Synergistic Use of LiDAR and Hyperspectral Data in Vineyard Classification: A Case Study from the Tokaj Region, Slovakia

Hyperspectral (HS) and LiDAR sensing provide complementary information for vineyard monitoring. HS imagery captures detailed spectral signals related to canopy phys-iology and biochemistry but is often affected by contamination from inter-row soil and weeds. LiDAR offers precise measurements of canopy structure yet lacks biochemical sen-sitivity. Their integration has the potential to overcome these limitations, and remotely pi-loted (unpiloted) aerial vehicles (UAVs) provide the flexible platform needed to collect both datasets at very high resolution over vineyards. In this study, UAV-based HS and LiDAR data were acquired in the Slovak part of the Tokaj wine region to evaluate their combined use for vine-versus-non-vine classification. Three classification scenarios were tested: (1) HS data reduced by principal component analysis (PCA), (2) PCA-transformed HS data combined with a LiDAR-derived canopy height model (CHM), and (3) simulated RGB bands combined with CHM. Accuracy was assessed using 1,000 independent field-validated refer-ence points. The HS-only classification achieved the lowest performance, with a vineyard producer’s accuracy of 48.9% and overall accuracy (OA) of 89%. Including CHM substan-tially improved results, yielding OA of 96%, κ = 0.85, and an F1-score of 0.88. The RGB+CHM scenario achieved intermediate results (OA 89%, κ = 0.60), showing that struc-tural information enhances even broadband UAV imagery. These findings demonstrate the added value of integrating HS and LiDAR data collected from UAVs. The approach enables reliable vineyard mapping, providing a foundation for applications such as vigour monitor-ing, yield estimation, and site-specific canopy management in precision viticulture.