Customer: Centre National d’Etudes Spatiales (CNES)

Programme: CO3D

Supply Chain: CNES > CS Group SPACE

Context

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

• Design, Development, Integration and Maintenance,

The features are as follows:

• Implementation of an operational processing chain in micro-services, orchestrated in the cloud via Zeebe.
• 2D and 3D production (stereo 3D reconstruction)
• CODIP deployed in several operational centers, in particular 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 Cloud Deployment

Technical characteristics

The solution key points are as follows:

• Algorithmic blocks of the chain cut into dockerized micro-services
• Service layer in Python
• Binding C++ for algo processing binaries
• Data exchange via S3 storage
• Service orchestration with Zeebe
• Strong scalability via 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