基于点云融合算法的无人机三维建模测量研究  

Research on UAV 3D Modeling and Measurement Based on Point Cloud Fusion Algorithm

作　　者：何跃[1] 张少斐 HE Yue;ZHANG Shaofei(Puyang Earthquake Monitoring Station,Puyang 457001,China)

机构地区：[1]濮阳市地震台,河南濮阳457001

出　　处：《微型电脑应用》2024年第6期172-175,共4页Microcomputer Applications

摘　　要：坐标残差过大是导致无人机三维建模准确性不达标的主要原因,为避免上述情况的发生,针对基于点云融合算法的无人机三维建模测量方法展开研究。确定2D样点、3D样点与目标视点的云对应关系,通过求解双线性融合差值的方式,建立点云融合算法模型,再联合无人机目标轮廓点定义条件,完成对目标Hu距的匹配,得到必要的无人机目标节点,实现基于点云融合算法的无人机目标提取。建立三维测量坐标系,并根据光学畸变差的取值范围,纠正三维影像,结合相关测量节点,推导点算子计算式,从而确定测量值平差的取值范围,完成无人机三维建模测量方法的设计。实验结果表明,点云融合算法作用下,X轴、Y轴、Z轴三个方向上的坐标残差均不超过10%,能够实现无人机三维测量模型的准确构建。Excessive coordinate residuals are the main reason for the substandard accuracy of UAV 3D modeling.In order to avoid the above situation,the UAV 3D modeling and measurement method based on point cloud fusion algorithm is studied.The cloud correspondence between 2D sample points,3D sample points and target viewpoint is determined.The point cloud fusion algorithm model is established by solving the bilinear fusion difference,and then is combined with the definition conditions of UAV target contour points to complete the matching of target Hu distance,then the necessary UAV target nodes can be obtained,and the UAV target extraction can be realized based on the point cloud fusion algorithm.A 3D measurement coordinate system is established,3D images are corrected according to the value range of optical distortion,and the calculation formula of the point operator is derived by combining the relevant measurement nodes,so as to determine the value range of the measurement adjustment,and complete the design of the 3D modeling and measurement method of the UAV.The experimental results show that under the action of point cloud fusion algorithm,the coordinate residuals of X axis,Y axis and Z axis are less than 10%,which can realize the accurate construction of UAV 3D measurement model.

关 键 词：点云融合算法 三维建模 目标Hu距 光学畸变差 影像纠正 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]