Contact

Discovering Invisible VIne Nutrition Elements

DiVine project aims to develop more detailed digital maps of (organic) vineyards that can distinctly identify areas likely affected by disease from those suffering from nutritional deficiencies. To reach this goal, our project will focus on the following objectives:

O1: Develop a framework (i.e., pipeline) for generating digital maps that provide more comprehensive information about the vineyard’s condition.

O2: Incorporate this pipeline into an accessible interface, such as a web application, ensuring ease of use for vineyard managers.

O3: Calculate the percentage of the vineyard suspected to be suffering from nutritional deficiencies, enabling targeted interventions.

O4: Combine the insights gained from remote sensing technology with data collected directly from the ground for a more holistic understanding of vineyard health.

Responsible organisation

Challenges and how they will be addressed

Traditional vineyard monitoring is a labor-intensive process, requiring extensive foot inspections that can take approximately 10 hours for every 10 hectares. In contrast, our technology can accomplish the same task in just up to 1 hour per 10 hectares. This conventional method not only demands significant manpower but also relies heavily on expert knowledge to identify the causes of grapevine stress. As climate change continues to introduce new diseases and symptoms, this becomes increasingly challenging.

At Veles Sense, we are developing a comprehensive pipeline for generating digital maps that provide detailed information about disease presence and highlight other areas under stress, such as those caused by nutrient deficiencies. This innovative approach offers vineyard owners deeper insights into the health of their crops, enabling them to save resources and boost yields. Our solution integrates remote sensing via drone imaging with ground-based activities to produce highly accurate maps. Additionally, we offer a user-friendly web application that makes it easy for vineyard owners to access and utilize this valuable information.

Tech components and data

A Veles Sense approach leverages remote sensing technologies that capture images beyond visible light (multispectral data), enabling the detection of subtle changes and hidden stress symptoms in vegetation. This imagery is integrated with advanced computer vision and machine learning techniques, supplemented by on-the-ground expert analysis. The primary objective of this project is to develop a comprehensive pipeline for creating detailed digital maps of organic vineyards, enhancing our understanding of areas with nutritional deficiencies. The process is segmented into three main parts: 

  1. Ground truth activities, data collection, and dataset creation: here, the multispectral images will be collected, labelled, and prepared as a training set for machine learning-based approaches.
  2. Image processing activities and Web App development: initial steps include automatic plant counting and detecting missing plants using object detection algorithms. Several vegetation indices will be calculated using spectral bands from the multispectral camera. By aligning vegetation index-based maps with ground truth data, areas unaffected by disease are identified. Assuming disease is not a factor, other anomalies likely indicate stress due to water or nutritional deficiencies. Enhanced understanding of ground conditions will be developed through these analytical steps. Together with the development of machine learning and image processing code we will develop a user-friendly web application.

Result evaluation: Specific locations suspected of nutritional stress are pinpointed for automatic digital sampling, determining the optimal number of samples for further analysis by using math-based and/or machine learning optimization approaches. Samples are collected and analyzed in the laboratory. An estimate of the percentage of the vineyard suffering from nutritional deficiencies is calculated and added to the final maps.  

Expected outcomes

The project is structured into 3 distinct phases, each expected to yield specific outcomes:

Design phase: The outcome of this initial phase is a comprehensive pipeline designed for the creation of more informative digital maps. This pipeline will encompass not only the computational aspects but also include all processes such as ground truthing activities, data collection, and expert evaluations. This holistic approach ensures that the pipeline is robust and tailored to the specific needs of organic vineyard monitoring. 

Development phase: The result of this phase is the development of a sophisticated software/code that integrates all necessary computer vision and machine learning modules. This will also include the deployment of a web application designed for user interaction and data management, providing a practical tool for vineyard managers and agronomists. 

Market phase: The final phase aims to produce a well-articulated market and sales strategy. This strategy will outline the approach for introducing the product to the market, identifying potential customers, and effectively selling the solution. 

Veles Sense is at the forefront of developing cutting-edge solutions for early grapevine disease detection. By integrating multispectral drone imagery with close-range imaging and advanced machine learning techniques, we aim to revolutionize the way vineyard health is monitored and managed. Our primary focus is on the detection of grapevine trunk diseases, with a particular emphasis on Esca—a devastating disease responsible for global economic losses estimated at 8-9 billion euros annually. In addition to disease detection, we are continuously enhancing our solution to identify vineyard stress caused by other factors, including irrigation issues, nutrient deficiencies, and additional diseases, ensuring comprehensive vineyard management and protection.

Drone Stakeholders survey (e.g. Drone manufacturers, Drone service providers, Software developers)

Start Typing

    Stay Connected
    We appreciate your feedback.

    Join our newsletter

      Join our newsletter

      Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or Research Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.

      ©2022 ICAERUS Project