docs /
examples /
clouds.py
Cloud mask in the Sentinel-2 data¶
This is a short demo of the cloud mask in the sentinel-2 dataset.
For a more sophisticated method of dealing with clouds, see the Sentinel-2 Cloud Masking with s2cloudless example.
[1]:
import ee
import folium
import pysoilmap.ee as psee
[2]:
psee.initialize()
The full dataset and documentation is available here:
https://developers.google.com/earth-engine/datasets/catalog/COPERNICUS_S2
When using a dataset, always make sure to refer to the documentation for resolution, band names and definitions, etc.
For the purpose of this example, we’ll select a specific image taken on a specific flyover to keep the computation time short:
[3]:
img = ee.Image("COPERNICUS/S2/20180430T103019_20180430T103520_T32UMU")
rgb = img.select(["B2", "B3", "B4"]).divide(10_000)
Bits 10 and 11 are opaque and cirrus clouds, respectively:
[4]:
clouds = img.select('QA60').bitwiseAnd(0xc00)
opaque = clouds.bitwiseAnd(0x400)
cirrus = clouds.bitwiseAnd(0x800)
Try to detect cloud shadows by projecting the clouds on the surface (adapted from the 3_Sentinel2_CloudAndShadowMask notebook in AGREE):
[5]:
shad_o = psee.cast_shadows(img, opaque)
shad_c = psee.cast_shadows(img, cirrus)
These derived images can now be added in javascript using Map.addLayer() or in python using e.g. geemap.Map.addLayer().
For the purpose of this notebook, we’ll add it to a folium.Map using pysoilmap.ee.add_map_layer():
[6]:
folium.Map.addLayer = psee.add_map_layer
[7]:
mask = (lambda mask: mask.updateMask(mask.neq(0)))
vis = (lambda color: {'min': 0, 'max': 1, 'palette': color})
[8]:
Map = folium.Map(location=psee.center(img), zoom_start=9)
Map.addLayer(rgb, {'min': 0, 'max': 0.3}, "RGB")
Map.addLayer(mask(opaque), vis('yellow'), name="Opaque Clouds")
Map.addLayer(mask(cirrus), vis('red'), name="Cirrus Clouds")
Map.addLayer(mask(shad_o), vis('teal'), name="Opaque Shadows")
Map.addLayer(mask(shad_c), vis('pink'), name="Cirrus Shadows")
Map.add_child(folium.LayerControl())
Map
[8]: