Near-real-time characterisation of vines

The importance of canopy structure and vigour assessment

The ability to accurately and efficiently assess canopy structure and vine vigour is essential for winegrowers aiming to optimise vineyard management. For instance, monitoring intra-block variability in canopy and vigour allows for the implementation of targeted management practices in specific areas. This approach helps reduce variability by applying management inputs where necessary. Alternatively, if winegrowers want to leverage this variability, they can implement strategies such as selective harvesting, separating the fruit into different quality classes, or selective fertilisation. These strategies can reduce costs and resources by applying nutrients only where needed.

Traditional methods for assessing canopy structure and vine vigour require considerable manual effort and expertise. While remote sensing technologies like drones and satellite imagery have improved accessibility, they introduce complexities related to data processing, legal compliance and weather conditions. To address these challenges, a research project was undertaken to explore advanced image-based techniques for vineyard monitoring to define canopy characteristics and vigour at the plant level.

Techniques evaluated

In this project, several non-destructive techniques were studied to determine canopy architecture and other key viticulture parameters (such as yield, water stress and pruning weight). The accurate and rapid quantification of these parameters can provide producers with a significant advantage, as they are closely related to vine balance and plant development during the season. This opens up possibilities for implementing management practices that improve grape quality through directed canopy manipulations, irrigation scheduling and pruning, all based on reliable and concrete information.

Main results

Canopy characterisation

For canopy characterisation (expressed as Leaf Area Index – LAI), using RGB, LiDAR and multispectral images (captured with a drone), showed strong correlations with destructive methods, with agreements of approximately 80%. Rough estimations of vigour on a large scale can be done using satellite information. However, quality-focused winegrowers are now managing premium blocks at a more detailed level, accounting for intra-block variability. Under these conditions, the techniques studied in this project could provide the precise and detailed information needed to achieve this goal. Figure 1 shows an example of the method used to process the LiDAR data.

FIGURE 1. Workflow of the methodology used to determine canopy volume from LiDAR data. This figure appears in one of the articles published as part of the project: https://www.actahort.org/books/1279/1279_34.htm.

Yield estimation

Yield estimation is a critical topic in viticulture, as traditional methods are complicated to implement and can generate significant errors, leading to logistical issues in the cellar. The results of this study confirm that RGB and RGB-D are effective methods for estimating bunch weight in both laboratory and field conditions (Figure 2). However, this approach has some practical limitations. In dense canopies, occlusion of bunches can lead to underestimations of the total weight at the vine level. Colour thresholding was very effective for classifying the bunches when a white panel was used; however, monitoring multiple vines with large structures can limit the applicability of this technique. In this regard, selecting target areas within a block can be a practical solution. Other approaches, such as complex machine learning models and nighttime imagery, should be considered for future studies.

FIGURE 2. Data acquisition of individual bunches in the field. (a) RGB image with full canopy (FC); (b) RGB image with leaf removal (LR); (c) RGB-D (Kinect mesh) with FC; and (d) RGB-D (Kinect mesh) with LR. This figure appears in one of the articles published as part of the project: https://www.mdpi.com/1424-8220/19/17/3652.

Water stress detection

The detection and quantification of water stress are crucial for sustainable viticulture in light of current climate change scenarios. In this study, water stress was assessed under different levels and conditions. Field spectroscopy and hyperspectral imaging (HSI), using both laboratory and field sensors, showed promising results. For field spectroscopy, the success of the method depends on a rigorous calibration and data acquisition protocol; small changes can cause significant differences in the spectral signature.

Although HSI is still costly and complex, it is evolving rapidly. This project evaluated HSI under contrasting water conditions (non-water-stressed vs. water-stressed vines), marking the first step in developing a methodology for HSI analysis and demonstrating the general feasibility of this technology. However, further experiments are needed to evaluate this technique under a broader range of real-world water stress conditions.

Pruning weight assessment

The concept developed for canopy characterisation was extended to assess pruning weight, an important parameter for evaluating vine balance. Weighing the pruning mass of a couple of vines is simple, but on a large scale, it becomes almost impossible. Several techniques and conditions were evaluated to identify the most suitable method. The results suggest that pruning weight can be accurately determined using image analysis (Figure 3).

FIGURE 3. Example of the method used to assess pruning weight by digital analysis.

Conclusions

The results obtained in this study indicate the potential for using non-destructive techniques (such as computer vision, HSI and spectroscopy) as vineyard monitoring tools. However, each method has its advantages and disadvantages in terms of applicability, cost and complexity. These factors must be carefully considered when evaluating and applying these techniques. Future projects in this field should focus on the operational aspects of these methods, considering elements such as automatic analysis and user interfaces.

Reference

Poblete-Echeverría, C., 2020. Final Report DVO 07: Near-real-time characterisation of vines for more efficient vineyard management. https://winetechlibrary.co.za/near-real-time-characterisation-of-grapevines-for-more-efficient-vineyard-management/.

For more information, contact Carlos Poblete-Echeverría at cpe@sun.ac.za.

Click here to get your copy of WineLand Magazine.