小型水库大坝无人机巡检和行人定位方法  

Unmanned aerial vehicle inspection and pedestrian location method for small reservoir dams

作　　者：赵思曾 康飞 何丽娜[3,4] 李俊杰 ZHAO Sizeng;KANG Fei;HE Lina;LI Junjie(College of water Conservancy and Hydropower Engineering,Hohai University,Nanjing 210098,China;Faculty of Infrastructure Engineering,Dalian University of Technology,Dalian 116024,China;School of Earth Sciences and Engineering,Hohai University,Nanjing 211100,China;State Key Laboratory of Geographic Information Engineering,Key Laboratory of Surveying and Mapping Science and Geospatial Information Technology,Ministry of Natural Resources,Beijing 100036,China)

机构地区：[1]河海大学水利水电学院,江苏南京210098 [2]大连理工大学建设工程学部,辽宁大连116024 [3]河海大学地球科学与工程学院,江苏南京211100 [4]地理信息工程国家重点实验室、测绘科学与地球空间信息技术自然资源部重点实验室,北京100036

出　　处：《江苏水利》2024年第6期12-18,32,共8页Jiangsu Water Resources

基　　金：江苏省水利科技项目(2022062);国家重点研发计划(2022YFB4703401,2022YFB4703404);国家自然科学基金(51979027,52079022);地理信息工程国家重点实验室、测绘科学与地球空间信息技术自然资源部重点实验室联合资助基金项目(2023-01-01)。

摘　　要：针对小型水库大坝缺少准确的坐标转换参数,无人机识别的行人目标难以精确三维定位的问题,提出无人机坐标系与坝体局部坐标系的转换方法。实际工程应用结果表明,坐标转换后生成的点云与激光扫描点云基本一致,对坝顶目标的非接触测量精度优于10cm。提出方法可以从任意角度和位置准确测量小目标行人在坝体结构上的位置,为小型水库大坝无人机常规巡检和建立电子围栏等提供高效且经济的方法。In view of the lack of accurate coordinate transformation parameters for small reservoir DAMS and the difficulty of accurate three-dimensional positioning of pedestrian targets identified by UAVs,a method of transforming UAVs coordinate system to local coordinate system of dam body is proposed.Practical engineering application results demonstrate that the point cloud generated after coordinate transformation is almost identical to the laser scanning point cloud,and the non-contact measurement accuracy for dam-top targets exceeds 10cm.The proposed method can measure small pedestrian targets on the dam structure from any angle and position,providing an efficient and cost-effective approach for routine drone inspections and establishing electronic fences for small reservoir dams.

关 键 词：无人机 小型水库大坝 摄影测量 运动恢复结构(SfM) 三维坐标转换 小目标检测 

分 类 号：TV123[水利工程—水文学及水资源]