What relevant information can be identified by experts on unmanned aerial vehicles' visible images for precision viticulture? - Agropolis Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

What relevant information can be identified by experts on unmanned aerial vehicles' visible images for precision viticulture?

Résumé

Unmanned aerial vehicles (UAV) offer interesting alternatives to satellites or airplanes regarding flight agility and image resolution. These sensor platforms may well be used to monitor vines field all throughout the vine's growing season at a very high spatial resolution. They could provide useful information, different to that normally considered in the literature. To identify the possible uses of UAV images in viticulture, a specific exploratory survey was put into place. This study aimed at identifying (i) relevant information that growers and advisers (G&A) can extract from UAV images and (ii) the added value that this information can have for both G&A's vineyard management decisions. This approach was conducted on an 11.3 ha commercial vineyard with soil, climate and a training system representative of vineyards in the south of France. UAV-based visible images (25 mm resolution) were acquired every two weeks from budburst to harvest by several UAV companies. Images were shown to a panel of G&As over six sessions during the growing season. Each of these sessions consisted of (i) an individual period during which images were first presented one at a time to each expert and then all together in chronological order from budburst to harvest, and (ii) a collective period during which G&As were asked to share and discuss their point of view. In this exploratory survey, the application of the proposed methodology demonstrated that most of the information on vine status, soil and vineyard environment could be extracted from UAV-based visible images by the experts, thus showing high interest in developing specific image processing techniques to extract this information from images. Results showed that this information was of great interest throughout the growing cycle of the vine, particularly for advisers, as a support to drive management strategies. The outputs of this exploratory study should be confirmed in other contexts than the Languedoc, France region to extrapolate the observed conclusions.
Fichier non déposé

Dates et versions

hal-02608581 , version 1 (16-05-2020)

Licence

Copyright (Tous droits réservés)

Identifiants

Citer

L. Pichon, C. Leroux, Catherine Macombe, James Taylor, Bruno Tisseyre. What relevant information can be identified by experts on unmanned aerial vehicles' visible images for precision viticulture?. 11th European Conference on Precision Agriculture (ECPA), Jul 2017, Edinburgh, United Kingdom. pp.278-294, ⟨10.1007/s11119-019-09634-0⟩. ⟨hal-02608581⟩
22 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More