UAS-Based Thermal Imaging for Water Stress Assessment in Precision Viticulture: a Case Study on a Vineyard In Tuscany, Italy

Water stress induced by drought represents a critical factor affecting vine physiology, grape yield, and wine quality. In this study, unmanned aerial system (UAS)-based thermal remote sensing is employed to assess vine water stress through a direct comparison between irrigated and non-irrigated vine rows in a Chianti Classico vineyard (Italy). High-resolution visible (VI) and thermal infrared (TI) imagery were acquired using a multirotor UAS equipped with RGB and thermal sensors. Thermal orthophotos were generated via photogrammetric processing and analysed to estimate canopy temperature using two different pixel-selection approaches: a temperature-based mask and a vegetation segmentation algorithm applied to the VI orthophoto. Results show that temperature masking allows discrimination between irrigated and non-irrigated rows, yielding canopy temperature differences of approximately 0.6 °C. However, this approach is sensitive to season-dependent thresholds and to the presence of vegetated soil and shadow pixels. By contrast, the proposed vegetation segmentation method effectively isolates canopy pixels, leading to a more robust temperature estimation and an increased temperature difference of up to 2.2 °C between irrigated and non-irrigated rows. The findings demonstrate the potential of UAS-based thermal imaging combined with canopy segmentation for improved water stress assessment in precision viticulture.