LiDAR三维重建中基于CSG方法的扩展研究

Expanding research on CSG in 3D reconstruction from LiDAR

针对机载激光雷达(airborne light detection and ranging,Li DAR)数据三维重建中运用构造表示法(constructive solid geometry,CSG)无法对复杂模型自动地进行基元分解、基元模型识别以及模型集成等问题,提出了一种基于CSG扩展的三维重建方法。该方法利用建筑物等高线的分族特性,通过等高线分族实现了CSG方法中复杂建筑物模型的层次划分和基元分解,在此基础上,结合建筑物分层等高线族特征和分层轮廓线的重建结果来辨别基元类型,实现了CSG方法中基元的自动识别,并依据其识别结果采用相应的重建方法完成模型基元的分部重建,最后依据一套有效的模型集成规则实现了复杂建筑物整体模型的自动建模。通过大范围的Li DAR数据进行试验,验证了基于CSG扩展方法对复杂建筑物三维建模的有效性。

To tackling such problems as primitive decomposition, primitive recognition, model integration in 3D model reconstruction with LiDAR data by using the CSG method, this paper proposes an expanding method for CSG. In this method, the clustering property of building contours is used for layers partition and primitives decomposition, then the styles of the primitives is recognized by combining the features of contours clusters and the contour reconstruction results. In this way, the process of primitives automatic recognition in CSG method is achieved. According to the primitive recognition result, the corresponding reconstruction method for the segmentation is selected, and the whole 3D model for complex building is automatically reconstructed by integrating the segmentation models based on a set of effective model integration rules at last. Experiment results of a wide range LiDAR data show that the proposed expanding CSG method is effective in the 3D reconstruction of complex buildings with LiDAR data.