极端干旱区稀疏荒漠植被信息遥感探测研究  

Sparse desert vegetation extraction in extreme arid region based on remote sensing imagery

作　　者：刘桂林[1,2] 张落成[1] 李广宇[1,2] 刘剑[1,2] 杨洪振[3] 

机构地区：[1]中国科学院南京地理与湖泊研究所,南京210008 [2]中国科学院研究生院,北京100049 [3]宁阳县农业局,宁阳271400

出　　处：《干旱区资源与环境》2013年第4期37-40,共4页Journal of Arid Land Resources and Environment

基　　金：国家自然科学基金项目(40971072);国家海洋公益专项(201005009-13)资助

摘　　要：基于Landsat 5 TM遥感影像,以塔里木河下游断流区的荒漠河岸林为例,选取植被指数随植被覆盖度增加的反映速率、植被指数的取值范围变化、植被与背景信息的区分度三个指标,对NDVI、SAVI、MSAVI、GEMI、EVI 5种植被指数探测低覆盖度植被信息的能力进行对比分析。研究表明:根据三个指标的综合分析,GEMI指数是塔里木河下游断流区探测低覆盖度植被信息的最佳植被指数。Taking desert riparian forest in the lower reaches of Tarim River as an example,five vegetation indices such as NDVI,SAVI,MSAVI,GEMI,and EVI were used to compare and analyze the ability of detecting low vegetation cover based on Landsat 5 TM imagery using three indicators such as change rate and the range of vegetation index,division degree.The results showed that GEMI is the best vegetation index to detect low coverage of vegetation information in the lower reaches of Tarim River according to the comprehensive analysis of the three indicators.

关 键 词：极端干旱区 稀疏荒漠植被 遥感 探测 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]