On Friday 24 January, the sustainability project Remote Sensing for Floriculture was launched at Landgoed Tespelduyn in the Netherlands in the presence of all partners involved. This project focuses on detecting plant diseases and crop abnormalities using artificial intelligence (AI), drones, satellites and other sensors.
Last year, researchers from the Dune and Flowerbulb Region successfully developed an AI model that allows drones to detect diseases in plants. This yielded promising results and was awarded a Computable Award in the Digital Transformation category. With the launch of this long-awaited sequel, Unmanned Valley, Greenport Duin- en Bollenstreek, NL Space Campus, Economic Board Duin- en Bollenstreek, Holland Rijnland and other organizations are once again joining forces to further accelerate the sustainability of floriculture.
Sustainability through innovation
In this new project, the bar has been raised. Besides drones, researchers will now use satellite images, soil sensors and camera images to detect plant diseases. By combining different technologies, it is expected to accurately recognize which plants need protection and, more importantly, which plants do not. The ultimate goal is that growers can easily integrate this technology into their operations so that diseases are detected at an earlier stage and (crop) protection products can be used more efficiently, sustainably and with high precision. This could potentially save a lot of costs and reduce the risk of crop losses.
New drones
Current drones fly relatively slowly and low over plots to ensure the quality of the images collected. This is because these images need to be accurate down to the millimeter. The next phase of the project will investigate how to speed up this process so that larger areas can be mapped. By using new acquisition techniques and combining these results with other data, the researchers hope to collect data faster and more efficiently.
They are also looking at using so-called 'drone boxes'. These automated systems enable drones to perform planned flights without a pilot. The drones are on standby in a protected 'box/box' near the fields and can take off at any time for real-time data collection.
"The first phase has taken us further and provided more insights than we had dared to hope, thanks in part to collaboration with all stakeholders. I believe that technological know-how in floriculture, with its complex business and sustainability challenges, can contribute not only to solutions, but also to new business models for the region," said Theo de Vries, director of Unmanned Valley. "Winning a prestigious technology award underlines that our approach is successful and that our results stand out even beyond the floriculture sector."
Data from space
The quality of satellite imagery has improved significantly in recent years. This offers potentially large-scale insights that can play a meaningful role in this type of precision agriculture. Moreover, satellites provide crucial data on environmental factors such as weather conditions and soil moisture. Satellite data offers opportunities for growers to accelerate the practical application of the results of this project.
Expand
The new research will be significantly expanded from the first project. In addition to drone and space data, other methods that could be efficient for collecting relevant data for the AI are being investigated. These include cameras on tractors, agricultural robots, and ground sensors. The combination of this data should eventually even make it possible to make predictions about the emergence and spread of disease states.
Whereas the initial focus was on recognizing botrytis in tulips and hyacinths, the scope is now being broadened. The researchers are now also looking at which other crops and disease patterns the model can be made suitable for. This will make the technology widely applicable inside and outside the ornamental plant cultivation sector.
The agriculture of the future
The coming year is all about collecting new data. Researchers can therefore be found in the fields a lot as soon as spring arrives. The new data collected will be used to further improve the model. They are also working on translating it into practice so that growers can easily apply the models with their existing equipment. The RS4F team expects to present the results around December 2025.
For more information:
Unmanned Valley
www.unmannedvalley.nl