基于面向对象分类技术的雅砻江流域地表信息提取与分析  

Landscape information extraction and analysis of the Yalong River Basin using object-oriented classification method

作　　者：谢帅[1,2] 陈建平[1,2] 于欢[3] 王哲[3] 向杰[1,2] 

机构地区：[1]中国地质大学(北京)地球科学与资源学院,北京100083 [2]北京市国土资源信息研究开发重点实验室,北京100083 [3]成都理工大学地球科学学院,四川成都610059

出　　处：《地质学刊》2016年第3期526-534,共9页Journal of Geology

基　　金：中国地质调查局项目"矿山环境变化自动监测示范研究"(1212011120029)

摘　　要：雅砻江是川西最重要的河流之一,雅砻江流域内的土地利用情况与流域内的生态环境及河流水质密切相关。以Landsat 8 OLI遥感影像为数据源,采用面向对象的分类方法,建立分类规则,获取研究区的土地利用情况,利用Fragstats 4.2软件进行区域景观空间分布特征分析。结果表明:采用面向对象的分类方法的分类总体精度达到8711%,Kappa系数达到0.855。雅砻江流域内的区域整体景观连接性较好。草地和林地是研究区的主导类型,整体看分布较为集中。The Yalong River is one of the most important rivers in western Sichuan Province. The land use of the Yalong River Basin is closely related to its ecological environment and water quality. In this study, using Landsat 8 remote sensing images as data source, we established the classification rule using an object-oriented classification method to obtain the land use of the study area, and further analyzed the spatial distribution characteristics of regional landscape with the Fragstats 4.2 software. The resuh shows that the overall classification accuracy was 87.11% by using an object-oriented classification method, and the Kappa coefficient was 0.855. The regional overall landscape in the Yalong River Basin has a good connectivity. Grass and forest lands were the dominant landscape types in the study area, with an overall concentrated distribution.

关 键 词：面向对象分类 土地利用 景观格局指数 Fragstats软件 eCognition软件 雅砻江流域 四川西部 

分 类 号：P237[天文地球—摄影测量与遥感]