Two vegetation indicators from 2D ground Lidar scanner compared for predicting spraying deposits on grapevine - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2017

Two vegetation indicators from 2D ground Lidar scanner compared for predicting spraying deposits on grapevine

Résumé

Terrestrial 2D Light Detection and Ranging (Lidar) has been shown to be a useful sensor for sensing vegetation in vineyards and orchards and has applications in managing crop protection. In this paper, two crop structure parameters are evaluated in vines with the objective to predict the proportion of sprayed phytosanitary products that is actually deposited in target vegetation. The Tree Area index (TAI) is based on the estimation of light beam interception’s probability within the plant cover. Leaf Wall Area by Point (LWApts) is derived from Pixelated Leaf Wall Area, a high resolution measurement applying the Leaf Wall Area concept. Repeatability at measuring these indicators has been assessed by analyzing their Relative Standard Deviations (RSD) for twenty cases with 5 repetitions in each. Spraying deposits on vineyard plots have been measured in order to study the relationships between deposits and the two crop indicators TAI and LWApts. Field trials were performed on five plots and at four times. Deposits were measured for two sprayers: a pneumatic arch and an air assisted face to face sprayer. For each time, plot and sprayer, Lidar measurements were performed on a 15m long section of two adjacent rows. The crop structure indicators were calculated from data acquired at frequency of 50Hz by the Lidar moved at 5km/h forward speed. Spray deposition was measured using a tracer (Tartrazine E102) sprayed on PVC collectors disposed according to a grid within 4 vigorous trees selected on the sections measured by Lidar. Both indicators were found to repeatable and useful to predict deposits. These promising results lead to the perspective of using Lidar systems to tune dose sprayed in the vineyards.
Fichier principal
Vignette du fichier
mo2017-pub00055510.pdf (1.11 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01735683 , version 1 (16-03-2018)

Identifiants

Citer

M. Bastianelli, V. de Rudnicki, Sébastien Codis, Xavier Ribeyrolles, O. Naud. Two vegetation indicators from 2D ground Lidar scanner compared for predicting spraying deposits on grapevine. EFITA 2017, Jul 2017, MONTPELLIER, France. 12 p. ⟨hal-01735683⟩
148 Consultations
169 Téléchargements

Partager

Gmail Facebook X LinkedIn More