Catalogue of Services

We provide a next-generation 3D plant analysis service designed to accelerate testing by turning raw plant scans into reliable, quantitative insights. Instead of relying on slow and error-prone manual measurements, our system uses advanced algorithms to automatically extract precise plant traits—helping you compare genotypes faster and more consistently. These scans are very useful to test your AI in plant segmentation. From a single scan—whether collected with research platforms like NPEC’s MaxiMarvin (see service NPEC S00134) or newer approaches such as Gaussian splatting—we deliver a fully segmented 3D model that distinguishes each component: main stem, side branches, leaves, pot, or support pole. A detailed skeleton of the plant structure is then generated, allowing automatic calculation of key parameters such as internode length, branching angle, and phyllotactic angle. This provides a standardised dataset to test and further improve your detection and segmentation AI. Beyond trait measurement, our skeleton-based models can also be used for synthetic data generation—modifying features such as leaf positions or stem lengths to expand AI training datasets or simulate growth patterns across different varieties for virtual robot testing.

Data analysis

Data augmentation

Provision of a phenotyping greenhouse with extensive sensor set - NPEC

NPEC will enable you to make accurate, high-throughput studies of plant performance possible in relation to relevant biotic (diseases, bacteria) and abiotic (light quantity and quality, nutrients, temperature, moisture, soil pH and atmospheric CO2 level) factors across a range of scales.

The experiments can be performed in a greenhouse compartment with an extensive set of sensors to measure the plant and it’s environment. In growth rooms smaller scale experiments can be performed, with a special chamber for studying the roots of plants using transparent growth media. In the open field crop measurements can be performed using the Traitseeker for RGB, spectral and 3D measurements.

AI model training

Data analysis

Provision of a virtual playground for robots

The service consists of a photorealistic digital greenhouse environment, built in a game engine.

The greenhouse is configurable to different sizes, light conditions (sun, supplementary lighting) or row spacing. In the greenhouse different crops can be placed to either test AI detection networks or complete robotic systems, as there is a ROS2 integration available. The environment is made in a way that physical properties of the objects have influence on other components or robot interactions.

AI model training

Data augmentation

Provision of datasets

Josephinum Research (JR)

Our service offers a comprehensive dataset tailored for training or testing your solutions in the technology and agricultural sectors. Whether you're looking to enhance existing models or build new ones, our versatile data can be adapted to fit your specific needs. For those requiring an even more robust solution, our service seamlessly integrates with our data augmentation and labelling service (S00172), ensuring that you have access to well-organized and precisely labelled data. This flexibility allows us to support a variety of sensors and use cases, utilizing both existing data records and capturing new data in your application area. Optimize your agricultural technology solutions with our data services, designed to propel your projects forward efficiently and effectively.

Data augmentation

Provision of datasets for simulation and visualisation

In order to develop digital twins for agriculture, reliable long-term datasets on the agronomy and ecology of fields are crucial. From several long-term experiments and agricultural systems (like the Farm of the Future), datasets of fields are available with registrations of crop operations, crop measurements (e.g., timeseries of vegetation indices), yield, soil, & weather data, and data on key performance indicators. In some cases, other sensor data (tractor data, soil moisture, insect monitoring) are also available. New datasets can be developed based on specific needs.

Provision of datasets for validation of AI algorithms for livestock farming

Thanks to 30 years of data gathering at Agro-innovation Centre De Marke, multiple data sets are available. During these years, data is gathered on individual cow events, barn emissions, yields of specific and typical livestock farm crops, soil samples were taken, and feeding strategies were developed. Due to the systematic approach of De Marke, this data can be easily reused for setting up specific validation of AI algorithms. Additionally, new data can be gathered as requested.

Provision of raw image datasets in field crops

In order to develop AI applications or vision functionalities in robots, one of the first steps is to have high-quality images of crops, weeds, nutrient deficiencies, diseases, or pests available. As the phenotype of plants can really differ between regions and seasons, providing balanced datasets with high-quality images and different illumination could save valuable (development and acquisition) time. The data can be used for annotation purposes or to have a first visual check if an existing trained algorithm works sufficiently.

