基于经验模型的Hyperion数据植被叶绿素含量反演  

Chlorophyll content retrieve of vegetation using Hyperion data based on empirical models

作　　者：丰明博 牛铮[1,2] 

机构地区：[1]中国科学院遥感与数字地球研究所,北京100101 [2]中国科学院遥感科学国家重点实验室,北京100101 [3]中国科学院大学,北京100049

出　　处：《国土资源遥感》2014年第1期71-77,共7页Remote Sensing for Land & Resources

基　　金：国家重点基础研究发展规划项目(编号:2010CB950603;2013CB733405);公益性行业(气象)科研专项经费项目(编号:GY-HY201006042);国家自然科学基金项目(编号:41201345)共同资助

摘　　要：对于反演植被叶绿素含量而言,基于Hyperion等高光谱传感器数据、利用经验方法建模是一种快速准确的方法。利用多种植被的实测数据以及Hyperion模拟数据,分析植被反射率及其变化形式与叶绿素含量的相关性,并进一步针对红边参数、植被指数等分析植被反射率与叶绿素含量的关系,选取最准确的经验建模方法。经过对比,改进的简单比值指数(modified simple ratio,MSR)与叶绿素含量相关性最高,其回归模型能比较准确地反演出叶绿素含量。通过Hyperion图像、利用MSR指数与实测叶绿素含量得到回归模型,建立区域叶绿素含量分布图;并对张掖地区植被叶绿素含量进行了反演,反演结果具有较高精度,相对误差低于5%。Modeling using empirical methods based on Hyperion is a fast and accurate way to retrieve vegetation chlorophyll content. In this paper, the measured spectra and simulated Hyperion spectra were analyzed, the correlation between chlorophyll content and reflectance with its change forms and the relation between chlorophyll content and red edge parameters as well as vegetation indexes were calculated to obtain the most accurate modeling method. The vegetation index of modified simple ratio（MSR） has a significant correlation with chlorophyll content, and its regression model can retrieve chlorophyll concentration accurately. Using MSR and measured chlorophyll content, the authors built the regression model based on Hyperion data and then established the chlorophyll concentration profile. The chlorophyll concentration profile of Zhangye City was computed and a high - accuracy was achieved,with its relative error less than 5%.

关 键 词：叶绿素含量 经验模型 特征波段 植被指数 Hyperion图像 多种植被 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]