基于成像光谱仪的冬小麦苗期冠层叶绿素密度监测

Monitoring Canopy Chlorophyll Density in Seedlings of Winter Wheat Using Imaging Spectrometer

利用自主研发的MSI200型成像光谱仪对冬小麦苗期叶绿素密度进行遥感监测,并与ASD Fieldspc FR2500型非成像地物光谱仪测定数据相比。结果表明,两种仪器所得R720、FD660、FD550、TCARI、GNDVI、PPR（550,450）、NRI等光谱特征参量均与叶绿素密度显著相关,拟合度较高,说明MSI200型成像光谱仪测定的作物光谱数据是可靠的。对筛选出的模型进行检验,ASD Fieldspc FR2500模型的预测精度达86.61%～92.79%,MSI200达91.26%～95.54%,其中PPR对叶绿素密度预测精度最高,RMSE分别为0.8391和0.0979。与传统非成像地物光谱仪相比,成像光谱仪能够提取纯冠层光谱信息,所得模型精度高、误差小,筛选的植被指数和特征波段对于航天、航空遥感器的定标有重要意义。

Imaging spectrometer is a new kind of remote sensing sensor, which receives images of ground objects and their spectrum components. It has great potentials in accurately quantitative analysis with remote sensing of high spatial and spectral resolutions. MSI200 is an imaging spectrometer manufactured by the authors. It was used in this study to monitor canopy chlorophyll density in winter wheat （Triticum aestivum L.） at seedling stage in comparison with ASD Fieldspc FR2500, a well-known non-imaging spectrometer. Correlation analysis showed that R720, FD660, FD550, TCARI, GNDVI, PPR （550, 450）, and NRI obtained by the 2 spectrometers were both significantly correlated （P〈0.01） with canopy chlorophyll density. This primarily indicated the reliability of MSI200. The precisions of predicted model on chlorophyll density were 91.26-95.54% and 86.61-92.79% for MSI200 and ASD Fieldspc FR2500, respectively. Among these parameters, PPR was the best one to monitor canopy chlorophyll density with root mean square error （RMSE） of 0.0979 and 0.8391 for MSI200 and ASD Fieldspc FR2500, respectively. Compared with the traditional non-imaging spectrometers, MSI200 predicted better the canopy chlorophyll density in wheat with smaller errors, and the selected wavelengths were important to spaceflight and airborne remote sense.