Atmospheric Correction Inter-comparison Exercise –
Cloud Masking Inter-comparison Exercise
1st workshop

Minimize Background

ACIX and CMIX are international collaborative initiatives to inter-compare a set of atmospheric correction (AC) and cloud masking (CM) algorithms for high-spatial resolution optical sensors. The exercises will focus on Sentinel-2 and Landsat-8 data over a set of test areas. The inter-comparison of the derived cloud masks is expected to contribute to a better understanding of strengths and weaknesses of different algorithms and ultimately to contribute to a reduction of a major error source for atmospheric correction and surface parameter retrievals. Similarly, the study of Bottom-of-Atmosphere (BOA) products will assist in understanding the different uncertainty contributors and improving the AC processors.


The first ACIX experiment started in June 2016 with the aim to bring together the developers of the state-of-the-art atmospheric correction (AC) processors and to study the variations amongst the different approaches. The input data were Landsat-8 and Sentinel-2A products over various sites of different land cover types around the world, i.e. agricultural, deserts, urban, snow and coastal areas. The description and the conclusions of this first experiment are summarised in Doxani et al. (2018). All the inter-comparison results can be found in the dedicated to ACIX I web page in CEOS Cal/Val portal. The improved versions of the participating processors and the increasing interest from AC developers to be part of the experiment stimulated the continuation of ACIX and its second implementation (ACIX II).

Following the recommendations of ACIX participants and other Earth Observation data users, an additional inter-comparison of cloud masking assessment was decided to be performed in parallel with ACIX. Cloud masking is a crucial step of the radiometric pre-processing of optical remotely sensed data and an important contributor to the retrieval of accurate surface reflectance within an atmospheric correction process. Therefore, it was considered essential to analyse these two processing chains together.

The test sites of the exercises will be redefined and more representative cases concerning land surface and atmospheric conditions, e.g. land/water, land cover, aerosols. Particular attention will be given also to aquatic sites, i.e. coastal and inland waters, which will be analysed as a separate sub-category. The scheme below describes the implementation of CMIX and ACIX II, which will run in parallel and follow the same timeline:



Minimize Objectives and Expected Outcomes

The exercises will bring together the developers of CM and AC algorithms and related software processors. The corresponding protocols will be defined together by coordinators and participants and it is mandatory to be followed by all the participants of the exercises. The input data will be Landsat-8, Sentinel-2A and -2B products of various test sites, covering various land cover types, i.e. agricultural, coastal, snow, water, deserts etc. The participants will generate the corresponding CM and BOA products which will be validated following the protocol.


Objectives of the 1st Workshop

To discuss and elaborate concepts, protocols and guidelines for the inter-comparison and validation of CM and BOA products.


Expected Outcomes of the 1st Workshop

  • Description of concept, protocols and procedures for inter-comparing and validating CM and BOA products
  • Definition of CMIX and ACIX II protocol
Minimize Programme
The programme of the 1st ACIX - CMIX workshop is available here
Minimize Schedule and Deadline

The exercises will be performed in three phases:

  1. Definition of the inter-comparison protocol: The coordinators will initially propose a protocol for the inter-comparison of CM and AC algorithms on which the participants are then invited to provide their feedback. All the proposals for the number and type of test sites, the number of scenes and the metrics for inter-comparison will be gathered before the 1st workshop of ACIX - CMIX. All the proposals will be discussed in the workshop and the final inter-comparison procedures shall be defined by all participants.
  2. Application of the AC processors: The participants will apply their processing schemes operationally on the test cases without tuning the processing parameters according to the scene features. The results will be submitted for analysis (3rd phase) to the coordinators.
  3. Analysis of the results: The coordinators will analyse the submitted results from all participants and assess the inter-comparison outputs based on the agreed metrics. All the results will be announced to participants prior to the 2nd ACIX II workshop, where they will be presented and discussed.

1st Workshop Announcement

15 June 2018

Distribution of the inter-comparison protocols (draft version)

30 July 2018

Deadline for ACIX participation

30 September 2018

Feedback on the AC inter-comparison protocol

30 September 2018

1st Workshop Registration Deadline and Preliminary programme

30 September 2018

1st Workshop of CEOS-WGCV ACIX-CMIX in Washington, D.C., USA

17-18 October 2018


Minimize Registration

Minimize Venue

Venue and accommodation information for the 1st workshop of ACIX-CMIX are available here

Minimize Organisation
Local Organiser    Organisation committee
Jean-Claude Roger, University of Maryland (USA) & NASA/GSFC (USA)
Eric Vermote, NASA/GSFC (USA)
Nima Pahlevan, NASA/GSFC (USA)
Sergii Skakun, University of Maryland (USA) & NASA/GSFC (USA)
Georgia Doxani, Serco and ESA/ESRIN (Italy)
Ferran Gascon, ESA/ESRIN (Italy)
Jean-Claude Roger, University of Maryland (USA) & NASA/GSFC (USA)
Eric Vermote, NASA/GSFC (USA)
Carsten Brockmann, Brockmann Consult GmbH (Germany)
Antoine Mangin, ACRI-ST (France)
Nima Pahlevan, NASA/GSFC (USA)
Sergii Skakun, University of Maryland (USA) & NASA/GSFC (USA)

Minimize Contact Information
     All Inquiries: