融合航空影像的LiDAR地物点云分类

Classification of LiDAR object points by fusing aerial image

点云滤波分类是LiDAR后续应用的基础工作,在点云滤波的基础上,以航空影像为辅助条件,结合点云高程信息,设计一套地物点云的分类方法。该方法首先融合航空影像与LiDAR数据,将对应RGB值赋予每个点,根据植被的光谱特征提取出部分植被点云;然后再根据文中定义的点云高程纹理,在剩余地物点云中提取出建筑物点,最后根据回波次数信息分离出剩余植被点,完成地物点云的分类。采用北京凤凰岭地区一组机载LiDAR数据进行实验。实验结果表明,该方法能够有效地将地物点云进行分类并且满足一定的精度要求,具有一定的实用价值。

Filtering and classification of point cloud is the base of LiDAR data application.A classification method of object point data is put forward in this paper.In this method,fusion of LiDAR data and aerial image is done firstly.Then,the vegetation points are attracted according to their spectrum feature.In succession,building points are classified from the remainder points according to the height texture.Finally,the remained tree points are attracted according to echo information.An experiment is performed with a group of LiDAR data in Beijing.The results presented in this paper have shown that the method is efficient and available.