基于无人机影像密集匹配点云的传统村落地面点提取及DEM生成——以湘西德夯村为例

目前,针对利用无人机技术在山地起伏大、山体植被密集区域,难以获取地面点及DEM等问题,本文提出了一种结合布料模拟算法和改进的局部最大值算法,利用树顶点、树高等植被信息,提取地面点,进而生成整个区域的DEM的方法。以中国传统村落德夯村为例,利用植被系数和高程信息将点云分割为植被密集区和非植被密集区两个部分。在非植被密集区,通过布料模拟算法和改进的局部最大值算法分别提取地面点和树顶点,计算平均树高;在植被密集区,通过该区域的树顶点推算得到植被密集区的近似地面点,最终将两部分的地面点云进行TIN插值得到该地区的DEM。试验结果表明,利用此方法生成的DEM均方根误差,在非植被密集区达0.037 m,植被密集区可达1.606 m,整体平均误差达1.492 m,总体精度较好,基本可以满足村落尺度空间分析的需求。

Ground point extraction and DEM generation of traditional village landing surface points based on dense matching point cloud of UAV image: taking Dehang village in western Hunan as an example

For the problems of using UAV technology to obtain ground points and DEM in mountainous area with large fluctuation and dense vegetation, a method combining cloth simulation filtering algorithm and improved local maximum algorithm is proposed.This method uses vegetation information such as tree vertices and tree heights to extract ground points, and then generates DEM for the entire area.In this paper, Dehang village, a traditional village, is taken as an example. The point cloud is divided into two parts:dense vegetation area and non-dense vegetation area by vegetation coefficient and elevation information.In non-dense vegetation area, the cloth simulation filtering algorithm and improved local maximum algorithm are used to extract ground points,tree vertices and calculate the average tree height; in dense vegetation area, the approximate ground point of the dense vegetation area is calculated from the vertex of the tree in the area, and finally perform TIN interpolation on the two parts of the ground point cloud to get the DEM of the area. The experimental results show that the root mean square error of DEM generated by this method can reach 0.037 m in non-dense vegetation areas, 1.606 m in dense vegetation areas, and the overall average error can reach 1.492 m. The overall accuracy is good, which can basically meet the needs of spatial analysis at the village scale.