【摘要】 以杨凌地区冬小麦为研究对象,使用六旋翼无人机搭载RedEdge多光谱相机进行叶绿素监测试验。共选取65个样本,每个样本为1 m×1 m的样地,在样地内选取小麦冠层的7片叶片,测量相对叶绿素含量SPAD值,取平均值作为实测值,GPS记录位置信息。地面数据测量与无人机飞行测量同步进行。用Pix4D mapper软件对无人机多光谱影像进行拼接处理,得到4个波段下小麦冠层叶片反射率光谱图像,并利用ENVI 5.1软件提取光谱反射率数据。选取8种常用光谱参数,其中与小麦SPAD相关性较高的有SAVI、EVI2、DVI、RVI、NDVI、EVI和ARVI共7种,相关系数均在0.67以上。用7种光谱参数和小麦SPAD实测值,使用一元线性回归法和多元线性回归法构建反演模型并进行精度分析,结果表明:一元线性回归法构建的SPAD-SAVI模型精度最佳,决定系数(R~2)为0.866,均方根误差RMSE为0.245,可作为无人机遥感快速、无损监测冬小麦叶绿素的技术手段。更多还原

【Abstract】 Taking winter wheat in Yangling area as the research object, a six-rotor uav(unmanned aerial vehicle) with RedEdge multispectral camera was used for chlorophyll monitoring test. A total of 65 samples were selected, each of which was 1 m×1 m sample plot. Seven leaves of wheat canopy were selected in the sample plot to measure the SPAD value of relative chlorophyll content. The mean value was taken as the measured value, and GPS recorded the location information. Ground data measurement and uav flight measurement were carried out synchronously. The Pix4 D mapper software was used to splicing multi-spectral images of uav, and wheat canopy reflectance spectra in 4 bands were extracted. Eight commonly used spectral parameters were selected, among which SAVI, EVI2, DVI, RVI, NDVI, EVI and ARVI had high correlation with wheat SPAD, and the correlation coefficients were all above 0.67. With 7 spectral parameters and wheat SPAD measured values, using unitary linear regression method and multiple linear regression method, an inverse model was constructed and precision analysis was carried out. The results showed that the SPAD-SAVI model constructed by unitary linear regression method had the best accuracy, the determination coefficient(R~2) was 0.866, and the root-mean-square error(RMSE) was 0.245. The research results can be used as a technique for rapid and nondestructive monitoring of chlorophyll in winter wheat by uav.更多还原