A new data collection campaign in Alsace<\/strong><\/h4>In 2026, the French Institute of Vine and Wine (IFV)<\/a>, based in Colmar, carried out a new data collection campaign across the vineyards of Use Case Pilot 1 (UCP1)<\/a> in Alsace, as part of the European STELLA<\/a> project. The campaign combined drone flights, tractor-mounted camera acquisitions and field surveys to collect complementary information on grapevine health.<\/p>The objective is to contribute to the development of digital tools capable of detecting two important grapevine diseases: Bois noir and Grapevine leafroll disease<\/strong>. By combining remote sensing technologies, artificial intelligence and field expertise, STELLA aims to improve disease monitoring and support more informed decision-making in viticulture.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t\t\t\t![\"\"](\"https:\/\/stella-pss.eu\/wp-content\/uploads\/2026\/11\/3.png\")
\t\t\t\t\t\t\t\t\t\t\tOnboard camera mounted on a tractor collecting canopy images<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t![\"\"](\"https:\/\/stella-pss.eu\/wp-content\/uploads\/2026\/11\/1.png\")
\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t\t\t\t![\"\"](\"https:\/\/stella-pss.eu\/wp-content\/uploads\/2026\/11\/2.png\")
\t\t\t\t\t\t\t\t\t\t\tDavid Kremer, the drone pilot of the Chamber of Agriculture<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\t\t\t\t\tCombining complementary data collection methods<\/strong><\/h4>Monitoring grapevine diseases across vineyard plots can be challenging, particularly when symptoms vary between vines and may be difficult to identify consistently across large areas. Combining different data collection methods offers an opportunity to capture complementary information and develop more reliable detection models.<\/p>
Three approaches are used in UCP 1 to collect images of the canopy:<\/p>
- Drone flights<\/strong>: Aerial imagery provides an overview of vineyard plots and allows the observation of spatial variability across the vineyard, collecting images in the RGB and hyperspectral. While disease symptoms can often be detected on the canopy, differentiating between specific grapevine diseases is difficult because symptoms may appear similar from an aerial perspective and fine diagnostic details are not always visible.<\/li>
- On-board camera acquisitions:<\/strong> images collected in the close to the vegetation provide a different perspective, capturing visual details of the grapevine canopy that may be useful for identifying disease symptoms.<\/li>
- Pedestrian prospections:<\/strong> field observations provided by the expert human eye provide essential reference data on grapevine health. These prospections help verify the presence or absence of symptoms and provide the ground-truth information needed to develop and evaluate disease detection models.<\/li><\/ul>
From field data to AI-based disease detection<\/strong><\/h4>The images collected during the campaign will contribute to the development of artificial intelligence models designed to detect Bois noir and Grapevine leafroll disease.<\/p>
The development of reliable detection models depends not only on the quantity of images collected, but also on the availability of high-quality, verified field observations. By linking images to information gathered during ground surveys, the project can provide the data needed to train and evaluate models under real vineyard conditions.<\/p>
Combining these different sources of information is a key step towards developing detection models that can make the most of digital imagery while taking into account the complexity of disease symptoms observed in the field.<\/p>
Once developed, these models are intended to be integrated into the STELLA Plant Protection Service Platform (PSS), contributing to digital tools that can support disease surveillance and help users better understand the health status of their vineyards.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t
\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t\t\t\t![\"\"](\"https:\/\/stella-pss.eu\/wp-content\/uploads\/2026\/11\/square-pic-blog--1024x1024.png\")
\t\t\t\t\t\t\t\t\t\t\tSymptoms of grapevine leafroll on leaves<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\n\t\t\t\t\n\t\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t\t\t\t![\"\"](\"https:\/\/stella-pss.eu\/wp-content\/uploads\/2026\/11\/square-pic-blog-1-1024x1024.png\")
\t\t\t\t\t\t\t\t\t\t\tA grapevine showing symptoms of Bois noir<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\n\t\t\t\t\t\n\t\t\t\t\n\t\t\t\t
The objective is to contribute to the development of digital tools capable of detecting two important grapevine diseases: Bois noir and Grapevine leafroll disease<\/strong>. By combining remote sensing technologies, artificial intelligence and field expertise, STELLA aims to improve disease monitoring and support more informed decision-making in viticulture.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t Monitoring grapevine diseases across vineyard plots can be challenging, particularly when symptoms vary between vines and may be difficult to identify consistently across large areas. Combining different data collection methods offers an opportunity to capture complementary information and develop more reliable detection models.<\/p> Three approaches are used in UCP 1 to collect images of the canopy:<\/p> The images collected during the campaign will contribute to the development of artificial intelligence models designed to detect Bois noir and Grapevine leafroll disease.<\/p> The development of reliable detection models depends not only on the quantity of images collected, but also on the availability of high-quality, verified field observations. By linking images to information gathered during ground surveys, the project can provide the data needed to train and evaluate models under real vineyard conditions.<\/p> Combining these different sources of information is a key step towards developing detection models that can make the most of digital imagery while taking into account the complexity of disease symptoms observed in the field.<\/p> Once developed, these models are intended to be integrated into the STELLA Plant Protection Service Platform (PSS), contributing to digital tools that can support disease surveillance and help users better understand the health status of their vineyards.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\tCombining complementary data collection methods<\/strong><\/h4>
From field data to AI-based disease detection<\/strong><\/h4>
\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t