基于可见光-近红外新光谱特征和最优组合原理的大麦叶片氮含量监测

Associating new spectral features from visible and near infrared regions with optimal combination principle to monitor leaf nitrogen concentration in barley

提出新的作物冠层叶片氮含量(LNC)高光谱遥感监测方法,以对氮素要求较高的大麦LNC监测为例,利用田间实测数据,从可见光-近红外区域的高光谱反射曲线中提取包含丰富多波段信息的斜率、夹角等新型特征参数,应用组合预测领域中的权重最优组合原理及其算法,实现对作物LNC的高光谱监测.研究表明,提出的高光谱反射曲线斜率和夹角等新型特征参数与作物LNC显著相关,并具有较好的定量响应关系,其中关键斜率参数(K re/K pb)和K pb以及夹角参数(Aδ/Aα)和(Aδ/Aθ)较好地描述了LNC的动态变化;而权重最优组合分析则表明(K re/K pb)和K nir1两个参数的组合最能响应LNC的光谱信息,有助于增强监测的稳定性并提高估测的精度.

The paper proposed a method to monitor LNC in crop with hyperspectral remote sensing. Taking the LNC monitoring of barley that is more demanding for nitrogen fertilization as a case, this study employs new spectral features such as slopes and angles extracted from the normalized reflectance curves in Visible-Near Infrared region to evaluate LNC, At the same time, the optimal combination principle that was widely used in the combinated forecasting domains was presented to estimate LNC. The analysis resluts proved that most of the new spectral features propsoed in the study exhibited significant correlations with LNC. Among the new spectral features, the key features of slopes (Kre/Kpb and Kpb) and angles (Aδ/Aα and Aδ/Aθ) could well describe the dynamic pattern of LNC changes in crop. The optimal combination algorithm determined the optimal combination with Kre/Kpb and Knir1, which could increase the spectral responding to LNC, strengthen the stability of models monitoring LNC and improve the accuracy of LNC estimates.