论文部分内容阅读
本文通过整个光谱范围内一阶导数光谱反射率与叶绿素、可溶性糖和可溶性蛋白的相关系数显著的波段,建立高光谱预监测水肥耦合条件下的夏玉米光合特性以及碳氮代谢,进而为玉米高产提供依据。在玉米拔节期和大喇叭口期选择596、1 025和924 nm,吐丝期和乳熟期选择638、1 068和965 nm这几个显著性波段的实测值来建立估测模型。研究结果表明,拔节和大喇叭口期叶片叶绿素SPAD值的估测模型为y=28832.45ρ596+39.34,可溶性糖含量的估测模型为y=640.54ρ1025+7.92,可溶性蛋白含量一阶导数光谱估测模型为y=4092.90ρ924+5.63,而吐丝期和乳熟期叶片叶绿素SPAD值的估测模型为y=134151.00ρ638+129.92,可溶性糖含量的估测模型为y=524.80ρ1068+9.20,可溶性蛋白含量一阶导数光谱估测模型为y=7321.61ρ965+36.64。所建立的高光谱预测模型在本试验所属的时空范围内能很好地预测和反演玉米生长状况。
In this paper, hyperspectral pre-monitoring of photosynthetic characteristics and carbon and nitrogen metabolism of summer maize under hyperspectral was carried out through the band of correlation coefficients between the first derivative of spectral reflectance and chlorophyll, soluble sugar and soluble protein in the whole spectral range. Provide evidence. At the jointing and boll stage of corn, 596, 1 025 and 924 nm were selected, and the measured values of 638, 1 068 and 965 nm at the silking and milking stages were selected to establish the estimation model. The results showed that the SPAD value of the leaf at jointing and boll stage was y = 28832.45ρ596 + 39.34, and the estimation of soluble sugar content was y = 640.54 ρ1025 + 7.92. The first derivative spectra of soluble protein The model was y = 4092.90ρ924 + 5.63, while the prediction model of leaf chlorophyll SPAD value at spinning and milking stage was y = 134151.00ρ638 + 129.92. The estimated model of soluble sugar content was y = 524.80 ρ1068 + 9.20, soluble protein The content of the first derivative spectral estimation model y = 7321.61ρ965 +36.64. The established hyperspectral prediction model can predict and inversely predict the growth of maize within the spatial and temporal scope of this experiment.