Water stress, which affects yield and wine quality, is often evaluated using the midday stem water potential (Ψ_stem). However, this measurement is acquired on a per plant basis and does not account for the assessment of vine water status spatial variability. The use of multispectral cameras mounted on unmanned aerial vehicle (UAV) is capable to capture the variability of vine water stress in a whole field scenario. It has been reported that conventional multispectral indices (CMI) that use information between 500–800 nm, do not accurately predict plant water status since they are not sensitive to water content. The objective of this study was to develop artificial neural network (ANN) models derived from multispectral images to predict the Ψ_stem spatial variability of a drip-irrigated Carménère vineyard in Talca, Maule Region, Chile. The coefficient of determination (R²) obtained between ANN outputs and ground-truth measurements of Ψ_stem were between 0.56–0.87, with the best performance observed for the model that included the bands 550, 570, 670, 700 and 800 nm. Validation analysis indicated that the ANN model could estimate Ψ_stem with a mean absolute error (MAE) of 0.1 MPa, root mean square error (RMSE) of 0.12 MPa, and relative error (RE) of −9.1%. For the validation of the CMI, the MAE, RMSE and RE values were between 0.26–0.27 MPa, 0.32–0.34 MPa and −24.2–25.6%, respectively.

The use of Unmanned Aerial Vehicles (UAVs) in viticulture permits the capture of aerial Red-Green-Blue (RGB) images with an ultra-high spatial resolution. Recent studies have demonstrated that RGB images can be used to monitor spatial variability of vine biophysical parameters. However, for estimating these parameters, accurate and automated segmentation methods are required to extract relevant information from RGB images. Manual segmentation of aerial images is a laborious and time-consuming process. Traditional classification methods have shown satisfactory results in the segmentation of RGB images for diverse applications and surfaces, however, in the case of commercial vineyards, it is necessary to consider some particularities inherent to canopy size in the vertical trellis systems (VSP) such as shadow effect and different soil conditions in inter-rows (mixed information of soil and weeds). Therefore, the objective of this study was to compare the performance of four classification methods (K-means, Artificial Neural Networks (ANN), Random Forest (RForest) and Spectral Indices (SI)) to detect canopy in a vineyard trained on VSP. Six flights were carried out from post-flowering to harvest in a commercial vineyard cv. Carménère using a low-cost UAV equipped with a conventional RGB camera. The results show that the ANN and the simple SI method complemented with the Otsu method for thresholding presented the best performance for the detection of the vine canopy with high overall accuracy values for all study days. Spectral indices presented the best performance in the detection of Plant class (Vine canopy) with an overall accuracy of around 0.99. However, considering the performance pixel by pixel, the Spectral indices are not able to discriminate between Soil and Shadow class. The best performance in the classification of three classes (Plant, Soil, and Shadow) of vineyard RGB images, was obtained when the SI values were used as input data in trained methods (ANN and RForest), reaching overall accuracy values around 0.98 with high sensitivity values for the three classes.

Aerial and terrestrial thermography has become a practical tool to determine water stress conditions in vineyards. However, for proper use of this technique it is necessary to consider vine architecture (canopy zone analysis) and image thresholding approaches (determination of the upper and lower baseline temperature values). During the 2014–2015 growing season, an experimental study under different water conditions (slight, mild, moderate, and severe water stress) was carried out in a commercial vineyard (Vitis vinifera L., cv. Carménè). In this study thermal images were obtained from different canopy zones by using both aerial (>60 m height) and ground-based (sunlit, shadow and nadir views) thermography. Using customized code that was written specifically for this research, three different thresholding approaches were applied to each image: (i) the standard deviation technique (SDT); (ii) the energy balance technique (EBT); and (iii) the field reference temperature technique (FRT). Results obtained from three different approaches showed that the EBT had the best performance. The EBT was able to discriminate over 95% of the leaf material, while SDT and FRT were able to detect around 70% and 40% of the leaf material, respectively. In the case of canopy zone analysis, ground-based nadir images presented the best correlations with stomatal conductance (gs) and stem water potential (Ψstem), reaching determination coefficients (r2) of 0.73 and 0.82, respectively. The best relationships between thermal indices and plant-based variables were registered during the period of maximum atmospheric demand (near veraison) with significant correlations for all methods.