摘要
以Worldview-2遥感影像作为数据源,对东洞庭湖湿地核心区域进行了LAI估算研究。首先对典型的植被指数与叶面积指数(VI-LAI)的相关性进行了分析;然后选择了7个与LAI之间存在显著相关性的植被指数(NDVI、RVI、DVI、SAVI、MSAVI、EVI、RDVI)作为VI-LAI的模型因子,采用多回归模型与LAI实测数据进行拟合分析,筛选出这回归模型的最优因子;然后利用实测数据作为检验样本,最终确立了以NDVI为模型变量的指数模型是用于LAI估测的最优模型,精度达到了74.34%。结果表明:本研究采用的多植被指数比较建立的湿地植被VI-LAI反演模型,是可以比较准确获取湿地区域叶面积指数特征的方法。
The research selects core area in wetland nature reserve of Dongting Lake in Hunan province as reserch object. In order to estimate the vegetation leaf area index（LAI） of core wetland region of Dongting Lake, high resolution image WORLDVIEW-2 data is utilzied to precisely leaf area index information. Then, Correlation between vegetation index and VI-LAI is analyzed. Seven vegetation indexes such as NDVI, RVI, DVI, SAVI, MSAVI, EVI and RDVI are selected as model factors of VI-LAI. Then multiple regression models（including unitary linear, quadratic polynomical, cubic polynomical, index model, logarithmic model and power function model） are adopted to take fitting analysis with measrued LAI data to select optimal factors of the six regression models. After that, 23 groups of measured data are used as test samples to finally determine that the the index model with NDVI as model variable is the optimal model for LAI estimation, Which makes the pricesion up to 74.34%.The experimental result shows that Vegetation Index Comparation mothod is an effective sway in wetland remote sensing leaf area index（LAI）Estimation.
出处
《中南林业科技大学学报》
CAS
CSCD
北大核心
2016年第5期11-18,共8页
Journal of Central South University of Forestry & Technology
基金
国家重大专项(21-Y30B05-9001-13/15-2)
国家高技术研究发展计划(863计划)(2012AA102001)
