Remotely sensed and ground measurament to detect and classify weed infestation of Ecballium elaterium (common name: squirting cucumber) on two dates and in two naturally infested hedgerow olive orchards under no tillage

Ecballium elaterium (common name: squirting cucumber) is an emerging weed problem in hedgerow or superintensive olive groves under no tillage. It colonizes the inter-row area infesting the natural or sown cover crops, and is considered a hard to control weed. Research in other woody crops has shown E. elaterium distribution in stands or patches and weed-free areas, which makes this weed susceptible to design a site-specific control strategy only addressed to E. elaterium patches. Therefore, our work adresses to develop a methodology based on the analysis of imagery acquired with an unmanned aerial vehicle (UAV) to detect and map E. elaterium infestations in hedgerow olive orchards for further design of site-specific control maps only for the E. elaterium infested areas. [Description of methods used for collection/generation of data] UAV flights and field work were conducted on 05/19/2021 and on 09/27/2021. Coordinates (XYZ) of the E. elaterium plants and patches were registered using a RTK GNSS. Aerial images were acquired with a quadcopter model Mavic Pro 2 (DJI, Shenzen, China) equipped with an RGB camera Hasselblad LID-20c with 20 Mp. The UAV was configured for flights at 50 m over the terrain following flight lines parallel to the olive trees lines. The aerial images had forward and side overlaps of 90% and 85%, respectively; UAV flights were carried around noon on sunny days with no wind. UAV images were processed with Agisoft Metashape Professional Edition software (Agisoft LLC, St. Petersburg, Russia) for the generation of the geomatic products used in the classification workflows: 1) the orthomosaic containing the spectral information, and 2) the Digital Surface Model (DSM) which provided height information. The orthomosaics and the DSMs from both fields and dates had a spatial resolution of 1.1 cm and 2.2 cm, respectively. The production of the geomatic products was automatic, with the exception of the manual measurement of 5 ground control points (GCPs) in the images (one in each corner plus one in the field center). GCP coordinates were measured in the field with a RTK GNSS receiver the day of flights.