基于ASD和FISS的植被叶片氮素含量反演研究

Inversion of nitrogen content of plant leaves based on ASD and FISS

地面成像光谱系统能够在更为微观的尺度上满足多元化应用需求，实现对植物群体、单个植株的快速精细探测，获取精细的植被生理参量的空间信息，从而弥补大尺度远距离遥感难以发挥作用的领域。本文采用ASD地物光谱仪和新型地面成像光谱系统FISS，以大豆叶片为研究对象，获取大豆叶片ASD光谱及FISS成像光谱数据，探讨大豆植株叶片的图像特征、反射率光谱、导数光谱、光谱指数等光谱特征变量与叶片氮素含量的相关性，建立植株叶片氮素含量的估算模型。研究结果表明：FISS地面成像光谱仪和ASD地面非成像光谱仪对氮素具有敏感指示的波段区间一致，480~520nlrl和680~720nin区间的波段范围与氮素含量具有较高的相关性；导数光谱特征所得精度要比原始光谱特征所得精度高，说明导数光谱对氮素含量的指示更为敏感，可以更加有效地提取氮素含量信息；反演模型对氮素含量反演精度的影响低于光谱特征对精度的影响。研究表明FISS数据由于其独特的测量方式和图谱合一的特点，可以更加准确地用于植被氮素含量的反演，实现植被叶片氮素含量的快速实时无损监测及图像可视化表达。

As the supplement of spaceborne and airborne imaging spectrometer system, Field Imaging Spectrometer System（FISS） spans a very broad range of applications to achieve fast fine detection of plant population and individual plant. Imaging spectrometer system of this new kind could provide vital information to obtain the fine spatial information of vegetation physiological parameters, especially for which spaceborne or airborne remote sensing could not be competent. In this study, we aimed to explore the relationships between image features, spectral reflectance, derivative spectra, spectral indices and nitrogen content of bean leaves with the combination of Field imaging spectrometer system （FISS） and Analytical Spectral Device （ASD）. The results showed that the spectral range of ASD that sensitive to nitrogen content was consistent with FISS. The spectral range of 480-520 nm and 680~720 nm was high related to nitrogen content. In addition, the derivative spectra could more accurately be used to inverse nitrogen content than spectral reflectance and is more sensitive to nitrogen content. It should pay more attention on the chosen of spectral features than inversion model of nitrogen content. The study demonstrates that for the advantages of unification of spectra and image, FISS could be used to more accurately estimate vegetation nitrogen content In order to realize the leaf nitrogen content in vegetation real-time non-destructive monitoring and image visualization.