Project
Title
Entwicklung eines intelligenten UAV-gestützten Unkrautmonitoringsystems für den selektiven und teilflächenspezifischen Herbizideinsatz
Acronym
weed-AI-seek
Start
28.05.2021
End
27.05.2024
Coordinating Institute
Leibniz-Institut für Agrartechnik und Bioökonomie e.V. (ATB)
Coordinator
Partner
CiS-GmbH
Hochschule Harz - Hochschule für angewandte Wissenschaften
Project team ATB
Allocated to research program
Summary
The objective of the weed-AI-seek project is to develop an intelligent real-time monitoring and mapping system for the detection of weed distribution in cereal stands. For this purpose, high-resolution aerial image data is captured at low flight altitude and classified directly on the drone using optimised onboard AI image recognition during the overflight. The innovative method enables species-specific recognition at the level of individual plants. With the help of derived application maps, the application of herbicides is to be more precisely localised and selectively adapted to the real and species-specific weed distribution in cereal stands. In this way, the project makes a significant contribution to reducing the use of plant protection products in arable farming and thus promotes sustainable agriculture. The Leibniz Institute of Agricultural Engineering and Bioeconomy e.V. (ATB) is involved in all aspects of the project. (ATB) is involved in all project modules. A major contribution will be the development of an annotation database for training and testing AI models for species-specific recognition of weeds. This database contains both the image data extracted from the UAV images and metadata generated by experts. This comprehensive database will represent the diversity of plant species and background in wheat stands and thus go beyond the characterisation of leading weeds alone. The ATB is also significantly involved in the creation of the image recognition model for an embedded system.
Funding
Bundesministerium für Ernährung und Landwirtschaft (BMEL)

Funding agency
Bundesanstalt für Landwirtschaft und Ernährung - Projektträger

Grant agreement number
28DK105A20