AIOPEN – An EO platform that integrates and combines AI/ML methods to support model development and exploitation of applications

Paper number

IAC-24,B5,1,3,x89164

Author

Mr. Leslie Gale, Space Applications Services, Belgium

Coauthor

Mr. Bernard Valentin, Space Applications Services, Belgium

Coauthor

Mr. Garin Smith, Telespazio UK, United Kingdom

Coauthor

Mr. Jonny Langstone, Telespazio UK, United Kingdom

Coauthor

Ms. Barbara Scarda, Serco Europe, Italy

Coauthor

Mr. Marco Pellengrini, SERCO S.p.A, Frascati, Rome, Italy, Italy

Coauthor

Mr. Francesco Ferrante, SERCO S.p.A, Frascati, Rome, Italy, Italy

Coauthor

Dr. Jakub Nalepa, KP Labs, Poland

Coauthor

Dr. Daniel Kostrzewa, KP Labs, Poland

Coauthor

Mr. Lukasz Tulczyjew, KP Labs, Poland

Coauthor

Mr. Mateusz Przeliorz, KP Labs, Poland

Coauthor

Mr. Jakub Sadel, KP Labs, Poland

Coauthor

Mr. Georg Zitzlsberger, Technical University of Ostrava, Czech Republic

Coauthor

Mr. Václav Svatoňe, Technical University of Ostrava, Czech Republic

Coauthor

Mr. Radek Halfar, Technical University of Ostrava, Czech Republic

Year

2024

Abstract

\documentclass[10pt]{article}
\usepackage[usenames]{color} %used for font color
\usepackage{hyperref}
\begin{document}
AI/ML is a transversal skill set now being applied to EO services and application development. Although AI developers and the EO science and service developer communities have become engaged in developing models support for creating integrated solutions was lacking. AIOPEN makes it possible to develop new and innovative applications and services applying established AI/ML software development capabilities. AIOPEN provides users with a solution that supports the end-to-end AI model development lifecycle.
 
This paper presents work performed in an ESA funded project to demonstrate how an existing EO data processing platform, EOPEN, can be extended to accommodate AI/ML developers. EOPEN was built using the Automated Service Builder (ASB) Framework developed by Space Applications Services. EOPEN is hosted in the ONDA Data and Information Access Service (DIAS).
 
AIOPEN represents a significant step towards harnessing the power of AI/ML technologies for the advancement of Earth Observation data analyses. To do so AIOPEN utilises cutting-edge technology and a cloud-native approach to address challenges in big data management, access, processing, and visualisation.
 
AIOPEN builds on existing technologies to extend the EOPEN platform combining technologies from the ASB (https://asb.spaceapplications.com), the ESA projects EOEPCA (\url{https://eoepca.org}) and AI4DTE (\url{https://eo4society.esa.int/projects/ai4dte-software-stack/}), OGC services and MLFlow (\url{https://mlflow.org}) Tracking Server.
 
AIOPEN is a robust platform providing collaboration services, including labelling campaign management, seamless model and data sharing, and efficient search across local and remote catalogues. It facilitates the hosting and sharing of models and training data, training of AI models, integration into new applications, and effective management and tracking of AI assets offering scientists and industry professionals public services capable of bringing together the processing and data access capabilities.
 
AIOPEN enables end-users (scientists and industry professionals) to leverage the vast amounts of EO data available and unlock valuable insights. Through community engagement activities AIOPEN fosters collaboration and gathers valuable feedback. 
 
To demonstrate and evaluate the capabilities of AIOPEN two showcases have been implemented: 1) Forest Cover Monitoring making use of a standard, well-established deep learning architectures like U-Nets for semantic image segmentation as the forest segmentation (in a single time point) is a binary semantic segmentation task.  2) Urban Change Detection using a Transformer Architecture EO data and Deep Neural Networks to detect urban related changes on the surface of the Earth to construct a digital twin of changes.

The paper will conclude with a discussion of the evaluation and future work.
\end{document}

Abstract document

IAC-24,B5,1,3,x89164.brief.pdf

Manuscript document

IAC-24,B5,1,3,x89164.pdf (🔒 authorized access only).

To get the manuscript, please contact IAF Secretariat.