Skip to content

Extracting cloudiness from Sentinel-2 data

Extraction of cloudiness from Sentinel-2 data

This step is optionnal if you are extracting data using Planetary, or don't wish to filter acquisitions on cloudiness. In this step, the percentage of pixels in the mask provided by the Sentinel-2 data provider is extracted for each acquisition. The results are exported in a csv file For THEIA, all pixels different to 0 in the CLM mask are considered cloudy For Scihub and PEPS, all pixels different to 4 or 5 in the SCL mask are considered cloudy

INPUT PARAMETERS

  • sentinel_dir (str) : Path of the directory containing Sentinel-2 data.
  • export_path (str) : Path to write csv file with extracted cloudiness.
  • tile_selection (list) : List of tiles from which to extract reflectance (ex : ["T31UFQ", "T31UGQ"]). If None, all tiles are extracted.
  • sentinel_source (str) : 'THEIA', 'Scihub' or 'PEPS' depending on the Sentinel-2 data provider.

OUPUT FILE

The results are exported in a csv file, with the columns :

  • area_name : the name of the Sentinel-2 tile
  • Date : the date of the Sentinel-2 acquisition
  • cloudiness : the percentage of cloudy pixels

Running this step

Using a script

from fordead.validation.extract_cloudiness import extract_cloudiness

extract_cloudiness(
    sentinel_dir = <sentinel_dir>, 
    export_path = <export_path>,
    tile_selection = <tile_selection>,
    sentinel_source = <sentinel_source>)

From the command invite

This step can also be ran from the command prompt. 

fordead extract_cloudiness [OPTIONS]

See detailed documentation for the command line prompt here