Using 3D Point Clouds Derived from UAV RGB Imagery to Describe Vineyard 3D Macro-StructureReport as inadecuate




Using 3D Point Clouds Derived from UAV RGB Imagery to Describe Vineyard 3D Macro-Structure - Download this document for free, or read online. Document in PDF available to download.

* Corresponding author 1 EMMAH - Environnement Méditerranéen et Modélisation des Agro-Hydrosystèmes

Abstract : In the context of precision viticulture, remote sensing in the optical domain offers a potential way to map crop structure characteristics, such as vegetation cover fraction, row orientation or leaf area index, that are later used in decision support tools. A method based on the RGB color model imagery acquired with an unmanned aerial vehicle UAV is proposed to describe the vineyard 3D macro-structure. The dense point cloud is first extracted from the overlapping RGB images acquired over the vineyard using the Structure from Motion algorithm implemented in the Agisoft PhotoScan software. Then, the terrain altitude extracted from the dense point cloud is used to get the 2D distribution of height of the vineyard. By applying a threshold on the height, the rows are separated from the row spacing. Row height, width and spacing are then estimated as well as the vineyard cover fraction and the percentage of missing segments along the rows. Results are compared with ground measurements with root mean square error RMSE = 9.8 cm for row height, RMSE = 8.7 cm for row width and RMSE = 7 cm for row spacing. The row width, cover fraction, as well as the percentage of missing row segments, appear to be sensitive to the quality of the dense point cloud. Optimal flight configuration and camera setting are therefore mandatory to access these characteristics with a good accuracy.

Keywords : uav vineyards structure 3D point cloud row characteristics





Author: Marie Weiss - Frederic Baret -

Source: https://hal.archives-ouvertes.fr/



DOWNLOAD PDF




Related documents