iMagine

iMagine provides a portfolio of free-at-the-point-of-use image datasets, high-performance image analysis tools empowered with Artificial Intelligence (AI), and Best Practice documents for scientific image analysis. These services and materials enable better and more efficient processing and analysis of imaging data in marine and freshwater research, accelerating our scientific insights about processes and measures relevant to healthy oceans, seas, and coastal and inland waters. By building on the computing platform of the European Open Science Cloud (EOSC) the project delivers a generic framework for AI model development, training, and deployment, which can be adopted by researchers for refining their AI-based applications for water pollution mitigation, biodiversity, and ecosystem studies, climate change analysis, and beach monitoring, but also for developing and optimizing other AI-based applications in this field. The iMagine compute layer consists of providers from the pan-European EGI federation infrastructure, collectively offering over 132,000 GPU hours, 6,000,000 CPU hours, and 1500 TB-month for image hosting and processing. The iMagine AI framework offers neural networks, parallel post-processing of very large data, and analysis of massive online data streams in distributed environments. 13 RIs will share over 9 million images and 8 AI-powered applications through the framework. Having representatives from so many RIs and IT experts, developing a portfolio of eye-catching image processing services together will also give rise to Best Practices. The synergies between aquatic use cases will lead to common solutions in data management, quality control, performance, integration, provenance, and FAIRness, contributing to harmonization across RIs and providing input for the iMagine Best Practice guidelines. The project results will be integrated into and will bring important contributions from RIs and e-infrastructures to EOSC and AI4EU.

EMSO ERIC is mainly involved in Marine Ecosystem Monitoring and is the leader of two tasks related to this domain and, therefore, responsible for the use case “Ecosystem monitoring at EMSO sites by video imagery”. The work that will be performed is focused on • Improving and simplifying data flows to databases by providing guidelines and standards for multidisciplinary data management, • Improving the pre-processing pipeline and methods for automatically processing video imagery for identifying interesting images for further analyses for purposes of ecosystem monitoring at EMSO sites, dealing with large volume and continuous video and further automation, to generate multidisciplinary databases.