A light-weight multi-spectral aerial imaging system for nitrogen crop monitoring

Lebourgeois Valentine, Bégué Agnès, Labbé Sylvain, Houles Marion, Martiné Jean-François. 2012. A light-weight multi-spectral aerial imaging system for nitrogen crop monitoring. Precision Agriculture, 13 (5) : 525-541.

https://doi.org/10.1007/s11119-012-9262-9

Article de revue ; Article de revue à facteur d'impact

Version publiée - Anglais
Accès réservé aux personnels Cirad
Utilisation soumise à autorisation de l'auteur ou du Cirad.
document_565361.pdf
Télécharger (810kB)

Quartile : Q1, Sujet : AGRICULTURE, MULTIDISCIPLINARY

Résumé : Image-based remote sensing is one promising technique for precision crop management. In this study, the use of an ultra light aircraft (ULA) equipped with broadband imaging sensors based on commercial digital cameras was investigated to characterize crop nitrogen status in cases of combined nitrogen and water stress. The acquisition system was composed of two Canon® EOS 400D digital cameras: an original RGB camera measuring luminance in the Red, Green and Blue spectral bands, and a modified camera equipped with an external band-pass filter measuring luminance in the near-infrared. A 5 month experiment was conducted on a sugarcane (Saccharum officinarum) trial consisting of three replicates. In each replicate, two sugarcane cultivars were grown with two levels of water input (rainfed/irrigated) and three levels of nitrogen (0, 65 and 130 kg/ha). Six ULA flights, coupled with ground crop measurements, took place during the experiment. For nitrogen status characterisation, three indices were tested from the closed canopy: the normalised difference vegetation index (NDVI), the green normalised difference vegetation index (GNDVI), and a broadband version of the simple ratio pigment index (hereafter referred to as the SRPIb), calculated from the ratio between blue and red bands of the digital camera. The indices were compared with two nitrogen crop variables: leaf nitrogen content (NL) and canopy nitrogen content (NC). SRPIb showed the best correlation (R 2 = 0.7) with NL, independently of the water and the N treatment. NDVI and GNDVI were best correlated with NC values with correlation coefficients of 0.7 and 0.64 respectively, but the regression coefficients were dependent on the water and N treatment. These results showed that SRPIb could characterise the nitrogen status of sugarcane crop, even in the case of combined stress, and that such acquisition systems are promising for crop nitrogen monitoring.

Mots-clés Agrovoc : Saccharum officinarum, azote, développement biologique, stress dû à la sécheresse, fertilisation, chlorophylle, imagerie, télédétection, Aéronef

Mots-clés géographiques Agrovoc : La Réunion, France

Classification Agris : U30 - Méthodes de recherche
F60 - Physiologie et biochimie végétale
H50 - Troubles divers des plantes
F01 - Culture des plantes

Champ stratégique Cirad : Axe 1 (2005-2013) - Intensification écologique

Auteurs et affiliations