Scanning as a Service for synthetic data generation and modelling

We offer a comprehensive 3D scanning and model preparation service tailored for agriculture and food applications, built on expertise developed through scanning diverse objects such as rapeseed plants, tomato seedlings, and fish. This service transforms real-world items into accurate, high-resolution 3D models, ready for use in synthetic data generation, robotics simulations, and plant phenotyping. Using tested scanning approaches—including photogrammetry with stationary or rotating cameras, DSLR, and smartphone captures—we can reliably acquire detailed image sets. Using these 3D models, your AI detection algorithms can be tested on high-quality 3D data. The data can be used to test detection algorithms, or it can be used in complete virtual environments to test robotic applications. Beyond static models, dynamic functionality can be added by assigning physics properties and enabling randomisation—altering geometry, texture, and patterns to expand dataset diversity. This allows models to be seamlessly integrated into simulation environments for testing AI, testing robotic systems, or conducting virtual experiments to test and improve the systems. Whether for research, product development, or automation, our solution delivers a scalable, future-ready platform for realistic and adaptable 3D tests.

Test and demonstration of (robotic sensor, implement, algorithm) solutions with end-users

Accessing the end-user in a country takes a lot of time and effort. Wageningen Research - Field Crops has, through its regional experimental farms, access to farmer networks and dealers and can bring you in contact with them. The researchers in our department have a practical approach and know what is needed to implement new technologies in farms, therefore fulfilling a bridge between agronomy, technology, and ecology. By testing and demonstrating your product or service to our farm managers and farmers in the region (for example, farmers in https://www.proeftuinprecisielandbouw.nl), we gather valuable feedback to speed up the development of your prototype.

Market research

Test and experimentation of robotic and AI solutions in field crops

The core business of Wageningen Research - Field Crops is to design and create experimental plots (e.g. at the Farm of the Future) and to collect objective data on crops, weeds, diseases, pests and quality and effectiveness of robotic and AI solutions, in which we provide statistical reliable datasets. Examples of experiments are e.g. testing the accuracy and quality of handsfree harvesting solutions, weeding robots and path following of autonomous implement carriers in field crops. We offer for example the following measurements : -Provide measurements on effectiveness of solutions on control of weeds, pests and diseases (e.g. effectiveness of a weed robot) -Provide timeseries of vegetation indices (drone) -Provide yield measurements (quantity and quality) -Provide plant measurements (emergence, flowering, senescence) -Provide weed countings (including species determination) -Provide sensor data (soil moisture, insect monitoring, etc.) Furthermore we have common agricultural machines and implements available. We can cover different field crops, but we also are specialized in new crop farming systems that are based on agro-ecological principles. Think of strip and mosaic farming. Get in touch to discuss the possibilities for your experiment or test question.

Test, validation, and data collection of robotic and AI systems and components for livestock solutions

Location

Netherlands

Livestock farming

In the setting of the Farms of the Future (De Marke), we can (physically) test and validate robotic and AI solutions that are dedicated to dairy solutions. In this context we are also able to collect appropriate data sets of these tests and validation experiments. Test and validation data will be compared to ground truth data sets.

Desk assessment

Testing of AI-based sensor performance

Josephinum Research (JR)

Our service thoroughly evaluates AI-based sensors by measuring key metrics like accuracy, mean squared error, and other tailored parameters. Testing is conducted in real-world environments where the sensor will be used, with reference data recorded or labelled for analysis. For certain sensors, we also utilise benchmark datasets to validate performance. We assess consistency under identical conditions, adaptability to environmental changes (e.g., light, temperature, or weather), and improvements in accuracy over time. Additionally, we measure power consumption, processing efficiency, and memory usage to ensure optimal resource utilisation. Response time between detection and action is evaluated for real-time applications, while extreme condition testing (e.g., bright light, rainy conditions, compacted soil) ensures robustness in challenging environments. We also verify data security and resilience against adversarial attacks. This comprehensive testing is essential, ensuring sensors meet high standards of precision, reliability, and efficiency. Benchmark datasets, standardised data collections used for validation, provide a reliable baseline for performance comparison. By addressing these aspects, our service ensures your sensor technology is ready for real-world deployment, delivering the performance and adaptability your application demands.