GRSG Conference 2022: Orbit to Outcrop

Poster Title: CALLISTO

Author: Georgia Karadimou

Abstract:

CALLISTO is an EU-funded project that started in January 2021, will last for 3 years, and has the ambition to bridge the gap between Copernicus Data and Information Access Services (DIAS) providers and application end users through dedicated Artificial Intelligence solutions. It will provide an interoperable Big Data platform integrating Earth Observation data with crowdsourced and georeferenced data and observations from Unmanned Aerial Vehicles, for several applications, based on the end users’ needs.

CALLISTO combines Earth Observation data from ONDA DIAS with data from heterogeneous distributed sources (including crowdsourced data, videos from Unmanned Aerial Vehicles and data from in-situ sensors) through machine learning and data fusion technologies.

Currently, the services of CALLISTO are planned to be tested in four real-life cases that are driven by the needs of the:

  • Virtual monitoring of the implementation of the Common Agricultural Policy (CAP) obligations,
  • Virtual presence in water resources for water quality assessment using EO and in-situ data,
  • Improvement of the existing approaches and the increase of the value of the current Border Surveillance Services,
  • Expansion of the use of satellite imagery and data from sensors for journalistic research and verification.

Some of the technologies and methodologies used in CALLISTO, are well known and are already or can be used as well for geological applications, (e.g., geohazards monitoring):

  • Artificial Intelligence: Machine Learning and Deep Learning techniques
  • Distributed computing: High Performance Computing for the application of AI-based techniques for modelling performance, neural network-based methods for detecting & predicting behaviour patterns
  • Unmanned Aerial Vehicles: Computer vision using UAVs and alert-driven UAVs path planning
  • Data fusion using satellite data, Galileo/GNSS data, Web and social media data, and in-situ hyperspectral measurements
  • 3D models: Reconstruction of models using satellite data
  • Semantic Image Segmentation & Ontologies: Representation of interlinked data in machine-readable data resources in graph structure
  • Named Entity Recognition: AI-based extraction of knowledge from unstructured texts

Within CALLISTO, the aim is the data fusion of hyperspectral satellite imagery with in situ measurements, field and social media data. A potential further use and development of it, can be regarding geological and natural hazards applications. The outcomes are semantically-enriched and served to the public in interactive interfaces, mobile and Mixed Reality apps, creating a novel and innovative immersive environment for the Copernicus market.