LEPMON Project Funded by BMBF

We are happy to report that we received the notice of grants for our LEPMON project on LEPidoptera MONitoring with camera light traps and AI for fine-grained visual species identification. It will be funded by the German Federal Ministry of Education and Research (BMBF) for three years within the BiodivKI funding program, which is part of the funding initiatives FEdA and FONA.

LEPMON originates from the joint work across two different research groups at the University of Jena over the last years, namely the collaboration between Gunnar Brehm from the Phyletic Museum Jena (Faculty of Biological Sciences) and Paul Bodesheim from the Computer Vision Group (Computer Science Institute), who also share the leadership of the whole project. During the next three years, this project will significantly contribute to the analysis of biodiversity and species richness in Germany by monitoring hundreds to thousands of nocturnal insect species (mainly Lepidoptera) with high-quality camera light traps and AI tools for automated analysis of the collected image data. Since many of these insects are important pollinators for a wide variety of different crop plants, insights from our monitoring initiative might also affect agriculture and food production.

Within the Computer Vision Group, the acquired grants enable us to further strengthen our research and expertise in fine-grained recognition with AI for biological and ecological applications. Hence, we can continue our work on nocturnal insect monitoring that started with the AMMOD project and further develop computer vision and AI methods for large-scale automated species recognition by incorporating domain knowledge in machine learning models for fine-grained visual categorization. This project, led by Paul Bodesheim, will also be closely linked to our research on foundation models like large vision models (LVMs) and vision language models (VLMs).

Together with the project partners Gunnar Brehm from the Phyletic Museum Jena (responsible for development and deployment of the camera light traps), Peter Grobe from the Leibniz Institute for the Analysis of Biodiversity Change (LIB) in Bonn (responsible for data management and building an insect monitoring database), Julie Koch Sheard from the University of Marburg (responsible for recruiting and training of citizen scientists to support the collection and annotation of image data), and Roel van Klink from the German Centre for Integrative Biodiversity Research (iDiv) in Leipzig (responsible for the statistical analysis of the acquired ecological data with respect to urban gradients), we aim at a nationwide insect monitoring across Germany.

Furthermore, the LEPMON project is closely linked to the EU COST Action InsectAI to expand and improve automated insect monitoring with computer vision and AI technology in Europe.

We are looking forward to an exciting interdisciplinary project with a high impact for our society.