基于区域生长法的无人机和无人船一体化测绘技术研究  

Research on Integrated Surveying Technology of Unmanned Aerial Vehicles and Unmanned Boats Based on the Region-Growing Method

作　　者：侍恒 曾贺[1,2] 苏磊 张宝森 刘昌辉[3] SHI Heng;ZENG He;SU Lei;ZHANG Baosen;LIU Changhui(Yellow River Institute of Hydraulic Research,Zhengzhou 450003,China;Research Center on Levee Safety and Disease Prevention,MWR,Zhengzhou 450003,China;Yellow River Engineering Consulting Co.,Ltd.,Zhengzhou 450003,China)

机构地区：[1]黄河水利科学研究院,河南郑州450003 [2]水利部堤防安全与病害防治工程技术研究中心,河南郑州450003 [3]黄河勘测规划设计研究院有限公司,河南郑州450003

出　　处：《人民黄河》2024年第11期137-141,148,共6页Yellow River

基　　金：国家重点研发计划项目(2022YFC3004403);黄河水利科学研究院推转基金资助项目(HKY-YF-2022-05);应急部揭榜中标攻关项目(应急厅函〔2021〕136号);水利部堤防安全与病害防治工程技术研究中心开放课题基金资助项目(LSDP202107)。

摘　　要：针对控导工程监测过程中采用的传统水下高程测量方法存在操作人员多、安全隐患大等问题,以神堤控导工程12坝—20坝为例,采用无人机、无人船等非载人式勘测技术,对河道水上地形和水下地形展开同步测量。引入坡度因子对区域生长法进行改进,以目视解译方法识别的林木区域为基准,分析对比区域生长法和考虑坡度的区域生长法的识别准确率。采用自然邻域法对水下地形数据进行插值并与水上地形数据融合,得到神堤控导工程水上水下数字高程模型(DEM),并与工程三维视图进行对比。结果表明:利用无人机和无人船一体化测绘生成的DEM成果,满足有关技术规范的要求,具有较高的精度;采用考虑坡度的区域生长法识别控导工程的防浪林和行道林,识别正确率达92.47%。In response to the issues of multiple operators and significant safety hazards in traditional underwater elevation measurement meth-ods during the monitoring and inspection process of river training works,taking the No.12th dam-No.20th dam of Shendi river training works as an example,non-manned survey technologies such as unmanned aerial vehicles and unmanned boats were used to simultaneously measure the topography of the river channel and underwater terrain.The slope factor was introduced to improve the regional growth method,and the forest area identified by visual interpretation method was used as a benchmark to analyze and compare the recognition accuracy of the regional growth method and the regional growth method considering slope.The underwater terrain data was interpolated by using the natural neighbor-hood method and fused with the above-water terrain data to obtain the water surface and underwater Digital Elevation Model(DEM)of Shendi river training works,which was compared with the three-dimensional view of the project.The results show that the use of unmanned aerial ve-hicles and unmanned boats for integrated mapping to generate DEM results can meet the requirements of relevant technical specifications and have higher accuracy.The region-growing method considering slope is used to identify the wave-proof forest and roadside forest of the river training works,with a recognition accuracy of 92.47%.

关 键 词：无人机 无人船 水下地形 水上地形 区域生长法 神堤控导工程 

分 类 号：P231[天文地球—摄影测量与遥感] P234.2[天文地球—测绘科学与技术]