【摘要】 针对城区大规模异源点云数据配准效率低的问题，提出一种城区机载和车载点云的快速配准算法。该算法首先以建筑物特征直线为基础，利用矩形邻域搜索对应线对，提高对应线对搜索效率；其次以线线间对应关系构建配准模型，完成平面配准；最后按1 s间隔选取车载轨迹点下方对应的高程点对，完成高程配准，减少搜索高程点对的时间和参与匹配的点对数量，提高配准效率。实测数据结果表明，该算法在平面距离偏差均值和均方根误差分别为17.231 mm, 23.874 mm,配准总用时为2.67 min,该算法可以实现城市区域机载和车载点云快速配准的同时保持有较高的配准精度。更多还原

【Abstract】 To solve the problem of low registration efficiency of large-scale heterogeneous point cloud data in urban areas, a fast registration algorithm for urban airborne and vehicle-borne point clouds is proposed. First, based on building feature lines, the algorithm searches for corresponding line pairs using rectangular neighborhood to improve the search efficiency of corresponding line pairs; then, the registration model is constructed based on the corresponding relationship between lines to complete the plane registration; finally, it selects the corresponding elevation point pairs below the vehicle track points at 1s interval to complete elevation registration, reduces the time for searching elevation point pairs, reduces the number of point pairs involved in matching, and improves the registration efficiency. The measured data show that the mean and root mean square errors of the plane distance deviation of the algorithm are 17.231 mm and 23.874 mm respectively, and the total registration time is 2.67 min. This algorithm can realize the rapid registration of airborne and vehicle-borne point clouds in urban areas, while maintaining a high-level registration accuracy.更多还原