Multi-source satellite image analysis – Participation in a technological challenge

The outcome should contribute to:

• an enhanced exploitation of satellite images, by automating the workflow of imagery intelligence production and enabling analysts to focus on value-added tasks;
• the EU technological edge and autonomy for defence-related satellite imagery analysis.

Imagery analysis and in particular satellite image analysis is an important component of defence intelligence. It enables to gather strategic and operational information on facilities, vehicles and forces on the ground and on the seas with a good geographic coverage across the Earth, and to assess the status of suspicious or hostile activities.

Satellite imagery analysis has long relied on traditional techniques, but the ever-increasing amounts of available satellite images has led to a need for automated analysis technologies. These technologies have been the subject of much research over several decades, and the steady progress of artificial intelligence (AI) and in particular of AI-based image recognition has led to new tools appearing on the market. However, much of the development work and technological challenges in this field have focused on image types other than satellite images, and there is a need to foster the progress of satellite image analysis technologies. In this context, the general objective of this call topic is to enhance these technologies by giving research teams the opportunity to benefit from the organisation of a technological challenge dedicated to them.

Specific objective

Optical and radar images of a given geographic area contain complementary information. While these different types of images result from very different physical principles, which make their combination quite challenging, such a combination can lead to improved information extraction and capabilities. This call topic therefore aims at progressing information fusion approaches towards this objective.

The proposals must address technological solutions to detect, identify and characterise relevant information from multi-source satellite images, in particular by taking advantage of the complementarity of optical and radar images. These technological solutions must be evaluated in the framework of the technological challenge organised under this call. Technologies should be integrated into demonstrators that can be tested by representative defence users on their own data.

Types of activities

The following table lists the types of activities which are eligible for this topic, and whether they are mandatory or optional (see Article 10(3) EDF Regulation):

Accordingly, the proposals must cover at least the following tasks as part of mandatory activities:

• Generating knowledge:
  • Research on satellite image analysis technologies.
  • Participation to the evaluation campaigns organised in the framework of the technological challenge, including:
    • exchanges with other stakeholders on the evaluation plans;
    • participation to the test campaigns whereby the performances of the technological modules are measured on the test data provided by the challenge organisers;
    • participation to debriefing workshops.

In addition, the proposals should cover the following tasks:

• Integrating knowledge:
  • Integration of technological modules into demonstrators that can be tested by representative defence users.

The proposals should substantiate synergies and complementarities with foreseen, ongoing or completed activities in the field of satellite imagery analysis, notably those that may be performed in the context of the EDIDP (e.g. under the call topic EDIDP-ISR-PEO-2019), the EDF (e.g. under the call topics EDF-2021-DIGIT-R-FL on Frugal learning for rapid adaptation of AI systems, EDF-2022-RA-DIGIT-DBIR on Shared databases and integrated systems for image recognition, EDF-2022-DA-SPACE-ISR on Innovative multi-sensor space-based Earth observation capabilities towards persistent and reactive ISR, or EDF-2024-RA-DIGIT-ASMEP on Automated Structural Modelling for Effect Prediction), the EU Space programme (e.g. feasibility studies on potential EU Earth-observation services for governmental use), Horizon Europe or within the European Defence Agency (EDA) (e.g. the Capability Technology Group on Space activities).

The proposals should include clear descriptions of the proposed criteria to assess work package completion. Criteria should include the completion of the tests performed in each evaluation campaign.

Functional requirements

The proposed solutions should meet the following functional requirements:

• Technological modules addressing the tasks defined in the framework of the challenge should be developed and submitted for evaluation (cf. preliminary evaluation plan).
• These technological modules should be integrated into demonstrators with a user-friendly interface. Any difference between the version evaluated through the challenge and a version integrated in the demonstrator should be documented.
• The demonstrators should be able to run locally, without a connection to a wide area network, and with reasonable resources in term of hardware size, weight, price, and energy consumption.
• The technological modules should be easy to configure and integrate into defence systems beyond the demonstrators produced in the framework of the challenge. They should follow as much as possible the relevant standards, best practices and guidelines, including those elaborated at the challenge level, in particular for input and output formats.