Prediction of SPAD value and distribution of rape leaf based on hyperspectral imaging technology

Chlorophyll content plays an important role in the growth of crop. Chlorophyll content of crop not only shows the situation of crop growth but also has important significance in disease diagnosis. By using hyperspectral imaging technology, chlorophyll content can be diagnosed non-destructively. Before the correlation analysis, the experiment gets the SPAD values and vegetation indexes. The result shows that VOG1 quadratic model has the best performance (R²=0.83 and RMSE=2.68). After using the best model for each pixel of the image, we can get a grayscale image, on which each pixel value is the SPAD value of this point. Then grayscale image is changed into pseudo-color image. As can be seen from the image, both from the veins to the mesophyll and from the bottom to the top, chlorophyll content is increased gradually. The study can provide the rape researchers with more information and a new thinking of rape studies.