论文部分内容阅读
机载激光雷达是DEM生产的重要技术手段之一。为了获取高分辨率的DEM,首先要将建筑物、车辆、植被等非地面点去除,这一过程称为滤波。本文在现有算法的基础上提出一种改进的形态学滤波算法。该算法逐步增大滤波窗口尺寸,并采用自适应的高差阈值,较好地保留了地形拓扑信息。试验采用ISPRS提供的参考数据进行测试分析,该数据代表典型的城市地区。试验结果表明上述算法可有效地去除绝大部分的非地面点。
Airborne lidar is one of the important technical means of DEM production. In order to obtain a high-resolution DEM, we first need to remove non-ground points such as buildings, vehicles and vegetation. This process is called filtering. In this paper, an improved morphological filtering algorithm is proposed based on the existing algorithms. The algorithm gradually increases the size of the filtering window and adopts the adaptive elevation difference threshold to better preserve topological topology information. Tests were conducted using the ISPRS-provided reference data representing typical urban areas. Experimental results show that the above algorithm can effectively remove most of the non-ground points.