基于知识的车载LiDAR地物自动分类

车载LiDAR已广泛应用于三维数字城市建模、道路信息数据采集等领域。海量点云信息中不同地物目标的自动识别和分类是LiDAR数据后处理的难点之一。根据不同地物目标物理特性、空间拓扑关系及其在点云中的相关特征知识，建立地物分类规则，依据分类知识进行地物自动识别和分类。通过实测数据分类试验，证明该方法可以较好实现建筑物、树木、线杆、行人等不同地物的自动识别和分类。

Knowledge-based automatic objects classification from mobile LiDAR data

The mobile LiDAR has been widely applied in data acquisition for 3D city reconstruction and road corridors. It is one of the key problems for LiDAR data processing that how to classify different kinds of objects automatically from the mass points cloud. A method is proposed to classify objects automatically based on knowledge of objects classification rules. The knowledge can be acquired by analyzing the objects’ physical characteristics, spatial topological relationships and those relevant characteristics on LiDAR data points. An experiment is conducted with real LiDAR data, and the results show that the knowledge-based classification method can automatically and successfully recognize and classify different objects, such as buildings, trees, street lamps and pedestrians.