Mask out clouds

Modified on Tue, 12 Dec 2023 at 01:17 PM

The Mask out clouds block removes images from the image collection that contain more than the specified percentage of clouds. 

You can change the maximum cloud cover percentage from the default of 40%.

This block is only used with optical datasets. It isn't needed for radar datasets that can 'see' through clouds.

Default method

Removes all image scenes with more than the specified percentage of cloud cover.

Custom method (Sentinel 2 Cloudless)

S2cloudless is a machine learning algorithm used to find clouds and create an image mask for the Sentinel-2 Multispectral dataset.

You can adjust the:

  • Cloud Probability Threshold: The maximum probability that a cloud exists in a pixel.  If the estimated probability of cloud is higher, the pixel is removed.  
  • Cloud Projection Distance: The maximum distance in kilometres from a cloud edge that the model looks for evidence of shadow.
  • NIR Dark Threshold: determines areas of cloud shadow. When the NIR band gets lower than this value the model assumes it is shadow.
  • Buffer: The distance in metres around the object where a cloud is identified and where pixels will be removed. 

Using a custom method will cause your workflow to run slower, as it requires more processing.

Learn more: 

Create a cloud-free image of your area of interest

Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select atleast one of the reasons

Feedback sent

We appreciate your effort and will try to fix the article