基于NDVI和LSMM的干旱区植被信息提取研究——以新疆吐鲁番市为例  

Extraction of Vegetation Information in Arid Area Based on NDVI and LSMM:A Case Study of Turpan

作　　者：徐娜[1,2] 丁建丽[1,2] 刘海霞[1,2] 

机构地区：[1]新疆大学资源与环境科学学院,新疆乌鲁木齐830046 [2]新疆绿洲生态重点实验室,新疆乌鲁木齐830046

出　　处：《测绘与空间地理信息》2012年第7期52-57,共6页Geomatics & Spatial Information Technology

基　　金：国家自然基金项目(40861020;40961008);霍英东教育基金项目(121018);教育部科学研究重点项目(030012);新疆自然科学基金项目(200821128)资助

摘　　要：植被是干旱区生态建设重要的组成部分,而植被覆盖度是生态环境变化的重要指示,是评价生态系统健康的前提条件。本文在遥感等技术的支持下,以landsatTM影像为数据源,选用归一化植被指数(NDVI)和线性光谱混合分析模型(LSMM)两种方法进行分析比较,提取吐鲁番市近20年植被覆盖度,并对该地区植被覆盖度的演变特征进行分析。结果表明:①LSMM方法能较好地提取干旱区植被信息,指标简单且分类精度较高。②NDVI方法提取植被时,受到很多限制,在干旱区不宜采用。Vegetation is an important part of ecological construction in arid areas,while vegetation cover is an important indication of environmental change,and is the premise condition of ecological health evaluation.With the support of remote sensing technology,based on landsatTM images,which is extracted using normalized difference vegetation index(NDVI) and linear spectral mixture model(LSMM),this paper Abstracts the vegetation coverage of Turpan city of the last 20 years,and analyzes the vegetation coverage evolution characteristics in Turpan.The results showed that: 1.The LSMM method can well extract vegetation cover information in arid area,the index is simple and classification accuracy is higher.2.The NDVI method is restricted when extracting the information of the vegetation,and should not be used in the arid area.

关 键 词：植被覆盖度 NDVI 光谱混合分析 遥感 干旱区 

分 类 号：TP75[自动化与计算机技术—检测技术与自动化装置]