潜在连接点集下拓扑推进三维建模算法研究

通过三维激光扫描仪可以获取到表示空间对象的点云数据。通过设置仪器的采样间隔可以获取到表示对象不同精密程度的点集。但是由于离散点之间缺乏相应的拓扑关系,限制了以点为基本图元的模型的应用。以空间离散特征点集为研究对象,研究了一种三维建模算法。算法基于距离较近的点之间比距离较远的点之间存在拓扑连接的可能性更大的前提,对点集进行预处理,然后在此基础上进行快速建模,最终获取到以三角形为基本面片的格网模型。所建立的模型易于进行各项空间分析和操作。该算法也能够有效地应用于更为复杂的建模任务中,以达到一体化建模的目的。

3D Modeling Algorithm Based on Potential Connection Topology

3D laser scanner is widely used in acquiring the point cloud data representing the spatial object. By setting the instrument sampling interval, different level of sophistication of the object can be obtained. However, due to the lack of topological relations between the discrete points, the using of the point cloud model is limited. In this paper, a 3D modeling algorithm was studied which based on the local topology relationships between the points on the object surface. A preprocessing to the dataset was carried to speed up the modeling algorithm, and a package of meshes was finally outputted. The final model could be easily used in spatial analysis and all kinds of operations. The algorithm also could be applied to more complex modeling tasks in order to achieve the integration of modeling purposed.