Abstract
A vast amount of Earth Observation data is produced daily and made available through online services and repositories. Contemporary and historical data can be retrieved and used to power existing applications, foster innovation, and improve EU citizens’ lives. However, an undersized audience follows this activity, leaving huge volumes of valuable information unexploited.
The EO4EU project aims to provide innovative tools, methodologies and approaches that would assist a wide spectrum of users, from domain experts and professionals to simple citizens to benefit from accessing EO data. It strives to deliver dynamic data mapping and labelling based on AI augmented modules, adding Fairness to the data and introducing an ecosystem for holistic management of EO data. EO4EU envisages to bridge the gap among domain experts and end users, while aims to bring in the foreground technological advances to address the market straightness towards a wider usage of EO data.
The project will support the wider exploitation of EO data by delivering: (i) Machine Learning (ML) methodologies for Semantic Annotation of existing and growing data sources, (ii) semantically enhanced knowledge graphs that will enable structuring of content around diverse topic areas and building step by step journeys from different sources into a unified approach, (iii) data fusion techniques to extend the scalability of existing distributed systems, (iv) Augmented and Virtual Reality for interactive user experience, and (v) advanced data analytics visualisations for improved learning and evidence-based interpretations of environmental observations. Its operational and technical capacity will be demonstrated within seven distinct pilots that cover different thematic areas, such as personalised health care, sea route planning, ocean monitoring, food security, food ecosystems, soil erosion, environmental pest, and crisis management. These thematic areas will engage a wide spectrum of involved stakeholders, from EO providers, policy makers and actors, researchers and academics to citizen scientists and the general public to join efforts and provide their multidisciplinary expertise to support the Commission’s strategic goals towards further exploitation of EO data.
The current document summarises a detailed work performed to identify the main driving lines on which the platform implementation work shall be organised around. Starting from the use cases to be implemented within the project, requirements in terms of data, processing tools, algorithms, results presentation and IPR have been collected and assessed. A detailed analysis on the gaps with respect to the state of the art has been performed and the results, per research area, have been aggregated based on the level of criticality.
The outcome showed that main criticalities remain on the access to relevant data and data sources, removing barriers to some datasets that prevent the development of services on them, and on the possibility to deploy a variety of computational resources close to the data. Besides traditional CPU- based processing technologies, GPU and HPC applications still remain not fully exploited and need some push also on the facilitation of the exploitation of these tools. The adoption of ML and AI tools is not yet widely diffused because these tools are not yet well known by the service developers than for technological gaps, thus there is a need of pushing awareness and knowledge of their functionalities and capabilities within the community.