Overview
QZ Solutions, a Polish technology company, is advancing precision agriculture with its SoilEO system, which uses hyperspectral imaging and AI algorithms to map soil nutrient content and optimise fertilisation strategies. To improve the accuracy of its models, QZ Solutions collaborated with agrifoodTEF Polish Node. The support included access to benchmark datasets and enriched annotations, enabling validation and optimisation of the SoilEO algorithms. This collaboration accelerates SoilEO’s market readiness and strengthens its potential impact on sustainable farming practices.
Arable farming
Food processing
Greenhouse
Horticulture
Livestock farming
Tree Crops
Viticulture
About the customer’s need
Modern farming requires precise soil management to increase yields while reducing fertiliser use and environmental impacts. Traditional soil testing methods are costly, time-consuming, and limited in scale. QZ Solutions aimed to provide farmers with accurate, AI-driven nutrient maps covering elements such as magnesium, potassium, phosphorus, and pH. However, their models required large-scale, high-quality, geolocated soil samples and consolidated datasets for validation and optimisation. To meet this challenge, QZ Solutions partnered with agrifoodTEF for support in benchmarking and data enrichment.
About the service
agrifoodTEF, through Lukasiewicz Poznanski Instytut Technologiczny (L-PIT) and Wielkopolska Agricultural Advisory Center in Poznań (WODR), provided two key services:
- Generation and provision of labeled datasets for benchmarking purposes (S00193): Over 2,000 geotagged soil samples were collected and analysed in the lab, creating a comprehensive benchmarking dataset with 5 single attributes and 20 double attributes covering nutrient levels and valuations.
- Curation and provision of annotated datasets (S00063): Soil samples were enhanced with geotagged photos, digital maps, and structured annotations, ensuring enriched datasets for algorithm validation.
The result was a robust, validated, and enriched dataset that significantly improved SoilEO’s AI models in detecting soil nutrients. Further collaboration with agrifoodTEF is ongoing to test SoilEO’s fertilisation map solutions in real-world conditions.
Discover the Curation and provision of annotated datasets service
Impact
The collaboration enabled QZ Solutions to improve the accuracy and reliability of SoilEO, providing farmers with more precise recommendations for fertilisation. Validated with high-quality datasets, the system is now ready for deployment, with at least 20 users expected by 2027. The enriched datasets reduced the time to deliver advisory services and enhanced the system’s ability to detect both macro- and microelements in diverse soil types.
By optimising fertiliser and water use, SoilEO supports sustainable agriculture, reduces environmental impacts, and contributes to rural development.
This success story showcases how AI and robotics, supported by agrifoodTEF, can benefit agri-businesses. Could your business be the next? Discover our services and join the technological revolution in agriculture.
