I am interested to visualize the result of Scene Classification (SC) that is automatically performed during the production of level-2A products. I am using GEE.

I have read this page and this page and I can see that there are 23 bands inside each sentine-2 level-2A product. I can see that half of these are "original" Bands of S2 and the rest are the "products" of S2 preprocessing algorithm.

What I can not figure out is how to combine the information of these separate layers/bands in order to visualize a classification map that represents the spatial distribution of the mentioned 12 groups (4 different clouds etc.)

In addition if there are information of min/max bounding values of each layer/band and its interpretation?

1 Answer 1


If I understand you correctly, you need to select the scl band, and then assign a palette to a visualization of a single band raster with values between 1 and 11.

These things look like .select('scl') either on an image or the image collection and visualization parameters will be {min:1, max:11, palette:['ff0004','868686','774b0a','10d22c','ffff52','0000ff','818181','c0c0c0','f1f1f1','bac5eb','52fff9']}

Remember that within EE you can type the name of the dataset into the search box at the top and quickly access the relevant collection/imagery details.

  • Thanks. You are right the SCL band contains integer values in the range [0,11] which represent classes. I was confused a bit by the information provided in this page where the intensity range of SCL band is said to be in the range [0,255]. docs.digitalearthafrica.org/en/latest/data_specs/… Commented Mar 1 at 11:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.