ExtremeEarth

The ExtremeEarth project develops extreme earth analytics techniques and technologies that scale to the petabytes of big Copernicus data, information and knowledge, and applies these technologies in two of the thematic exploitation platforms of the European Space Agency: the one dedicated to Food Security and the one dedicated to the Polar regions.

ExtremeEarth aims to close the gaps in research and innovation in the Copernicus context with significant impact for the European industry.

ExtremeEarth concentrates on developing techniques and software that will enable the extraction of information and knowledge from big Copernicus data using deep learning techniques and extreme geospatial analytics, and the development of two use cases based on this information and knowledge and other relevant non-EO data sets. ExtremeEarth will impact developments in the Integrated Ground Segment of Copernicus and the Sentinel Collaborative Ground Segment. ExtremeEarth tools and techniques can be used for extracting information and knowledge from big Copernicus data and making this information and knowledge available as linked data and, in this way, allow the easy development of applications by developers with minimal or no knowledge of EO techniques, file formats, data access protocols etc.

In the last few years, there have been four highly successful European research projects that have pursued this idea: the FP7 projects TELEIOS, LEO and Melodies, and the H2020 project Copernicus App Lab. These projects, under the lead of partner UoA, pioneered the use of linked geospatial data in the EO domain, and demonstrated the potential of linked data and semantic web technologies by developing prototype environmental and business applications. ExtremeEarth builds on the advances of these projects.

The data and information processed and disseminated puts Copernicus at the forefront of the Big Data paradigm, giving rise to all relevant challenges: volume, velocity, variety, veracity and value.

VOLUME

The repository of Sentinel products managed by the European Space Agency has so far published more than 11 million products, and it has more than 191 thousand users who have downloaded more than 106 PB of data.

VELOCITY

Copernicus data has to be delivered and processed in a short time frame. By the end of 2017, 10 TB of data were generated and 93 TB of data were disseminated every day from the Sentinel product repository.

VARIETY

The Sentinel satellites have different types of sensors and different levels of processing. Along with aerial imagery, in-situ data and other collateral information, this wealth of data is processed by EO actors to extract information and knowledge.

VERACITY

Decision-making and operations require reliable sources. Thus, assessing the quality of the data is important for the whole information extraction chain.

VALUE

The Copernicus Market report of 2016 estimates that the overall investment in Copernicus will reach EUR 7.4 billion in the years 2008-2020, while the cumulative economic value generated by it in the same period will be around EUR 13.5 billion, and it will support 28.030 job years in the EO sector.

Latest News

Project Summary

ExtremeEarth aims to close the gaps in research and innovation in the Copernicus context with significant impact for the European industry.

The deep neural network architectures classify effectively and efficiently multimedia images using extremely large benchmark datasets and utilizing the power of Big Data, Cloud and GPU technologies.

New tools have been developed in the projects TELEIOS, LEO, Melodies and Copernicus App Lab, for knowledge discovery and data mining from satellite images and related geospatial data sets, as well as tools for linked geospatial data integration, querying and analytics. However, none of these tools scales to the many petabytes of data, information and knowledge present in the Copernicus context.

Contrary to multimedia images, for which highly scalable Artificial Intelligence techniques based on deep neural network architectures have been developed by big North American companies such as Google and Facebook recently, similar architectures for satellite images, that can manage the extreme scale and characteristics of Copernicus data, do not exist in Europe or elsewhere today. Training datasets consisting of millions of data samples in the Copernicus context do not exist today and published deep learning architectures for Copernicus satellite images typically run using one GPU and do not take advantage of recent advances like distributed scale-out deep learning.

To make the Web of Data a reality, it is important to have this huge amount of data on the Web, available in a standard format, reachable and manageable by Semantic Web tools.

The technologies to be developed will extend the European Hops data platform of partners Logical Clocks and KTH to offer unprecendented scalability to extreme data volumes and scale-out distributed deep learning for Copernicus data. The extended Hops data platform will run on a DIAS selected after the project starts and will be available as open source to enable its adoption by the strong European Earth Observation downstream services industry.

ExtremeEarth is a H2020 proposal in the area of ICT addressing the call “ICT-12-2018-2020: Big Data technologies and extreme-scale analytics”. The main objective of ExtremeEarth is to develop extreme earth analytics techniques and technologies that scale to the petabytes of big Copernicus data, information and knowledge, and apply these technologies in two of the ESA TEPs: Food Security and Polar. The scientific and technical objectives of ExtremeEarth are the following:

  • To develop scalable deep learning and extreme earth analytics techniques for Copernicus big data.
  • To develop very large training datasets for deep learning architectures targeting the classification of Sentinel images.
  • To develop techniques and tools for linked geospatial data querying, federation and analytics that scale to big Copernicus data, information and knowledge.
  • To extend the capabilities for EO data discovery and access with semantic catalogue services that scale to the big data, information and knowledge of Copernicus.
  • To integrate the big data and extreme earth analytics technologies of all previous Objectives in the Hops data platform and deploy them in the selected DIAS and the two TEPs.