ee.Image.updateMask

Aktualizuje maskę obrazu we wszystkich pozycjach, w których istniejąca maska nie ma wartości zero. Obraz wyjściowy zachowuje metadane i ślad obrazu wejściowego.

Wykorzystanie	Zwroty
Image.updateMask(mask)	Obraz

Argument	Typ	Szczegóły
to: image	Obraz	Obraz wejściowy.
mask	Obraz	Nowa maska obrazu jako liczba zmiennoprzecinkowa z zakresu [0, 1] (nieprawidłowa = 0, prawidłowa = 1). Jeśli ten obraz ma 1 pasmo, jest używany we wszystkich pasmach obrazu wejściowego. W przeciwnym razie musi mieć taką samą liczbę pasm jak obraz wejściowy.

Przykłady

Edytor kodu (JavaScript)

// A Sentinel-2 surface reflectance image.
var img = ee.Image('COPERNICUS/S2_SR/20210109T185751_20210109T185931_T10SEG');
var trueColorViz = {
  bands: ['B4', 'B3', 'B2'],
  min: 0,
  max: 2700,
  gamma: 1.3
};
print('Sentinel-2 image', img);
Map.setCenter(-122.36, 37.47, 10);
Map.addLayer(img, trueColorViz, 'Sentinel-2 image');

// Create a Boolean land mask from the SWIR1 band; water is value 0, land is 1.
var landMask = img.select('B11').gt(100);
print('Land mask', landMask);
Map.addLayer(landMask, {palette: ['blue', 'lightgreen']}, 'Land mask');

// Apply the single-band land mask to all image bands; pixel values equal to 0
// in the mask become invalid in the image.
var imgMasked = img.updateMask(landMask);
print('Image, land only', imgMasked);
Map.addLayer(imgMasked, trueColorViz, 'Image, land only');

// Masks are band-specific. Here, a multi-band mask image is used to update
// corresponding input image band masks.
var imgBandSubset = img.select(['B4', 'B3', 'B2']);
var bandSpecificMasks = imgBandSubset.gt(200);
var imgBandSubsetMasked = imgBandSubset.updateMask(bandSpecificMasks);
print('Multi-band mask image', bandSpecificMasks);
print('Image, variable band masks', imgBandSubsetMasked);
Map.addLayer(bandSpecificMasks, null, 'Multi-band mask image');
Map.addLayer(imgBandSubsetMasked, trueColorViz, 'Image, variable band masks');
// Note that there is only a single alpha channel for visualization, so when
// the ee.Image is rendered as an RGB image or map tiles, a masked pixel in any
// band will result in transparency for all bands.

// Floating point mask values between 0 and 1 will be used to define opacity
// in visualization via Map.addLayer and ee.Image.visualize.
var landMaskFloat = landMask.add(0.65);
var imgMaskedFloat = img.updateMask(landMaskFloat);
print('Image, partially transparent', imgMaskedFloat);
Map.addLayer(imgMaskedFloat, trueColorViz, 'Image, partially transparent');

Konfiguracja Pythona

Informacje o interfejsie Python API i używaniu geemap do interaktywnego programowania znajdziesz na stronie Środowisko Python.

import ee
import geemap.core as geemap

Colab (Python)

# A Sentinel-2 surface reflectance image.
img = ee.Image('COPERNICUS/S2_SR/20210109T185751_20210109T185931_T10SEG')
true_color_viz = {
    'bands': ['B4', 'B3', 'B2'],
    'min': 0,
    'max': 2700,
    'gamma': 1.3,
}
display('Sentinel-2 image', img)
m = geemap.Map()
m.set_center(-122.36, 37.47, 10)
m.add_layer(img, true_color_viz, 'Sentinel-2 image')

# Create a Boolean land mask from the SWIR1 band water is value 0, land is 1.
land_mask = img.select('B11').gt(100)
display('Land mask', land_mask)
m.add_layer(land_mask, {'palette': ['blue', 'lightgreen']}, 'Land mask')

# Apply the single-band land mask to all image bands pixel values equal to 0
# in the mask become invalid in the image.
img_masked = img.updateMask(land_mask)
display('Image, land only', img_masked)
m.add_layer(img_masked, true_color_viz, 'Image, land only')

# Masks are band-specific. Here, a multi-band mask image is used to update
# corresponding input image band masks.
img_band_subset = img.select(['B4', 'B3', 'B2'])
band_specific_masks = img_band_subset.gt(200)
img_band_subset_masked = img_band_subset.updateMask(band_specific_masks)
display('Multi-band mask image', band_specific_masks)
display('Image, variable band masks', img_band_subset_masked)
m.add_layer(band_specific_masks, None, 'Multi-band mask image')
m.add_layer(
    img_band_subset_masked, true_color_viz, 'Image, variable band masks'
)
# Note that there is only a single alpha channel for visualization, so when
# the ee.Image is rendered as an RGB image or map tiles, a masked pixel in any
# band will result in transparency for all bands.

# Floating point mask values between 0 and 1 will be used to define opacity
# in visualization via m.add_ee_layer and ee.Image.visualize.
land_mask_float = land_mask.add(0.65)
img_masked_float = img.updateMask(land_mask_float)
display('Image, partially transparent', img_masked_float)
m.add_layer(img_masked_float, true_color_viz, 'Image, partially transparent')
m