基于多源数据融合的弃渣场体积高效定量技术研究OA
Research on efficient quantitative technology for spoil disposal site volume based on multi-source data fusion
为提高弃渣场体积监测精度,解决单一监测手段数据格式不统一、标准不一致的问题,本研究基于地形测绘学原理、空间坐标匹配理论及精度分层控制方法,整合无人机航测、三维激光建模与 Cyclone+Smart 3D 数据处理平台,提出了基于多源数据融合的弃渣场体积高效定量技术.采用 DJI M300 RTK 无人机搭载 5E 镜头相机系统获取全景高分辨率影像,结合 Scan P50 型徕卡高精度地面三维激光扫描仪获取密集点云数据,创新"航测三维模型-地面高精度点云数据"耦合算法,通过坐标系统一、特征点匹配、点云配准等技术实现多源数据精准融合.以霍山县和金寨县 26 个弃渣场为研究对象,系统分析点云精度对土方量计算精度的影响规律.平缓区弃渣场采用 100 mm 点云精度时计算误差为 4.02%,满足工程要求;起伏区弃渣场需采用 50 mm 点云精度才能将误差控制在 3.42%.通过合理选择精度等级,可在保证计算准确性的前提下大幅提升数据处理效率,平缓区和起伏区分别可减少约 99%和 98%的数据量.研究构建了"点云补正模型+模型补全点云"的逆向建模机制,实现了空地数据优势互补,为高效准确获取弃渣场方量提供了可靠的技术支撑.
To improve the accuracy of monitoring the volume of spoil disposal sites and address the issues of inconsistent data formats and standards inherent in single monitoring methods,this study,based on the principles of topographic surveying,spatial coordinate matching theory,and precision hierarchical control methods,integrates UAV aerial surveying,3D laser scanning,and the Cyclone+Smart 3D data processing platform to propose an efficient quantitative technique for spoil disposal site volume based on multi-source data fusion.A DJI M300 RTK UAV equipped with a 5E lens camera system was employed to acquire panoramic high-resolution images,while a Leica Scan P50 high-precision terrestrial 3D laser scanner was used to obtain dense point cloud data.An innovative"aerial survey 3D model-ground high-precision point cloud data"coupling algorithm was developed,achieving precise fusion of multi-source data through techniques such as coordinate system unification,feature point matching,and point cloud registration.Taking 26 spoil disposal sites in Huoshan county and Jinzhai county as the research subjects,the study systematically analyzed the influence pattern of point cloud density on the accuracy of earthwork volume calculation.The results indicated that for spoil disposal sites in gentle terrain areas,a point cloud accuracy of 100 mm yields a calculation error of 4.02%,meeting engineering requirements;for sites in undulating terrain areas,a point cloud accuracy of 50 mm was necessary to control the error within 3.42%.By rationally selecting the accuracy grade,data processing efficiency can be significantly improved while ensuring calculation accuracy,with data volume reduced by approximately 99%for gentle areas and 98%for undulating areas.The research constructed a reverse modeling mechanism of"point cloud correction model+model completion point cloud,"realizing the complementary advantages of aerial and ground data,and providing reliable technical support for the efficient and accurate acquisition of spoil disposal site volumes.
王诏楷;夏小林;陈磊;赵黎明;陈康祚
安徽省·水利部淮河水利委员会水利科学研究院,安徽 合肥 230088||安徽省水科学与智慧水利重点实验室,安徽 合肥 230088安徽省·水利部淮河水利委员会水利科学研究院,安徽 合肥 230088||安徽省水科学与智慧水利重点实验室,安徽 合肥 230088安徽省·水利部淮河水利委员会水利科学研究院,安徽 合肥 230088||安徽省水科学与智慧水利重点实验室,安徽 合肥 230088安徽省·水利部淮河水利委员会水利科学研究院,安徽 合肥 230088||安徽省水科学与智慧水利重点实验室,安徽 合肥 230088安徽省·水利部淮河水利委员会水利科学研究院,安徽 合肥 230088||安徽省水科学与智慧水利重点实验室,安徽 合肥 230088
天文与地球科学
弃渣场多源数据融合三维激光扫描无人机航测
spoil disposal sitemulti-source data fusion3D laser scanningUAV photogrammetry
《江淮水利科技》 2026 (2)
19-25,7
安徽省自然科学基金联合基金项目(2308085US04)安徽省自然科学基金青年项目(2408085QE162)安徽省(水利部淮河水利委员会)水利科学研究院科技攻关计划项目(KJGG202502)
评论