基于LANDSAT 8卫星影像的马鞍山市土地利用分类法比较的研究  

Comparative study of land use classification methods in Maanshan City based on LANDSAT 8 satellite images

作　　者：熊敏 XIONG Min(Fujian Provincial Institute of Geological Surveying and Mapping,Fuzhou 350011,China)

机构地区：[1]福建省地质测绘院,福建福州350011

出　　处：《经纬天地》2023年第5期76-81,共6页Survey World

摘　　要：目前,遥感技术越发成熟,使用遥感技术采集的遥感影像质量也越来越高,因此土地利用影像信息绝大多数来源于卫星遥感影像。本文研究所采用的数据是马鞍山市基于LANDSAT 8卫星获取的多光谱遥感影像,将下载的卫星影像做预处理,并对处理过的数据影像使用K-Means、IsoData、最大似然法、最小距离法和决策树5种不同分类方法进行土地利用分类,根据该地区土地利用现状以及国家土地利用现状分类标准,将当地的土地类型分为水体、植被、农田、居民区和道路5大类,最后对每个分类结果进行精度验证,并加以对比分析,得出对于该地区最优的分类方法为决策树分类法。有利于更加充分地管理利用土地,避免国土资源的浪费和破坏。At present,remote sensing technology is becoming more and more mature and the quality and clarity of remote sensing images collected through remote sensing technology are becoming higher and higher.Therefore,the vast majority of land use image information comes from satellite remote sensing images.Considering the factors of cloud cover,the data used in this study is a multispectral remote sensing images obtained from a certain area based on LANDSAT 8 satellite.The downloaded satellite image is preprocessed and the processed data image is classified by using five different classification methods for land use.At the same time,according to the current land use situation in the region and the national land use status classification standards,the local land types are divided into water,bodies,vegetation,farmland and roads.In the end,the accuracy of each classification result is verified and compared for the five major categories of residential areas and roads.The optimal classification method for this area is the decision tree classification method.

关 键 词：监督分类 非监督分类 决策树 精度验证 土地利用 

分 类 号：F301.2[经济管理—产业经济]