城市复杂地形环境下倾斜模型结合LiDAR点云的大比例尺地形图更新方法北大核心CSCD

针对无人机倾斜摄影技术受遮挡影响较大和难以穿透植被茂密地区的问题,本文提出了城市复杂地形环境下倾斜模型结合LiDAR点云进行小区域大比例尺数字地形图的更新方法。首先采用五镜头六旋翼无人机分别以垂直和平行主要建筑物楼群方向进行2次全区域拍摄,以及无人机机载激光雷达全区域采集点云,并对高度不足10 m的别墅区进行单镜头低空补飞。然后融合倾斜影像点云与机载激光点云建模,经过3种建模方案对比,融合建模的倾斜三维模型的位置精度和模型质量均最优。最后基于此模型进行测图。精度评定结果表明,城市复杂地形环境下在飞行方案和像控点布设合理的情况下,通过倾斜三维模型采集的数字地形图的平面和高程精度完全满足并优于深圳市1∶1000数字地形图动态更新的精度指标。

Large scale topographic map updating method based on tilt model and LiDAR point cloud in urban complex terrain environment

UAV tilt photography technology is greatly affected by occlusion and is difficult to penetrate the densely vegetated area, this paper proposes a method to update the large-scale digital topographic map of a small area with the tilt model combined with LiDAR point cloud in the urban complex terrain environment. Firstly, the UAV of five lens and six rotor is used to take two full-scale photos of the experimental area in the direction of vertical and parallel to the main buildings, and the unmanned airborne laser radar collects the point cloud in the whole area, and carries out a single lens low altitude make-up flight for the villa area less than 10 m. After aerial triangulation, the inclined image point cloud and airborne laser point cloud are fused for modeling. Comparing three modeling schemes, the inclined 3D model of the fusion modeling is the best in position accuracy and model quality. Finally, the mapping is carried out based on this model. The accuracy evaluation shows that under the condition of reasonable flight scheme and image control point layout in the urban complex terrain environment, the plane and elevation accuracy of the 1:1000 digital topographic map collected by the inclined three-dimensional model fully meets and is better than the accuracy index of the dynamic update of the 1:1000 digital topographic map in Shenzhen.