论文部分内容阅读
结合山区道路的空间分布特点和激光点云特征,提出了一种从机载LiDAR数据中快速提取山区道路的方法。首先,利用形态学滤波方法进行点云滤波,以去除原始数据的非地面点(建筑、输电线路以及植被等)。在此基础上,采用基于多规则区域生长算法提取道路点并进行优化、然后采用Freeman链编码方法定位追踪道路边界,并利用数学形态学方法进一步细化道路中心线,进而提取完整的道路信息。利用山区机载LiDAR点云数据进行试验并与其他方法的处理结果进行比较,结果表明:本文方法能够有效地从激光点云中提取道路信息:提取道路的完整度为93.87%,正确率为93.84%,质量为88.43%。
Combining with the spatial distribution of mountain roads and the characteristics of laser point cloud, a method of rapidly extracting mountain roads from airborne LiDAR data is proposed. First, point cloud filtering is performed using morphological filtering to remove non-ground points (buildings, transmission lines, vegetation, etc.) of the original data. Based on this, the road points are extracted and optimized based on the multi-rule region growing algorithm. Then the Freeman chain coding method is used to track the road boundary. The mathematical morphology method is used to further refine the road centerline and extract the complete road information. The experimental results on mountain airborne LiDAR point cloud data are compared with those of other methods. The results show that the proposed method can effectively extract road information from laser point clouds: the completeness of the extracted road is 93.87% and the correctness is 93.84 %, The mass is 88.43%.