基于高光谱的苹果树冠层磷素状况估测模型研究

Estimation of Phosphorus Content in Apple Tree Canopy Based on Hyperspectrum

利用高光谱技术估测了苹果树冠层的磷素含量。先用ASD Field Spec3型地物光谱仪测定了春梢停止生长期苹果树冠层的高光谱反射率,并对光谱数据进行了多种变换处理。然后对其与磷素含量进行了相关分析,找出了与磷素相关性较显著的光谱参量,并通过逐步回归分析建立了磷素估测模型。结果表明,近红外波段是苹果树冠层磷素的敏感波段;808 nm、921 nm、1195 nm、1272 nm及其组合的归一化红外光谱指数与苹果树冠层磷素高度相关。在构建的估测模型中,以808 nm、921 nm、1195nm、1272 nm及其组合的归一化红外光谱指数为自变量构建的高光谱估测模型的估测效果最佳。该研究实现了苹果树冠层磷素含量的快速估测,同时也为苹果的实时营养诊断提供了理论依据。

The phosphorus content of apple tree canopy is estimated by using a hyperspectral remote sensing technology. Firstly, a Field Spec 3 type field spectrometer from ASD company is used to measure the hyperspectral reflectance of apple tree canopy in the shoot growth stopping period in spring. The measured spectral data are processed in a variety of transformation forms. Then, correlation of the data with the phosphorus content is analyzed. The spectral parameters more remarkably correlated with the phosphorus content are found. A phosphorus content estimation model is established after the regression analysis is made. The result shows that the near infrared band is the sensitive band of the phosphorus content of apple tree canopy. The wavelengths of 808 nm, 921 nm, 1195 nm and 1272 nm and the normalized infrared spectral index combined by these wavelengths are highly correlated with the phosphorus content of apple tree canopy. This is verified in the established estimation model. This research not only realizes the fast estimation of phosphorus content of apple tree canopy, but also provides the theoretical basis for the real-time diagnosis of nutrition of aDDle trees.