从DSM数据中自动提取建筑物的方法研究

The Research about Automatic Extraction of Buildings from DSM

现今,DSM的获取日趋方便、快捷、精确,充分利用DSM所提供的信息,可以解决很多实际问题。本文提出了一种从DSM自动提取建筑物的方法,首先从DSM中提取DTM,差值后得到消除了地形影响的规格化的DSM(即NDSM),然后对NDSM进行改进型"标记控制的分水岭"分割,得到地上物目标;最后根据这些目标的各自特点进行过滤,祛除非建筑物对象,留下我们所要的建筑物对象。

Nowadays, the acquisition of DSM （ Digital Surface Model） is becoming more convenient and more accurate. It is very probable that lots of practical problems would be solved well by making the best of the information from the DSM. This paper concerns an algorithm about the automatic extraction of buildings from the DSM. First, the DTM （Digital Terrain Model） is extracted from the DSM, and then the Normalized - DSM is derived by subtracting DTM from DSM. It only presents the object heights in the ground. Second, the NDSM is segmented using the algorithm of the marker - controlled watershed, and the off - ground objects are obtained. Finally, filtering the targets with their own particular characters, the no - buildings are removed from the off - ground objects.