随机森林模型在遥感水深反演中的应用  

Satellite-Derived Bathymetry Using Random Forest Model

作　　者：邱耀炜 沈蔚[1,2] 纪茜 QIU Yao-wei;SHEN Wei;JI Qian(College of Marine Science,Shanghai Ocean University,Shanghai 201306,China;Shanghai Engineering Research Center of Estuarine and Oceanographic Mapping,Shanghai 201306,China)

机构地区：[1]上海海洋大学海洋科学学院,上海201306 [2]上海河口海洋测绘工程技术研究中心,上海201306

出　　处：《海洋技术学报》2019年第5期98-103,共6页Journal of Ocean Technology

基　　金：上海市科委重点科研资助项目(14590502200)

摘　　要：随着我国浅海测绘需求的日益增长,文中利用四波段的WorldView-2高分辨率遥感影像,选取我国南海西沙群岛中的甘泉岛和台湾南湾地区作为典型试验区,开展水深反演研究。引入随机森林算法构建了随机森林水深反演模型,并同常用的3种水深反演模型进行精度对比。结果表明,在甘泉岛和南湾地区随机森林模型反演的水深值和真实水深值的RMSE分别为0.85 m和1.59 m,MRE分别为8%和12%,均优于其他3种模型。With the growing demand for shallow sea surveying and mapping in China,this paper uses four-bands high resolution WorldView-2 images to conduct satellite-derived bathymetry experiments in the Ganquan Island in Xisha Islands and the Nanwan area in Taiwan.The random forest satellite-derived bathymetry model is constructed by using the random forest algorithm,and the accuracy is compared with three classic satellite-derived bathymetry model.The results show that the bathymetry accuracy of the random forest model is optimal in the Ganquan Island and Nanwan area,with RMSE of 0.85 m and 1.59 m,and MRE of 8%and 12%,respectively.

关 键 词：水深遥感 Worldview-2 随机森林 甘泉岛 台湾南湾 

分 类 号：P237[天文地球—摄影测量与遥感]