Infra-red thermal image analysis for grapevines

Abstract

Infrared thermal images (IRTI) have been used for grapevine research since the early 90’s. Even though its promising results in the assessment of canopy stomatal conductance and plant water status, from the beginning and recent research publications, it has not been fully applied on a commercial scale yet. It is believed that the bottleneck for this technology is the lack of reliable automation tools for IRTI analysis. Accurate and reliable automation techniques will allow the use of this technique to assess the spatial variability of physiological processes within the canopy using infrared cameras mounted on moving vehicles, drones, octocopters or robots. Automated analysis systems are requirement of The Vineyard of The Future initiative, which is an international effort to establish fully monitored vineyards in the most prominent viticultural and winemaking areas in the world. In this work, a semi-automated IRTI analyses performed using a code written in MATLAB® for estimate dry and wet references excluding non-leaf temperatures was compared with evaporimeter (EvapoSensor, Skye Instruments Ltd, Powys, UK) measurements used to provide dry and wet references from IRTIs. Results obtained from this research (grapevines cv. Tempranillo) showed good and statistically significant correlations between temperature references obtained from IRTI analysis and measured values. This work constitutes one additional step forward to the implementation of thermal imaging as an automated routine technique for physiological vineyard assessment from proximal sensing and unmanned aerial vehicles (UAV) platforms.