基于多源高分辨率遥感影像的典型自然资源要素提取  

Extraction of typical natural resource elements based on multi-sourcehigh-resolution remote sensing images

作　　者：马锦山 贾国焕 张赛 张炯 MA Jinshan;JIA Guohuan;ZHANG Sai;ZHANG Jiong(Xining Land Survey and Planning Research Institute Co.,Ltd.,Xining 810000,China;Zhongke Beiwei(Beijing)Technology Co.,Ltd.,Beijing 100192,China)

机构地区：[1]西宁市国土勘测规划研究院有限公司,青海西宁810000 [2]中科北纬(北京)科技有限公司,北京100192

出　　处：《测绘通报》2024年第3期123-126,150,共5页Bulletin of Surveying and Mapping

摘　　要：利用高分辨率遥感数据具有高空间分辨率的特性,本文以青海省西宁市0.3和1 m多源高分辨遥感影像为数据源,基于卷积神经网络深度学习算法进行典型自然资源要素提取。结果表明,0.3 m遥感影像提取耕地、林地准确率均在85%以上,召回率在89%以上;1 m遥感影像提取耕地林地准确率在90%以上,召回率在91%以上,研究成果可用于西宁市自然资源典型要素智能提取。Using high-resolution remote sensing data with high spatial resolution characteristics,typical natural resource elements are extracted based on traditional convolutional neural network deep learning algorithms using multi-source high-resolution remote sensing images of 0.3 and 1 m in Xining,Qinghai province as data sources.The results show that the accuracy of extracting farmland and forest land from 0.3 m remote sensing images is over 85%,with a recall rate of over 89%.The accuracy of extracting farmland and forest land from 1 m remote sensing images is over 90%,with a recall rate of over 91%.The research results can be used for intelligent extraction of typical elements of natural resources in Xining.

关 键 词：高分辨率 卷积神经网络 深度学习 遥感解译 

分 类 号：P237[天文地球—摄影测量与遥感]