基于ESP2的面向对象分类方法在高铁线路提取中的应用  

Application of Object-Oriented Classification Based on ESP2 in High-Speed Railway Line Extraction

作　　者：周秀芳 龚循强[2,3] 李泽春 邱万锦 孙坤 ZHOU Xiufang;GONG Xunqiang;LI Zechun;QIU Wanjin;SUN Kun(Research Center for Ecological Civilization Construction System of Jiangxi Province,East China University of Technology,Nanchang 330013,China;Key Laboratory of Mine Environmental Monitoring and Improving Around Poyang Lake of Ministry of Natural Resources,East China University of Technology,Nanchang 330013,China;Faculty of Geomatics,East China University of Technology,Nanchang 330013,China)

机构地区：[1]东华理工大学江西生态文明建设制度研究中心,江西南昌330013 [2]东华理工大学自然资源部环鄱阳湖区域矿山环境监测与治理重点实验室,江西南昌330013 [3]东华理工大学测绘工程学院,江西南昌330013

出　　处：《测绘地理信息》2024年第4期20-23,共4页Journal of Geomatics

基　　金：江西生态文明建设制度研究中心开放基金(JXST2104);国家自然科学基金(42101457)

摘　　要：以高分二号影像为原始数据,采用eCognition软件中的ESP2工具预测影像最佳分割尺度参数,通过k近邻、分类与回归树、支持向量机3种面向对象分类方法提取高铁线路,并引入总体精度、Kappa系数、完整率、正确率和提取质量5个指标对提取的高铁线路进行精度评价。结果表明,3种方法的5个评定指标均在0.9以上,这表明面向对象分类方法在高铁线路提取中具有可行性。The ESP2 tool in eCognition software is used in combination with the GF-2 images as the original data to predict the optimal segmentation scale parameters of the image,and the high-speed railway lines are extracted through three object-oriented classification methods,i.e.,k-nearest neighbor,classification and regression tree,and support vector machine.Five indicators of overall accuracy,Kappa coefficient,completion rate,correct rate and extraction quality are introduced to evaluate the accuracy of the extracted high-speed railway lines.The experimental results show that the five mentioned extraction indexes of the three methods are all above 0.9,which indicates that the object-oriented classification method is feasible in the field of high-speed railway line extraction.

关 键 词：面向对象分类 高铁线路提取 ESP2 影像分割 精度评定 

分 类 号：P208[天文地球—地图制图学与地理信息工程] P237[天文地球—测绘科学与技术] U238[交通运输工程—道路与铁道工程]