融合激光点云与影像点云的精细化DEM生成

针对地面激光扫描及无人机航摄技术在实际外业测量中受视场角限制或遮挡等因素的影响而难以获取待测区域完整的点云数据的问题，本文在经典ICP算法的基础上，提出了一种顾及高程差异和点云密度的激光点云与影像点云融合方法。通过差分数字高程模型对点云进行分块，并基于点云密度选取融合范围，将分块后的影像点云配准到激光点云的孔洞和稀疏区域。本文方法能够提高激光点云与影像点云的融合效果，保持激光点云的精度并保留更多的细节特征，实现激光点云与影像点云的高质量融合。

Refined DEM generation based on fusion of laser point cloud and image point cloud

In view of the problem that the ground laser scanning and UAV aerial photography technology is difficult to obtain the complete point cloud data of the area to be measured due to factors such as field angle limitation or occlusion in actual field measurement, based on the classical ICP algorithm, this paper proposes a fusion method of laser point cloud and image point cloud considering the elevation difference and point cloud density, divides the point cloud into blocks by differential digital elevation model, selects the fusion range based on the point cloud density, and registers the segmented image point cloud to the holes and sparse areas of the laser point cloud. The method in this paper can improve the fusion effect of laser point cloud and image point cloud, keep the precision of laser point cloud and retain more detailed features, and realize the high-quality fusion of laser point cloud and image point cloud.