| Full name: | Advanced Multimodal Sensing and Data Fusion for Early Digital Detection of Plant Stress Symptoms (EIG CONCERT-Japan) |
| Start date: | 2025. 08. 01. |
| End date: | 2028. 07. 31. |
| Participants: | Leibniz Universität Hannover, Germany
|
| Project homepage: | MULTIFUSE homepage |
| Coordinator: | Leibniz Universität Hannover |
Climate change is intensifying pressures on agriculture through extreme weather events as well as alterations in distribution and host-interactions of plant pathogens. At the same time, the transition to more sustainable agricultural practices requires the reduction of chemical pesticides and the use of fertilizers.
In order to meet these potentially conflicting requirements, advanced sensing systems are needed that complement the visual inspection of crops and thus allowing precise treatment of single plants by efficient phenotyping and closer monitoring. Ensuring food security and promoting sustainable agriculture necessitates precise early detection and differentiation of biotic and abiotic plant stress conditions to inform actionable recommendations in the framework of precision agriculture practices. Multi- and hyperspectral imaging currently represent the gold standard for non-contact plant monitoring. However, these methods lack the precision and reliability needed to consistently link disease symptoms to their causal stressors, thus, limiting their utility to support breeders and farmers in continuous crop monitoring.
To address this challenge, MULTIFUSE aims to integrate international expertise across plant sciences, optical sensor technology, and data processing to harness the potential of advanced multisensory systems for proximal plant status evaluation. The fusion of this sensor data will allow a digital system for stress detection and differentiation, which will benefit from the synergies of the individual methods.
This interdisciplinary approach seeks to enhance the accuracy digital phenotyping, allowing a precise linkage between detected plant stress symptoms and causal stressors. Both plant breeding and, in the longer term, plant monitoring will benefit from this progress, facilitating the advancement and adoption of digital phenotyping in agriculture.
In the project, HUN-REN SZTAKI provides and maintains the central digital research infrastructure, the research data repository and cloud services that are needed for the international data exchange among the project partners. The Department of Distributed Systems (HUN-REN SZTAKI DSD) is working on the research data management, metadata handling and prediction AI models. Data pre-processing and multi-level data fusion will take place in the shared “border area” between experimental lab research and data science.
