Researchers from Unmanned Valley and Greenport Duin- en Bollenstreek, together with NL Space Campus, have successfully developed an AI model that allows drones to detect botrytis, a plant disease. This model uses artificial intelligence (AI) to analyse large volumes of data collected by drones, which results in a highly detailed map that precisely identifies which plants are diseased or at risk, providing millimetre-level accuracy. This innovation could significantly reduce crop protection agents’ use. In November 2024, Unmanned Valley, Greenport Duin- en Bollenstreek, and NL Space Campus received a Computable Award in the Digital Transformation category for their collaborative, ambitious, and innovative project. The second part of the project, RSvS2, officially started Friday January 24th.
Preserving Biodiversity and Innovative Solutions for Growers
The Remote Sensing for Floriculture (RSF) project combines drones, satellite data, sensors, and artificial intelligence to detect plant diseases early, targeting issues like botrytis in tulips and hyacinths. With the help of advanced sensor technology, the project accurately identifies which plants need protection products and which do not. This approach leads to more efficient and sustainable use of resources, ultimately lowering costs and minimising environmental impact, an exciting development for growers, as it offers potential cost savings by reducing the risk of crop failures, decreasing the purchase of the amount of crop protection products, and eliminating the necessity for manual field inspections, which can’t cover every plant.
Excessive use of crop protection products threatens water quality and biodiversity. By detecting harmful conditions affecting crops early and accurately, crop protection products can be applied more precisely, significantly reducing their overall usage. The European Union plans to substantially decrease the use of these products in the upcoming years as part of the European Green Deal, which seeks to make Europe the world’s first climate-neutral continent by 2050.
Remote Sensing for Floriculture (RSF): The Second Phase
After reaching the initial milestones, the project is entering its second phase. Researchers will explore additional resources and enhance the capabilities of this type of precision agriculture. The following steps include improving the existing artificial intelligence and collecting new data in the spring and summer. The integration of weather data and satellite imagery will also be further evaluated, specifically assessing whether higher-resolution images captured from greater altitudes offer added value. Flying relatively inexpensive drones at greater heights and higher speeds can improve the efficiency of drone data collection. Additionally, exploring the use of drone boxes in the field may prove beneficial. These boxes can conduct pre-programmed automated flights, allowing for the collection of real-time data on demand and investigating the integration of value-added data from various sources, such as soil samples, camera images taken in and around fields, or mounted on tractors, as well as other robotics for data collection (e.g., H2L).
To clearly define points of interest, researchers will create task maps that focus on assessing specific areas in the field. While the initial model concentrated on detecting botrytis in tulips and hyacinths, the upcoming project studies will include tests for other diseases in ornamental crops, such as gladioli, daffodils, and alliums. These will serve as demonstrable use cases for the agricultural sector, exploring whether the AI model can be applied more broadly beyond ornamental floriculture.
The project will run until the end of the year and is supported by the NL Space Campus, Unmanned Valley, Greenport Duin-en Bollenstreek, Economic Board Duin- en Bollenstreek and Holland Rijnland.
