Landsat和GF数据面向对象土地覆盖分类研究  

Exploring the object-oriented land cover classification based on Landsat and GF data

作　　者：尚明 马杰[2,3] 李悦[1] 赵菲 顾鹏程 潘光耀 李倩 任阳阳 SHANG Ming;MA Jie;LI Yue;ZHAO Fei;GU Pengcheng;PAN Guangyao;LI Qian;REN Yangyang(School of Earth Science and Engineering,Hebei University of Engineering,Handan 056038,China;Xinjiang Key Laboratory of Water Cycle and Utilization in Arid Zone,Xinjiang Institute of Ecology and Geography,Chinese Academy of Sciences,Urumqi 830011,China;School of Architecture and Art,Hebei University of Engineering,Handan 056038,China;China Satellite Communications Co.,Ltd.,Beijing 100190,China;China Land Surveying and Planning Institute,Beijing 100871,China;Jinglang Ecological Environment Technology(Wuhan)Co.,Ltd.,Wuhan 420111,China;State Key Laboratory of Desert and Oasis Ecology,Xinjiang Institute of Ecology and Geography,Chinese Academy of Sciences,Urumqi 830011,China;Hebei Geographic Information Group Co.,Ltd.,Shijiazhuang 050000,China)

机构地区：[1]河北工程大学地球科学与工程学院,邯郸056038 [2]新疆干旱区水循环与水利用重点实验室,中国科学院新疆生态与地理研究所,乌鲁木齐830011 [3]河北工程大学建筑与艺术学院,邯郸056038 [4]中国卫通集团股份有限公司,北京100190 [5]中国国土勘测规划院,北京100871 [6]景朗生态环境技术(武汉)有限公司,武汉420111 [7]荒漠与绿洲生态国家重点实验室,中国科学院新疆生态与地理研究所,乌鲁木齐830011 [8]河北省地理信息集团有限公司,石家庄050000

出　　处：《自然资源遥感》2024年第3期240-247,共8页Remote Sensing for Natural Resources

基　　金：邯郸市科学技术研究与发展计划项目“面向对象的邯郸市土地覆被变化及生态服务功能评价”(编号:21422903273);新疆干旱区水循环与水利用重点实验室开放课题“干旱区融雪径流模拟研究——以玛纳斯河流域为例”(编号:XJYS0907-2023-11);国家自然科学基金项目“华北平原农田最大羧化速率和光合色素的关系及其生产力模拟研究”(编号:32001130);河北省自然科学基金项目“基于叶片光合能力和叶绿素含量同步观测的河北冬小麦-夏玉米生产力模拟研究”(编号:C2021402011)共同资助。

摘　　要：针对中分辨率遥感数据面向对象分类,以河北省北部山区和南部平原Landsat8 OLI,Landsat5 TM及高分一号(GF1)数据为研究对象,对支持向量机(support vector machine,SVM)、随机森林(random forest,RF)、决策树(decision tree,DT)及朴素贝叶斯(naive Bayes,NB)4种分类器的土地覆盖分类效果进行对比,并分析SVM,RF和DT中关键参数对分类结果的影响。结果表明:在2个研究区,各分类器结果略有差异,从整体上看其优劣排序为SVM,NB,RF和DT。其中SVM和DT分类精度随参数变化波动较大:对于SVM,当参数C取值不小于103且gamma不大于10^(-1)时,无论哪种情况其分类精度均优于90%;对于DT,当参数树深(Depth)大于3时,各情况下的分类精度相对较高且趋于稳定。RF分类精度随参数变化波动较小且没有明显的变化规律。研究结果可为中分辨率遥感数据面向对象土地覆盖分类研究提供参考。This study aims to explore the object-oriented classification based on moderate-resolution remote sensing data.Using the Landsat8 OLI,Landsat5 TM,and GF1 data obtained from the northern mountainous area and the southern plain area in Hebei Province,this study compared the land cover classification effects of four classifiers:support vector machine(SVM),random forest(RF),decision tree(DT),and naive Bayes(NB).Moreover,it analyzed the impacts of critical parameters in SVM,RF,and DT on the classification results.The findings indicate that the classification results of the classifiers vary slightly in the two study areas,with their effects decreased in the order of SVM,NB,RF,and DT.The classification accuracies of SVM and DT fluctuated significantly with parameter changes.With C values not below 103 and gamma values not exceeding 10^(-1),SVM can yield classification accuracies above 90%in all cases.With depth values over 3,DT exhibits relatively high and stable classification accuracies.With parameter changes,RF manifests slightly varying classification accuracies with nonsignificant variation patterns.The results of this study serve as a reference for exploring the object-oriented land cover classification based on moderate-resolution remote sensing data.

关 键 词：面向对象分类 分类器 LANDSAT 高分一号 土地覆盖 

分 类 号：P407.8[天文地球—大气科学及气象学] TP79[自动化与计算机技术—检测技术与自动化装置]