http://www.gissky.net- GIS空间站

基于高光谱遥感数据Hyperion和植物冠层反射光谱,应用指数法、回归统计法和基于光谱位置变量的方法对矿区植被生物量和叶绿素浓度(SPAD)进行估算。结果表明:植被指数R752/R548与植物鲜重相关性最高,相关系数为0.88;选用植物像元光谱,基于植被指数R752/R548,利用三次函数法构建植物鲜重估算模型精度较高,多重判定系数R2达0.883;植被指数DVI[752,640]与植物干重相关性最高,相关系数为0.42;基于植被指数DVI[752,640],应用线性回归法构建植被干重估算模型精度较低,多重判定系数R2为0.177;基于四点内插法提取的红边位置与叶绿素浓度显著相关,相关系数为0.433;Datt(1)和Datt(2)植被指数与叶绿素浓度存在显著相关,相关系数分别为0.871和0.868;基于红边位置(REP)、Datt(1)和Datt(2)植被指数构建植物叶绿素浓度估算模型精度较高,多重判定系数R2分别为0.814、0.805和0.781。应用高光谱遥感技术可有效地检测矿区受损生态环境下的植被,为矿区植物生态修复工程提供本底资料。 更多还原

【Abstract】 In order to testify the availability of estimation of bio-chemical and bio-physical parameters of vegetation by using hyper-spectral remote sensing,biomass and chlorophyll concentration were estimated effectively by using index method,linear regression,and red edge method.Research results showed that there was the highest correlation between vegetation index R752/R548 and pixel spectra extracted from Hyperion data,the R was 0.88;fresh weight estimation model by cubic function method was establis...