We use cookies on our website. Some are necessary for the operation of the website. You can also allow cookies for statistical purposes. You can adjust the data protection settings or agree to all cookies directly.
Development and practical test of a UAV system with image processing algorithms as a basis for site specific herbicide application
Project
Project code: JKI-A-08-1222
Contract period: 01.01.2016
- 31.12.2018
Purpose of research: Applied research
For the realisation of a site specific herbicide management under the acronym KOPTHER a method of species-selective weed recognition will be developed. For this purpose an autonom flying platform (Unmanned Aerial Vehicle - UAV) with an autopilot and new methods for the photogrammetric estimation of single weed plants will be used. KOPTHER`s aim is to establish an autonomous system in agricultural practice to map the spatial distribution of weed species in cropland. KOPTHER make available information for application maps which are necessary for site specific herbicide application. Automatic weed recognition should contribute to reduce pesticides per unit of area and to optimize operations. This enables savings on operating resources, shorter working hours and environmental protection. Creating application maps requires so far manuell weed counting. This is very time and cost consuming and can therefore only be used for small sample size, but not for agricultural practice. Estimates can be made by geostatistical methods for areas not covered by sampling. A high density of data points, estimated by UAV can help to improve accuracy of weed estimation. A small unmanned aircraft vehicle with a modified RGB camera was tested taking images from agricultural fields. A microcopter with six rotors was applied. The hexacopter in particular is GPS controlled and operates within predefined areas at given altitudes. Different scenarios of photogrammetrically weed detection have been carried out regarding to variable altitude, image resolution, weed and crop growth stages. First experiences with UAVs showed a high potential for site-specific weed control. Images analyses with regards to a recognition of weed patches can be used to adapt herbicide applications to varying weed occurrence across a field.
Section overview
Subjects
- Crop Protection
- Agricultural Engineering Plant Production
Framework programme
Funding programme
Excutive institution
Institute for Plant Protection in Field Crops and Grassland (JKI-A)