Smartagb: Aboveground Biomass Estimation of Sorghum Based on Spatial Resolution, Machine Learning and Vegetation Index

This work aims to explore the feasibility of predicting and estimating aboveground biomass (AGB) sorghum using multispectral images captured by UAVs, clarify quantitative relationship between vegetation index AGB based on different spatial resolutions, build an estimation model UAV under resolutions. Combining resolution, index, machine learning, a training set is used train model, verification verify select best prediction corresponding The three models resolutions are classic learning models. 1) when resolution 0.017m, precision obtained from random forest R2=0.8961, MAE=26.4340, RMSE=32.2459. 2) 0.024m, accuracy Lasso algorithm R2=0.8826, MAE=31.106, RMSE=40.2937; 3) 0.030m, decision tree R2=0.8568, MAE=30.3373, RMSE=40.8082; 4) model's decreases with decrease resolution. results show that combination effective, fast, accurate method.