Context

MAJA (MACCS ATCOR Joint Algorthm) is a level 2A processing chain that:

detects clouds and their shadows,
estimates the optical thickness of aerosols, the amount of water vapour,
corrects atmospheric effects.

Historically the MACCS (Multi-sensor Atmospheric Correction and Cloud Screening) chain was developed jointly by CESBIO and CNES for Venus mission and extend progressively to other missions (SPOT-5, Formosat-2, Landsat 5, 7, 8 and Sentinel-2A/B). CESBIO design the algorithms and develop a prototype, while CNES took in charge the operational version of the chain, which CESBIO largely contributed to validate. More recently, CNES+CESBIO and DLR decided to pool their efforts to develop the MAJA chain (MACCS-ATCOR Joint Algorithm). This chain is an evolution of the MACCS chain in which methods from the DLR’s ATCOR chain will be progressively added. In order to celebrate this agreement, MACCS 6.0 version has been renamed to MAJA V1.0.

The MAJA chain (currently V3.X) is used operationally by the THEIA initiative in the CNES MUSCATE production centre.

MAJA 4.X will be relased in open source in 2020 with a new architecture which allow better integration with OTB and possibility to CESBIO and other researchers to reuse and update algorithms.

MAJA is now used on multiple programs, including Venus, Sentinel2, Landsat, Formosat-2, SPOT5.

CS promotes the MAJA chain as a product (cf. MAJA)

CS Group responsabilities for MACCS ATCOR Joint Algorithm are as follows:

Algorithms development and validation into MAJA or OTB based on CESBIO ATBB.
Develop new functionnalities or interfaces (new sensors or product format).
Improve processing performance.
Maintain and support integration of MAJA in CNES operational context

Main Picture

The features are as follows:

Detect clouds and clouds shadows
Detect snow and water bodies
Estimate atmospheric and aerosol parameters
Perform atmospheric correction on Earth Observation product
Perform envrionmental and topograpic correction Earth Observation product

Project implementation

The project objectives are as follows:

Develop and maintain an up to date and operational processing chain to generate L2A EO products

The processes for carrying out the project are:

Iterative development
Continuous development and integration (coding/unit testing)
Automatic Test Plan Generator with test execution and insertion of results in a CDash instance

Technical characteristics

The solution key points are as follows:

C++ software build on OTB libraries
Python
CMake/CTest/CDash
Multiple plugin handling using generic factories

Archi Picture

The main technologies used in this project are:

Domain	Technology(ies)
Operating System(s)	Linux 64 bits, RedHat/CentOS 7./8., Ubuntu 18.04
Programming language(s)	C++, Python
Interoperability (protocols, format, APIs)	SENTINEL SAFE, MUSCATE, LANDSAT product format
Production software (IDE, DEVOPS etc.)	GCC
Main COTS library(ies)	ORFEO Tool Box (OTB), Intersecteur3D, GDAL, HDF, OpenJPEG