UAV Hyperspectral Characterization of Vegetation Using Entropy-Based Active Sampling for Partial Least Square Regression Models

Optimization of agricultural practices is key for facing the challenges modern agri-food systems, which are expected to satisfy a growing demand food production in landscape characterized by reduction cultivable lands and an increasing awareness sustainability issues. In this work, operational methodology characterization vegetation biomass nitrogen content based on close-range hyperspectral remote sensing introduced. It unsupervised active learning technique suitable calibration partial least square regression. The proposed relies innovative usage Shannon’s entropy allows set-up incremental monitoring framework from scratch aiming at minimizing field sampling activities. Experimental results concerning estimation grassland returned RMSE values 2.05 t/ha 4.68 kg/ha, respectively. They comparable with literature, mostly relying supervised frameworks confirmed suitability environments.