• Reduce text

    Reduce text
  • Restore text size

    Restore text size
  • Increase the text

    Increase the text
  • Print

Grapevine flavescence dorée symptoms © Sandrine Eveillard

Grapevine flavescence dorée

Drones to monitor outbreaks of grapevine "flavescence dorée"

French vineyards, an important component in territorial and economic development (70 Departments and 16 Regions concerned) have to deal with numerous diseases; these include “flavescence dorée”, which affects nearly 450,000 ha (approximately 60% of French vineyards). To enable the early detection of outbreaks, the Joint Research Unit for Agroecology in Dijon is involved in a project on drone-borne imaging.

Updated on 01/04/2016
Published on 09/07/2015

Flavescence dorée: an increasingly common disease

This grapevine disease is caused by a phytoplasma (a bacterium devoid of cell walls) which lives in the phloem of the plant (through which the sap circulates). This pathogenic agent is transmitted by a sucking insect, the grape leaf hopper (Scaphoideus titanus), which feeds off sap and releases the bacteria that causes important losses in vineyards.  Flavescence dorée, a notifiable quarantine disease throughout France, causes a variety of symptoms such as discolouration of the leaves, wood that is little or not lignified, dead inflorescences, or a wilting of grape clusters, etc.

Early detection that is necessary but time-consuming

To control this highly contagious and incurable disease, it is essential to detect any outbreaks as early as possible. During the period of detection - which is short and coincides with harvest - observations are made vine by vine, and it is not certain that the entire vineyard will be checked.  Other detection methods could be envisaged, such as on-board systems or remote sensing, but because of their cost, their effects on crops (injuries, soil compaction) or the risk of disseminating the disease, drones may prove to be more appropriate.

The pertinence of drone-borne detection: initial parameters evaluated during a feasibility study

To validate the choice of drone-borne detection, the scientists tested several parameters.  Image capture, in particular, by a multirotor drone equipped with an Olympus EPL-2 12 Mpixels camera, was performed at different heights (from 3 to 9 metres above the ground), at a speed of 2 metres/sec., with a view angle of 45° to 90° (depending on longitudinal or transversal flights across the vineyard) and finally at different rates of exposure (1/800s – 1/1000s). Similarly, the pre-processing algorithms applied to the images were tested and compared with visual annotations of the plots, in order to evaluate detection - or not - of the disease, and the pertinence of coupling texture/colour/light spectrum data.  When the plots were very severely infested, it was found that pre-processing was sufficient, but this was not the case with respect to mildly infested plots.

Characterising the disease by drone-borne imaging

In order to perfect this detection, the scientists are now initiating a partnership with the companies Novadem and Global Sensing Technologies (GST). The former is developing a drone equipped with a high-resolution, multi-spectrum sensor and a field guidance and perception system to analyse grapevine diseases. This robotic tool will be coupled with a software for the analysis and geographical representation of outbreaks, designed by GST which requires pre-processing performed by the Joint Research Unit for Agroecology in Dijon. To overcome the technological obstacles (monitoring of the terrain and flight safety, visual detection and classification of the disease), several configurations will be evaluated.
These initial results, which now need to be refined and finalised, were achieved in close collaboration with the Bureau Interprofessionnel des Vins de Bourgogne (BIVB) (Burgundy Wine Board). They have demonstrated the potential economic importance to the wine industry of drone-borne imaging to detect vines infested by flavescence dorée: a reduction in the time required for diagnosis, an improved targeting of interventions, more effective action on the pathogen, and reductions in pesticide use and in the removal of contaminated plants.  This feasibility study has now given rise to a government-sponsored project called DAMAV on the automated detection of grapevine diseases (Détection automatique des maladies de la vigne), being carried out by the Pégase, Risques, Aerospace Valley and Vitagora competitiveness clusters


  • Frédéric Cointault et al., « Détection de foyers infectieux de flavescence dorée par imagerie par drone », Les Rencontres du Végétal, 8ème édition, Session Sciences et technologies de l’information et de la communication au service du végétal spécialisé, , 12-13 January 2015, Angers.