New approach for predicting nitrogen and pigments in maize from hyperspectral data and machine learning models

Abstract Fast diagnostics from hyperspectral data and machine learning (ML) models to predict nitrogen (N) content and pigments in maize crop is a challenge to optimize nitrogen fertilization. Therefore, this field research identifies which ML algorithms accurately estimate nitrogen, chlorophyll, and carotenoid contents in maize crop leaves at different phenological phases using hyperspectral band data. The treatments were arranged in a factorial scheme having four N doses (0, 54, 108, and 216 kg ha − 1 ) combined with five leaf collection seasons at phenological stages V6 to V14. The ML models tested were artificial neural networks – ANN, decision tree adapted for prediction problems – M5P, REPTree decision tree, random forest - RF, polynomial support vector machine – SVMP, and ZeroR - ZR (control). Pearson correlations were estimated between in situ variables and those predicted by ML models. Spectral bands 530–550 nm and 690–750 nm are effective wavelengths as well as RF and SVM are effective models for the estimation of nitrogen and pigment contents. These methods can be applied in precision farming practices and assist in decision making for N management in maize crop.