Complémentarité des images Pléiades et drone pour la viticulture de précision dans le cadre du programme EarthLab
GERMAIN, Christian
Ecole Nationale Supérieure des Sciences Agronomiques de Bordeaux-Aquitaine [Bordeaux Sciences Agro]
< Reduce
Ecole Nationale Supérieure des Sciences Agronomiques de Bordeaux-Aquitaine [Bordeaux Sciences Agro]
Language
en
Communication dans un congrès
This item was published in
Pléiades Days 2014, 2014-04-01, Toulouse. 2014
English Abstract
The potentiality of the Pléiades product has been assessed in the framework of the EarthLab programme set up by Telespazio France. « EarthLab Vine » aims at providing vineyard owners and wine sector with various products ...Read more >
The potentiality of the Pléiades product has been assessed in the framework of the EarthLab programme set up by Telespazio France. « EarthLab Vine » aims at providing vineyard owners and wine sector with various products and services dedicated to accurate and regular monitoring of vineyards in order to improve the technical management and increase the value of the production. Telespazio France has based the EarthLab products on the different very high resolution acquisition techniques available nowadays: satellite, aerial and UAV (Unmanned Aerial Vehicle). Thanks to its resolution in the panchromatic mode, the Pléiades sensor provides images at a resolution that is compatible with the resolution required for precision agriculture activity especially viticulture where the order of magnitude of the vine elements (rows) is about 1 metre. The multispectral information is brought through a pans harpening method so as to get the information at the highest possible resolution. This step is necessary to highlight the various elements of the target scene especially the rows and inter-rows that are not distinguishable on the native resolution images (2 metres). Then the pansharpened images are processed in order to generate various products especially a vegetation index map, vegetation index maps for the rows and for the inter-rows, as well as a map showing missing or dead vine stocks. This study has shown on some vineyard plots that results obtained from the Pléiades images are spatially consistent with the results obtained from very high resolution images acquired by an UAV. Consequently, the same products can be derived from both image sets: vigour maps, row detection, missing or dead stocks. However, small detail detection such as missing or dead stocks is more difficult on the Pléiades images than on the UAV images. On vineyards with narrow rows, Pléiades pixels are very often mixed pixels composed of inter-row vegetation (grass) and vine components, which highlights the limit of the sensor resolution compared to the UAV image resolution. Consequently, UAV imagery is still more adapted in some spatial configurations. This study has shown the potentiality of Pléiades for precision viticulture applications thanks to their high resolution compared to SPOT satellite family, as well as the synergy with very high resolution imagery (UAV or aerial acquisitions). Beyond the very high resolution capabilities of Pléiades, the spatial coverage as well as the short revisit time are of high interest compared to the constraints of the aerial techniques.Read less <
Origin
Hal imported