期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2003
卷号：XXXIV-3/W13
出版社：Copernicus Publications
摘要：In this paper, we present a method for the automated generation of 3D building models from directly observed point clouds generated by LIDAR. First, building regions are detected automatically. After that, roof planes are detected by applying a curvature-based segmentation technique. These roof planes are grouped in order to create polyhedral building models. In this grouping process, the shapes of the roof plane boundaries are determined, and the whole model is improved by an overall adjustment using all available sensor information and, if possible, geometric constraints. We describe the existing modules and those still in the implementation phase, and we discuss the issue of integrating aerial images into the reconstruction process to increase the geometric quality of the reconstructed models. Intermediate results are presented for a test site located in the City of Vienna, Austria.
关键词：3D building models; Building extraction; Data acquisition; LIDAR processing; Multi-sensor processing
