基于Landsat-7 ETM的地物精准分类提取方法分析  

Analysis on the Methods of Accurate Classification and Extraction of Features Based on Landsat-7 ETM Data

作　　者：雷宝佳 蒋好忱 刘兰兰 方睿红 LEI Baojia;JIANG Haochen;LIU Lanlan;FANG Ruihong(The First Institute of Mapping and Geographic Information of MNR,Xi′an 710054,China;The First Topographic Surveying Brigade of MNR,Xi′an 710054,China)

机构地区：[1]自然资源部第一地理信息制图院,陕西西安710054 [2]自然资源部第一地形测量队,陕西西安710054

出　　处：《测绘与空间地理信息》2023年第7期26-28,33,共4页Geomatics & Spatial Information Technology

基　　金：陕西测绘地理信息局科技创新专项基金项目(SCK2021-02)资助。

摘　　要：居民地是人与自然相互作用的产物,快速掌握居民地及其他地物的空间分布和变化特征对城市化进程及区域可持续发展具有重要意义。以四川省绵阳市为研究区域,采用Landsat-7 ETM数据,利用基于像元和面向对象两种方法分别对研究区的居民地及其他地物进行提取,并对结果进行了比较分析。结果表明,三种方法都能较为准确地提取出居民地及其他地物信息,但在提取精度、工作量投入、应用场景方面各有优缺点,面向对象的分类方法具有一定的优势,在大范围生产中可根据实际情况选用不同的方法。Residential areas are the product of the interaction between humans and nature,and quickly grasping the spatial distribution and change characteristics of residential areas and other land features is of great significance for the urbanization process and regional sustainable development.Taking Mianyang in Sichuan province as the research area,using Landsat-7 ETM data,this paper uses pixel based and object-oriented methods to extract residential areas and other surface features in the research area,and compares and analyzes the results.The results show that all three methods can accurately extract residential and other land feature information,but each has its own advantages and disadvantages in terms of extraction accuracy,workload investment,and application scenarios.The object-oriented classification method has certain advantages,and different methods can be selected according to the actual situation in large-scale production.

关 键 词：居民地 遥感 对象 像元 

分 类 号：P237[天文地球—摄影测量与遥感]