ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume IV-2/W5
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., IV-2/W5, 179–185, 2019
https://doi.org/10.5194/isprs-annals-IV-2-W5-179-2019
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., IV-2/W5, 179–185, 2019
https://doi.org/10.5194/isprs-annals-IV-2-W5-179-2019

  29 May 2019

29 May 2019

DETECTING RUMEX OBTUSIFOLIUS WEED PLANTS IN GRASSLANDS FROM UAV RGB IMAGERY USING DEEP LEARNING

J. Valente1,2, M. Doldersum1, C. Roers3, and L. Kooistra1 J. Valente et al.
  • 1Laboratory of Geo-Information Science and Remote Sensing, Wageningen University & Research, 6708 PB, The Netherlands
  • 2Information Technology Group, Wageningen University & Research, Wageningen 6700 EW, The Netherlands
  • 3Naturschutzzentrum im Kreis Kleve e.V., 46459 Rees-Bienen, Germany

Keywords: DJI Phantom, Grasslands, Machine vision, Deep learning, Rumex, Weeding, Plant detection, Aerial surveying

Abstract. Broad-leaved dock (Rumex obtusifolius) is a fast growing and spreading weed and is one of the most common weeds in production grasslands in the Netherlands. The heavy occurrence, fast growth and negative environmental-agricultural impact makes Rumex a species important to control. Current control is done directly in the field by mechanical or chemical actuation methods as soon as the plants are found in situ by the farmer. In nature conservation areas control is much more difficult because spraying is not allowed. This reduces the amount of grass and its quality. Rumex could be rapidly detected using high-resolution RGB images obtained from a UAV and optimize the plant control practices in wide nature conservation areas. In this paper, a novel approach for Rumex detection from orthomosaics obtained using a commercial available quadrotor (DJI Phantom 3 PRO) is proposed. The results obtained shown that Rumex can be detected up to 90% from a 6 mm/pixel ortho-mosaic generated from an aerial survey and using deep learning.