I. State of the art review:
The project started with an extensive investigation of the satellite data and sources available, with a focus on identifying suitable time frames aligned with the project's objectives. This led to the initial selection of the most promising Earth Observation (EO) parameters, taking into account both ease-of-use and their potential correlation with economic and epidemic data.
II. Analysis and selection of AI methods: We successfully developed and tested computer vision and image feature extraction algorithms tailored specifically for detecting airplanes, ships, containers, vehicles and umbrellas in satellite images with diverse resolutions.
III. Analysis and definition of economic key indicators and indexes: The general framework about econometric indicators was defined and developed within the declared methodological approach. Territorial/grey/statistical indicators were identified and selected to be correlated against EO parameters.
IV. Analysis and definition of epidemic key indicators and indexes: Epidemic key indicators and models were selected and justified. An epidemic analysis was carried out for the region of Macedonia in Greece on which the selected epidemic models were combined with weather information from satellite data.
V. Econometric model analysis and trade-off: The development and application and deployment of STeMA Territorial Impact Assessment (TIA) methodology and techniques were succesfully implemented.
Logical chain and GIS operational tool was deployed succesfully, including co-research and development of statistical and geographical appropriate indicators, for measuring and assessing various policies, local practices and impacts of epidemy at regional and local level. Two cases studies were carried out for Rome (Italy) and Wronclaw (Poland) in order to test the STeMA-TIA methodology. Moreover a selection of quali-quantitative statistical/geographical indicators, Building of composite indices, cooperation to Data collection at local level (per case) was carried out in order to build Case-studies to validate the developed tools and methodologies .
VI. Epidemic model analysis and implementation: The epidemic model was completed and the software was analysed and tested for later integration in the EYE platform.The model was validated through a use-case in Macedonia (Greece).
The EYE project main result includes the development of the EYE-SENSE platform through which local and public authorities will be able effortlessly gain access to Earth Observation (EO) parameters of a selected area related to atmospheric and water quality, night-light activity, land-surface temperature and object detection. Those EO parameters can be used to directly analyze a region’s socio-economic activity.
Other main results are:
1. Serverless Architecture Implementation: The platform adopts a serverless architecture, leading to potential cost savings and improved scalability.
2. Additional Modules: Development of a media sentiment analysis tool, epidemic models, and a territorial analysis model.
3. Datasets: Generation of valuable datasets, including earth observation data and in-situ economic data.
The exploitable assets of the EYE project can be categorized into four (4) groups as described below:
1. EYE Platform and modules: This includes any custom algorithms, AI/Computer vision models, Natural Language Processing (NLP) models, STEMA-TIA models, Epidemic models, data and processing pipelines, or analytical tools developed as part of the project.
2. Data Products and Datasets: The project generated valuable datasets, including earth observation data, and in-situ economic data. These datasets can be packaged and distributed as data products for research, planning, and decision-making purposes.
