Validation of Copernicus Sentinel-2 Cloud Masks Obtained from MAJA, Sen2Cor, and FMask Processors Using Reference Cloud Masks Generated with a Supervised Active Learning Procedure

Louis Baetens,Olivier Hagolle,Camille Desjardins

doi:10.3390/rs11040433

Abstract

The Sentinel-2 satellite mission, developed by the European Space Agency (ESA) for the Copernicus program of the European Union, provides repetitive multi-spectral observations of all Earth land surfaces at a high resolution. The Level 2A product is a basic product requested by many Sentinel-2 users: it provides surface reflectance after atmospheric correction, with a cloud and cloud shadow mask. The cloud/shadow mask is a key element to enable an automatic processing of Sentinel-2 data, and therefore, its performances must be accurately validated. To validate the Sentinel-2 operational Level 2A cloud mask, a software program named Active Learning Cloud Detection (ALCD) was developed, to produce reference cloud masks. Active learning methods allow reducing the number of necessary training samples by iteratively selecting them where the confidence of the classifier is low in the previous iterations. The ALCD method was designed to minimize human operator time thanks to a manually-supervised active learning method. The trained classifier uses a combination of spectral and multi-temporal information as input features and produces fully-classified images. The ALCD method was validated using visual criteria, consistency checks, and compared to another manually-generated cloud masks, with an overall accuracy above 98%. ALCD was used to create 32 reference cloud masks, on 10 different sites, with different seasons and cloud cover types. These masks were used to validate the cloud and shadow masks produced by three Sentinel-2 Level 2A processors: MAJA, used by the French Space Agency (CNES) to deliver Level 2A products, Sen2Cor, used by the European Space Agency (ESA), and FMask, used by the United States Geological Survey (USGS). The results show that MAJA and FMask perform similarly, with an overall accuracy around 90% (91% for MAJA, 90% for FMask), while Sen2Cor’s overall accuracy is 84%. The reference cloud masks, as well as the ALCD software used to generate them are made available to the Sentinel-2 user community.

Highlights

Thanks to their open access policy, their systematic and frequent revisit, and their data quality, the Landsat [1] and Copernicus Sentinel-2 [2] missions have revolutionized the optical Earth observation at a high resolution
We used three methods to validate the reference cloud masks provided by the Active Learning Cloud Detection (ALCD) method
There is for instance no cirrus class for FMask, or the absence of distinction between a medium probability cloud and a high probability cloud in MAJA and FMask, which is present in Sen2Cor

Summary

Introduction

Thanks to their open access policy, their systematic and frequent revisit, and their data quality, the Landsat [1] and Copernicus Sentinel-2 [2] missions have revolutionized the optical Earth observation at a high resolution. Before this open access era, most users only had access to a very limited number of images per year on their sites and used to process the data manually or at least in a very supervised manner. The detection of cloud shadows is complex, as a similar low reflectance range can be frequently observed on targets that are not obscured by clouds. In the case of semi-transparent clouds, shadow detection is even more challenging [5]

Methods

Results

Conclusion