基于光谱归一化的马尾松LAI遥感估算研究

Masson's Pine LAI Estimation Based on Spectral Normalization Using Remote Sensing Data

通过对比不同传感器间光谱响应函数的差异,研究基于光谱响应函数的不同传感器相似波段的归一化方法,探讨归一化后植被指数在马尾松叶面积指数(LAI)估算中的应用。以某一传感器为基准,根据波段总辐射率比值关系将其他卫星传感器归一化为基准传感器,然后计算其植被指数,建立LAI反演模型。为验证方法可行性,选取永安地区2008年3月获取的BJ-1CCD、IRS-P6LISS3和MODIS数据作为研究对象,根据三者的光谱响应函数差异,将BJ-1CCD和IRS-P6的LISS3的红光和近红外波段归一化为MODIS的相应波段,并分别计算归一化前后的NDVI值。结果表明归一化后不同传感器的植被指数关系与理想的关系y=x更加接近。利用归一化后的IRS-P6影像的NDVI反演马尾松LAI,并将其应用于MODIS和BJ-1传感器,得到归一化后不同传感器的植被指数值基本相等,表明归一化以后的植被指数应用于LAI的估算具有一定的普适性,能适用于多种传感器。

In this paper,the spectral response functions among the different sensors were compared,and the normalization method for similar bands of different sensors was studied.The application of normalized VI on Masson＇s pine LAI estimation was studied.One specific sensor was set as a reference and other sensors were normalized to it according to the integral ratio relationship,then the vegetation index could be calculated and LAI inversion model was established.To validate its feasibility,Yongan city in Fujian province was taken as the study area,and the data of red and near-infrared from BJ-1 CCD sensor and IRS-P6 LISS3 sensor were normalized consistently to corresponding bands of MODIS sensor.And then the NDVI values before and after normalization were calculated.The results showed that after normalization,the relationship of vegetation indices among the different sensors were closer to the ideal relationship y=x.Moreover,NDVI calculated from normalized IRS P6 data was used to estimate Masson＇s pine LAI,which was applied to MODIS and BJ-1 sensors.As the LAI among different sensors were nearly equal after normalization,the conclusion could be reached that the normalized VIs of different sensors could be generally applied to estimate LAI to a certain extent.