Assessing day and night UAV-Based estimates of crop characterization using LiDAR

Authors: Flynn, Kyle; BAATH, GURJINDER - Texas Agrilife Research; SAPKOTA, BALA RAM - Texas Agrilife Research; Smith, Douglas

Submitted to: Smart Agricultural Technology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/14/2026
Publication Date: 1/17/2026
Citation: Flynn, K.C., Baath, G., Sapkota, B., Smith, D.R. 2026. Assessing day and night UAV-Based estimates of crop characterization using LiDAR. Smart Agricultural Technology. https://doi.org/10.1016/j.atech.2026.101805.
DOI: https://doi.org/10.1016/j.atech.2026.101805

Interpretive Summary: Advancements among remote sensing technologies and platforms have provided opportunity to collect spatially continuous biophysical parameters for the benefit of precision agriculture. Specifically, unmanned aerial vehicles with mounted LiDAR sensors have promise in characterizing crop canopy height and leaf area index (LAI). This study aimed to demonstrate the assessment of crop height and LAI using UAV LiDAR-based methodologies under varying light conditions (e.g. day or night). The findings within provide the foundation for future works aiming to incorporate day and/or night LiDAR-based precision agriculture as both LiDAR-based crop height and LAI measures were promising no matter the time of day.

Technical Abstract: Advancements among remote sensing technologies and platforms have provided opportunity to collect spatially continuous biophysical parameters for the benefit of precision agriculture. Specifically, unmanned aerial vehicles with mounted LiDAR sensors have promise in characterizing crop canopy height (CH) and leaf area index (LAI). This study aimed to demonstrate the assessment of CH and LAI using UAV LiDAR-based methodologies under varying light conditions (e.g. day or night). The methods incorporated included the use of the cloth simulation filter (CSF) to assign ground/vegetation status of the LiDAR point cloud (CH calculations), and then further benefited from the inclusion of laser penetration and intensity-based LiDAR indices to estimate LAI. These methods were replicated with identical day and night flights for a multi-planting date (seven dates) corn study conducted in Temple, Texas. Findings suggest that both CH (R2=0.91; RMSE=15.43) and LAI (R2=0.78, RMSE=0.06) measurements are highly correlated when comparing in-situ and LiDAR-based measurements. When comparing between LiDAR data collected during day and night flights, CH had very little difference ('R2=0.01) and were found to be highly replicable. LAI measures presented more differences ('R2=0.05) between the day and night collections that we attribute to leaf turgor due to the relatively minor differences among models. The findings within provide the foundation for future works aiming to incorporate day and/or night LiDAR-based precision agriculture.