Customer: Centre National d’Etudes Spatiales (CNES)

Programme: CO3D

Supply Chain: CNES > CS Group SPACE

Context

CS Group responsabilities for 3D CO3D Image Processing Optical Constellation are as follows:

• Design, Development, Integration and Maintenance,

The features are as follows:

• Implementation of an operational treatment chain in micro-service, orchestrated in the cloud via Zeebe.
• 2D and 3D production (stereo 3D reconstruction)
• CODIP deployed in several operational centers, notably on the cloud.

Project implementation

The project objectives are as follows:

• Development for the CO3D constellation of the CODIP image processing chain: CO3D Image Processing
• Combined application of the Agile Scrum method and the DevOps approach

The processes for carrying out the project are:

• Agile scrum method
• Continuous deployment on the cloud

Technical characteristics

The solution key points are as follows:

• Algorithmic blocks of the chain cut into dockerized micro-service
• Python service layer
• Binding C ++ for Algo treatments
• Data exchange via S3 storage
• Services orchestration with Zeebe
• Strong scalability via the cloud deployment

The main technologies used in this project are:

Domain	Technology(ies)
Hardware environment(s)	Cloud (Flexible Engine)
Operating System(s)	GNU/Linux
Programming language(s)	GO, Python, C++
Interoperability (protocols, format, APIs)	S3, API REST, JSON
Production software (IDE, DEVOPS etc.)	Gitlab, Jenkins, SonarQube, Artifactory, Docker, Kubernetes, Zeebe, HashiCorp Vault, Vaultwarden
Main COTS library(ies)	CARS, GDAL, Taskflow, ALCIDE, MARIO, PESTO, boto3, pytest, ASTRIDZ, LIBGEO, DISIMUL, cookiecutter, pandas, pygdal, laspy, taskflow, boto3, testfixtures, jsonschema, numpy, rasterio, black, isort, pre-commit, pylint, pytest, clang, cython, cmake, scikit-image, scikit-learn, sphinx, wheel, proj, libgeotiff