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

被引:170
作者
Baetens, Louis [1 ]
Desjardins, Camille [2 ]
Hagolle, Olivier [1 ]
机构
[1] Univ Toulouse, CNRS, CESBIO, CNES,INRA,IRD,UPS, 18 Ave E Belin, F-31401 Toulouse 9, France
[2] Ctr Natl Etud Spatiales, 18 Ave E Belin, F-31401 Toulouse 9, France
关键词
Sentinel-2; cloud mask; cloud shadow; validation; active learning; MAJA; Sen2Cor; FMask; LANDSAT; SHADOW; COVER; WATER; IMAGERY; SCIENCE;
D O I
10.3390/rs11040433
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
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.
引用
收藏
页数:25
相关论文
共 37 条
[1]   Discriminating clear sky from clouds with MODIS [J].
Ackerman, SA ;
Strabala, KI ;
Menzel, WP ;
Frey, RA ;
Moeller, CC ;
Gumley, LE .
JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES, 1998, 103 (D24) :32141-32157
[2]  
[Anonymous], 2017, P 2017 9 INT WORKSHO, DOI DOI 10.1109/MULTI-TEMP.2017.8035219
[3]  
Baetens L., ACTIVE LEARNING CLOU
[4]   A PRECAUTION REGARDING CIRRUS CLOUD DETECTION FROM AIRBORNE IMAGING SPECTROMETER DATA USING THE 1.38-MU-M WATER-VAPOR BAND [J].
BENDOR, E .
REMOTE SENSING OF ENVIRONMENT, 1994, 50 (03) :346-350
[5]   Random forests [J].
Breiman, L .
MACHINE LEARNING, 2001, 45 (01) :5-32
[6]  
Bréon FM, 1999, J APPL METEOROL, V38, P777, DOI 10.1175/1520-0450(1999)038<0777:CDFTSP>2.0.CO
[7]  
2
[8]   SPECTRAL SIGNATURE OF ALPINE SNOW COVER FROM THE LANDSAT THEMATIC MAPPER [J].
DOZIER, J .
REMOTE SENSING OF ENVIRONMENT, 1989, 28 :9-&
[9]   Sentinel-2: ESA's Optical High-Resolution Mission for GMES Operational Services [J].
Drusch, M. ;
Del Bello, U. ;
Carlier, S. ;
Colin, O. ;
Fernandez, V. ;
Gascon, F. ;
Hoersch, B. ;
Isola, C. ;
Laberinti, P. ;
Martimort, P. ;
Meygret, A. ;
Spoto, F. ;
Sy, O. ;
Marchese, F. ;
Bargellini, P. .
REMOTE SENSING OF ENVIRONMENT, 2012, 120 :25-36
[10]   Cloud detection algorithm comparison and validation for operational Landsat data products [J].
Foga, Steve ;
Scaramuzza, Pat L. ;
Guo, Song ;
Zhu, Zhe ;
Dilley, Ronald D., Jr. ;
Beckmann, Tim ;
Schmidt, Gail L. ;
Dwyer, John L. ;
Hughes, M. Joseph ;
Laue, Brady .
REMOTE SENSING OF ENVIRONMENT, 2017, 194 :379-390