运用遥感指数的崩滑检测方法  

Research on Landslides Detection Method Based on Remote Sensing Index

作　　者：冯淦[1] 何政伟[1] 邓磊[2] 赵银兵[1] 胡步清[1] 徐鹏杰[2] 

机构地区：[1]成都理工大学 地质灾害防治与地质环境保护国家重点实验室,四川成都610059 [2]首都师范大学 三维信息获取与应用教育部重点实验室,北京100048

出　　处：《地理空间信息》2012年第2期16-18,179,共3页Geospatial Information

基　　金：科技部支撑资助项目(2008BAK49B02)

摘　　要：以汶川地震时出现的崩滑为研究对象,以该地区震前/后TM影像为研究数据,利用变化检测的方法,比较和分析了归一化植被指数(NDVI)、归一化建筑指数(NDBI)和归一化燃烧率(NBR)应用在崩滑检测中的可行性和优劣性。结果表明,3种遥感指数均可以用于崩滑检测,但效果不同:NDVI对崩滑检测最全面但可信度最低,NBR对崩滑检测最可信但遗漏最多,NDBI处于中间水平。Based on the landslides occurs on the Wenchuan earthquake and pre/post-earthquake TM image data,this paper used the change detection method to compare and analyze the feasibility,advantages and disadvantages between the Normalized Difference Vegetation Index(NDVI),the Normalized Difference Built-up Index(NDBI) and the Normalized Burn Ratio(NBR) in the field of landslide detection.Results showed that all the three kinds of remote sensing index could be used in the field of landslide detection,but the effect was different.NDVI had the highest comprehensiveness and lowest credibility,NBR had the highest credibility and the lowest comprehensiveness,and NDBI in intermediate level.

关 键 词：遥感 变化检测 NDVI NDBI NBR 崩滑 

分 类 号：P237[天文地球—摄影测量与遥感]