Toward a Holistic Framework Using Machine Learning Approaches for Volume Estimation of Olive Trees

This study explores UAV–LiDAR applications for the precise estimation of olive tree canopy volume in agricultural settings. Our framework employs 3D point cloud analysis with a workflow comprising aerial imagery for tree detection and extraction of key morphometric parameters. Comparing multiple machine learning approaches (random forest, linear regression, and decision tree algorithms) against ground measurements, the random forest model demonstrated superior performance with an R² of 91%. Results confirm that UAV–LiDAR systems effectively capture detailed tree structural measurements with high accuracy, offering significant potential for precision irrigation decision support. This methodology provides agronomists with reliable tools for the quantitative assessment of tree canopy volumes, enabling data-driven orchard management decisions.