苹果叶片氮含量高光谱反演方法对比

杨福芹,冯海宽,李振海,潘洁晨,谢瑞

Comparison of Hyperspectral Remote Sensing Inversion Methods for Apple Leaf Nitrogen Content

Fuqin Yang,Haikuan Feng,Zhenhai Li,Jiechen Pan,Rui Xie

表7 基于三边光谱指数的叶片氮含量预测模型及验证

Tab 7 Modeling and verification of leaf nitrogen content based on spectral parameter

光谱指数 回归方程 建模（n=180） 验证（n=297） R2 RMSE RE% R2 RMSE RE% λr y = -0.0962x + 72.015 0.60 0.38 10.44 0.46 0.35 11.09 Db y = 1.704ln(x) + 13.662 0.75 0.30 8.26 0.40 0.35 10.89 Dy y = 2766.6x + 2.1353 0.68 0.34 9.19 0.32 0.38 11.80 Rg y = 23.509x + 0.6495 0.58 0.39 10.81 0.00 0.45 13.88 SDb y = 1.6986ln(x) + 8.6703 0.72 0.31 8.34 0.13 0.44 14.36 SDy y = -58.941x + 1.3219 0.78 0.28 7.46 0.41 0.36 11.70 SDr/SDb y = -1.741ln(x) + 7.5622 0.71 0.32 8.55 0.20 0.43 14.12 SDr/SDy y = 0.1131x + 5.1234 0.75 0.30 8.42 0.40 0.39 13.06 Rg/Ro y = 2.7425ln(x) + 0.9405 0.75 0.30 7.91 0.33 0.41 13.71 (SDr-SDb)/(SDr+SDb) y = -11.617x + 12.969 0.68 0.34 8.73 0.21 0.42 13.79 (SDr-SDy)/(SDr+SDy) y = 14.562ln(x) + 1.3012 0.75 0.30 8.00 0.47 0.35 11.56 (Rg-Ro)/(Rg+Ro) y = 6.521x + 0.6827 0.75 0.30 7.95 0.32 0.41 13.92