Researchers in the Duin– en Bollenstreek region have developed a smart method that can significantly reduce the use of crop protection agents.
Researchers from Unmanned Valley, Greenport Duin- en Bollenstreek, and NL Space Campus have succeeded in developing an AI model that can recognise botrytis using drones. The model uses artificial intelligence to analyse large amounts of drone data. The result is a clear map showing precisely which plants are diseased or at risk, down to the millimetre. This can significantly reduce the use of crop protection agents. The results of ‘Remote Sensing for Ornamental Horticulture’ were presented on 6 March at Unmanned Valley.
The European Union has set a goal to significantly reduce the use of crop protection agents in the coming years. This is part of the European Green Deal, through which the EU aims to make Europe the world’s first climate-neutral continent by 2050. Excessive use of crop protection agents poses risks to water quality and biodiversity. By detecting conditions that require crop protection early and accurately, these agents can be applied very precisely, reducing their use significantly.
Model Suitable for Botrytis Recognition
The model is suitable for recognising botrytis in tulips and hyacinths. It is expected that, with some relatively minor adjustments, other diseases can also be detected in other crops. This is a very interesting development for growers, not only environmentally but also cost saving. For example, because large quantities of crop protection agents do not need to be purchased, the risk of crop failure is reduced, and there is no need for people to go into the field to inspect every plant, which they can never do comprehensively. In addition, the researchers used an easily obtainable and relatively inexpensive drone. The drone can even carry out missions fully automatically, although currently, there still needs to be a pilot present who has visual contact with the drone.
The Project Continues
‘Remote Sensing for Ornamental Horticulture,’ as the project is called, will continue. Within the project, efforts are being made to improve the accuracy of measurements by combining drone data with satellite imagery and current data on soil and weather. Several large companies from the agricultural sector have expressed interest in investigating whether the technology can also recognize other diseases in other crops. Scalability and the development of the business model are also being examined.
This project was made possible in part by the municipality of Katwijk, Holland-Rijnland, and Drone Engineering & Operations students from the ROC van Amsterdam. More information about this project will be published on the website and social media of Unmanned Valley.
Featured image: Freepik
