR clouds data. So as to create D-TIN faster and better, how to effectively and fully make use of compute resource has become an imminent task. The main purpose of this paper is to seek effective methods, which can improve the efficiency of the TIN algorithm about huge mass LiDAR data and also can realize terrain visualization. The main research work of this paper can be summarized as follows:(1) The characteristics and organization of the LiDAR data are analyzed; the methods of creating digital elevation model (DEM) and some problems are summarized; besides, the possibility for improving the efficiency of TIN algorithm by using highly spatial coherence of LiDAR data is also discussed.(2) The concept of Delaunay triangle and its features, together with the algorithm ideas and implementation steps about conventional D-TIN incremental insertion algorithm and the concept and idea of local optimization are introduced. What's more, the merits and drawbacks of several geography oriented three-dimensional models are compared, and the object-oriented modeling method is introduced.(3) The advanced triangulation algorithm based on stream processing proposed by Dr. Isenburg is further studied in this paper. Meanwhile, the algorithm of building point stream with spatial finalization and the algorithm of identifying final triangles are analyzed and concluded in detail, and the procedure of these algorithms are put forward. In local triangulation optimization algorithm, the LOP algorithm which focus on the directed edges is analyzed, and also Qi function is proposed to improve this algorithm. Besides, storing the generated TIN information in OBJ format can increase the transplantation and universality. With VC++6.0, the above mentioned algorithm is realized, and by comparing this algorithm with Classic-by-point insertion and Quadtree-based incremental insertion algorithm, it is easy to find that the time complexity of this algorithm is close to the linear complexity. Finally, time complexity is analyzed.(4) In this paper, a solution based on existing three-dimensional modeling technical characteristics and OBJ standard model files and object-oriented 3D TIN data model is put forward for simulating real terrain quickly. Moreover, TIN visualization is achieved by using OpenGL and VC++6.0, thus, the accuracy and efficiency of this model is proved.In this paper, improvement of Based on the introduction and study of abroad advanced algorithm, and combined with the project needs, the algorithm is improved in this paper, that is storing the output results in OBJ files and modeling the TIN using the object-oriented approach. This method has certain practical value in quickening the speed of 3D modeling and enhancing the popularity of three-dimensional model.
Keywords/Search Tags:LiDAR, TIN, Stream Processing, Algorithm, OBJ File, Object-oriented
PDF Full Text Request
Related items