AI-Supported Bee Monitoring for Sustainable and Biodiverse Agriculture
OCELI project completed
Research Focus: Applied Artificial Intelligence
Spectacular results from the OCELI project: AI, bees, and data analysis show new ways for biodiversity-friendly agriculture and provide impetus for research and practice.
The OCELI joint project (“Bee-based biomonitoring to explore the synergistic mechanisms of agriculture and pollinating insects”), funded by the German Federal Ministry of Agriculture, Food and Regional Identity, has developed an innovative monitoring system that uses honeybees and bumblebees as bioindicators. The aim is to better understand environmental influences and make agricultural practices more sustainable.
Automated bee monitoring with AI
The core of the now completed project is the technology developed by apic.ai GmbH. The company developed energy-autonomous, networked camera systems that continuously record the arrival and departure of pollinators. Artificial Intelligence analyzes pollen colors, recognizes markers on individual bees, and determines their behaviour. In collaboration with the FZI Research Center for Information Technology, powerful algorithms for color analysis and recognition were developed. The FZI also contributed to system integration and real-time evaluation in the field.
Practical field studies and risk assessment
The CRO Eurofins Agroscience Services conducted extensive field studies under real-world conditions. Among other things, visual AI-based camera systems were used to record the collection of pollen, the return behavior of tagged bees, and temporary inactivity after pesticide exposure. A particular focus was placed on measuring repellent effects and the behaviour of newly hatched bees. Marking and tracking enabled new insights to be gained into the foraging behaviour and the colonies’ response to stress.
Validation through BEEHAVE simulation model
The Helmholtz-Centre for Environmental Research (UFZ) validated the collected data using the BEEHAVE simulation model. Effects, such as those caused by pesticides, could be reproduced in the model and better understood. Of particular importance was the combination of activity and pollen information with simulation scenarios, enabling reliable conclusions to be drawn about colony development and food availability.
AI-supported analyses with disy Cadenza
Disy Informationssysteme GmbH integrated the OCELI data into its Cadenza analysis platform. Machine learning methods analyzed over 70 environmental and biodiversity characteristics, including temperature, humidity, land use, and CO₂ levels. Explainable AI — in particular the game theory approach SHapley Additive exPlanations (SHAP) — made it possible to illustrate the influence of individual factors on bee activity transparently. The analyses showed that time of day, temperature, season, and biodiversity significantly influence bee activity and explain up to 80 percent of the observed variability.
Practice relevance
OCELI closes an important gap between research and practice: the data obtained helps to make landscapes more pollinator-friendly in a targeted manner — a benefit for agriculture, politics, and biodiversity. Discussions with pesticide manufacturers and farmers have already revealed great interest in the findings.
About the FZI
The FZI Research Center for Information Technology, with headquarters in Karlsruhe and a branch office in Berlin, is a non-profit institution for information technology application research and transfer. It delivers the latest scientific findings in information technology to companies and public institutions and qualifies individuals for academic and business careers or the leap into self-employment. Supervised by professors from various faculties, the research groups at the FZI develop interdisciplinary concepts, software, hardware, and system solutions for their clients and implement the solutions found as prototypes. The FZI House of Living Labs provides a unique research environment for application research. The FZI is an innovation partner of the Karlsruhe Institute of Technology (KIT) and a strategic partner of the German Informatics Society (GI).