Estimation Model For Chlorophyll And Canopy Index Of Maize Based On Hyperspectral Remote Sensing

Hyperspectral remote sensing is characterized by high spectrall resolution, marvelous continuity of wavebands and considerable contents of spectral information. To monitor some growth indexes instantaneously can reasonably conduct crop management and enhance the output and quality of crops, reflecting the requirement of modern agriculture.Based on analysis of canopy spectral reflectance and physiological and biochemical test technology, different stages(jointing stage, booting stage, blossom stage, filling stage, milk stage) hyperspectral data and growth indexes(LAI, FPAR, Chi, Chi.a, Chl.b, Chl.d) of two varieties spring maize in different densities were obtained with hyperspectral detection technology, the characteristics of canopy spectral reflectance were analysed, correlation analysis were made and prediction equations were established between LAI/FPAR and the original hyperspectral reflectance and different forms with vegetation index. The results are as follows:1.Simulation model of Leaf area index (LAI) can be estimated by building simulation model with the first derivative spectral reflectance at 851nm,the simulation equation is LAI=3.6944e76927x(r=0.9385**), the correlation coefficient between simulating data and testing data is significant at 0.01 level(r=0.8987**,RMSE=0.4400).2.Simulation model of Fraction of Absorbed Photosynthetically Active Radiation (FPAR) can be estimated by building simulation model with NDVI at (855,758),thc simulation equation is FPAR=-76.966x2+9.297x+0.6405(r=0.6686**), the correlation coefficient between simulating data and testing data is significant at 0.01 level(r=0.9045**,RMSE=0.0136).3.Simulation model of Chlorophyll content (Chi) can be estimated by building simulation model with SAVI at (997,759),the simulation equation is Chl=91.93x2 17.363x+2.7406 (r=0.7983**), the correlation coefficient between simulating data and testing data is significant at 0.01 level(r=0.7543**,RMSE=0.3042).4.Simulation model of Chlorophyll a content (Chi.a) can be estimated by building simulation model with the original spectral reflectance at 359nm,the simulation equation is Chl.a=1.9665e-4407x(r=0.6115**), the correlation coefficient between simulating data and testing data is significant at 0.01 level(r=0.6032**,RMSE=0.1527).5.Simulation model of Chlorophyll b content (Chl.b) can be estimated by building simulation model with NDVI at (982,760),the simulation equation is Chl.b= 67.23x" 11.198x+0.778(r=0.8944**). the correlation coefficient between simulating data and testing data is significant at 0.01 level(r=0.7740**,RMSE=0.2382).6.Simulation model of Chlorophyll dense (Chl.d) can be estimated by building simulation model with SAVI at (974,760),the simulation equation is Chl.d= 11.055x2-5.2275x+0.769l(r=0.7127**), the correlation coefficient between simulating data and testing data is significant at 0.01 level(r=0.6614**,RMSE=0.1570).