Detection of Nitrogen Content Changes of Rape Leaf Using Hyperspectral Imaging

Abstract

Hyperspectral imaging (400-1000nm) and artificial neural network (ANN) techniques were investigated for the detection of nitrogen content changes of rape leaf. Measuring SPAD value of rape leaf by using SPAD (Soil and Plant Analyzer Development).A hyperspectral imaging system was established to acquire hyperspectral data. Principal component analysis(PCA) was used to obtain principal component images, as well as to select the optimal wavelength(s). ANN was applied to establish the model between the spectral reflection values and SPAD values. The prediction results were obtained for the nitrogen content of rape leaf with the correlation of prediction of R=0.9237. The results show that the hyperspectral imaging has good classification on different nitrogen content of rape leaf.

Research on potato (Solanum tuberosum L.) nitrogen nutrition diagnosis based on hyperspectral data

“…Zou et al. (2011) explored the application of hyperspectral imaging (400–1000 nm) and artificial neural network (ANN) technology in detecting changes in nitrogen content in Brassica napus leaves. They used ANN to create a model that was highly correlated ( R 2 [coefficient of determination] = 0.924) to SPAD meter readings.…”

“…Wang et al (2020a) utilized hyperspectral imaging and partial least squares regression to create a model that can be used to estimate the NNI of Camellia sinensis. Zou et al (2011) explored the application of hyperspectral imaging (400-1000 nm) and artificial neural network (ANN) technology in detecting changes in nitrogen content in Brassica napus leaves. They used ANN to create a model that was highly correlated (R 2 [coefficient of determination] = 0.924) to SPAD meter readings.…”