This paper deals with some techniques that can be used to improve quality and geometric consistency of data driven 3D city models. It focuses on two practical tasks with respect to roof polygons: finding better edges and planarization. To increase the quality of edges, models are merged with true orthophotos and cadastral data. With mixed integer linear programs, the number of right angles and symmetry are maximized. Linear optimization problems also help to correct non-planar roof facets and to avoid artificial step edges within roofs. Since these methods change the positions of vertices, self-intersections of polygons have to be healed, and building walls have to be re-arranged. The techniques focus on roof geometries that can be represented in 2.5D. In general, this is true for CityGML models in level of detail 2 that do not have detailed facades or overhangs. However, techniques can also be adapted to other 2.5D modeling tasks.

本文讨论了一些可用于提高数据驱动3D城市模型的质量和几何一致性的技术。它着重于与屋顶多边形有关的两个实际任务：找到更好的边缘和平面化。为了增加边缘的质量，将模型与真实的正射照片和地籍数据合并。使用混合整数线性程序，可以最大程度地增加直角和对称性。线性优化问题还有助于校正非平面屋顶面，并避免在屋顶内形成人为的台阶边缘。由于这些方法会改变顶点的位置，因此必须修复多边形的自相交，并且必须重新布置建筑物墙。这些技术着重于可以用2.5D表示的屋顶几何形状。一般来说，对于细节级别2中没有详细外观或悬垂的CityGML模型，这是正确的。但是，技术也可以适用于其他2.5D建模任务。