ee.Image.reduceRegions

在指定集合中，針對每個地圖項目的區域套用縮減器。

Reducer 的輸入數量必須與輸入圖片的頻帶數量相同。

傳回輸入特徵，每個特徵都會附加對應的縮減器輸出內容。

用量	傳回
Image.reduceRegions(collection, reducer, scale, crs, crsTransform, tileScale, maxPixelsPerRegion)	FeatureCollection

引數	類型	詳細資料
this：image	圖片	要縮小的圖片。
collection	FeatureCollection	要縮減的功能。
reducer	縮減函式	要套用的縮減函式。
scale	浮點值，預設值：空值	要使用的投影方式的標稱公尺比例。
crs	投影，預設值：null	要使用的投影方式。如果未指定，系統會使用圖像第一個頻帶的投影。如果除了比例外，還指定了其他值，則會將圖片重新調整為指定的比例。
crsTransform	清單，預設值：null	CRS 轉換值清單。這是 3x2 轉換矩陣的列優先排序。這個選項與「scale」互斥，且會取代投影設定的任何轉換。
tileScale	浮點值，預設值：1	用於縮減匯總圖塊大小的縮放比例；使用較大的 tileScale (例如 2 或 4) 可能會讓計算機在預設情況下耗盡記憶體。
maxPixelsPerRegion	Long，預設值：null	每個區域可減少的像素數量上限。

範例

程式碼編輯器 (JavaScript)

// A Landsat 8 SR image with SWIR1, NIR, and green bands.
var img = ee.Image('LANDSAT/LC08/C02/T1_L2/LC08_044034_20210508')
              .select(['SR_B6', 'SR_B5', 'SR_B3']);

// Santa Cruz Mountains ecoregions feature collection.
var regionCol = ee.FeatureCollection('EPA/Ecoregions/2013/L4')
                    .filter('us_l4name == "Santa Cruz Mountains" || ' +
                            'us_l4name == "San Mateo Coastal Hills" || ' +
                            'us_l4name == "Leeward Hills"');

// Display layers on the map.
Map.setCenter(-122.08, 37.22, 9);
Map.addLayer(img, {min: 10000, max: 20000}, 'Landsat image');
Map.addLayer(regionCol, {color: 'white'}, 'Santa Cruz Mountains ecoregions');

// Calculate median band values within Santa Cruz Mountains ecoregions. It is
// good practice to explicitly define "scale" (or "crsTransform") and "crs"
// parameters of the analysis to avoid unexpected results from undesired
// defaults when e.g. reducing a composite image.
var stats = img.reduceRegions({
  collection: regionCol,
  reducer: ee.Reducer.median(),
  scale: 30,  // meters
  crs: 'EPSG:3310',  // California Albers projection
});

// The input feature collection is returned with new properties appended.
// The new properties are the outcome of the region reduction per image band,
// for each feature in the collection. Region reduction property names
// are the same as the input image band names.
print('Median band values, Santa Cruz Mountains ecoregions', stats);

// You can combine reducers to calculate e.g. mean and standard deviation
// simultaneously. The resulting property names are the concatenation of the
// band names and statistic names, separated by an underscore.
var reducer = ee.Reducer.mean().combine({
  reducer2: ee.Reducer.stdDev(),
  sharedInputs: true
});
var multiStats = img.reduceRegions({
  collection: regionCol,
  reducer: reducer,
  scale: 30,
  crs: 'EPSG:3310',
});
print('Mean & SD band values, Santa Cruz Mountains ecoregions', multiStats);

Python 設定

請參閱「 Python 環境」頁面，瞭解 Python API 和如何使用 geemap 進行互動式開發。

import ee
import geemap.core as geemap

Colab (Python)

# A Landsat 8 SR image with SWIR1, NIR, and green bands.
img = ee.Image('LANDSAT/LC08/C02/T1_L2/LC08_044034_20210508').select(
    ['SR_B6', 'SR_B5', 'SR_B3']
)

# Santa Cruz Mountains ecoregions feature collection.
region_col = ee.FeatureCollection('EPA/Ecoregions/2013/L4').filter(
    'us_l4name == "Santa Cruz Mountains" || '
    + 'us_l4name == "San Mateo Coastal Hills" || '
    + 'us_l4name == "Leeward Hills"'
)

# Display layers on the map.
m = geemap.Map()
m.set_center(-122.08, 37.22, 9)
m.add_layer(img, {'min': 10000, 'max': 20000}, 'Landsat image')
m.add_layer(
    region_col, {'color': 'white'}, 'Santa Cruz Mountains ecoregions'
)
display(m)

# Calculate median band values within Santa Cruz Mountains ecoregions. It is
# good practice to explicitly define "scale" (or "crsTransform") and "crs"
# parameters of the analysis to avoid unexpected results from undesired
# defaults when e.g. reducing a composite image.
stats = img.reduceRegions(
    collection=region_col,
    reducer=ee.Reducer.median(),
    scale=30,  # meters
    crs='EPSG:3310',  # California Albers projection
)

# The input feature collection is returned with new properties appended.
# The new properties are the outcome of the region reduction per image band,
# for each feature in the collection. Region reduction property names
# are the same as the input image band names.
display('Median band values, Santa Cruz Mountains ecoregions', stats)

# You can combine reducers to calculate e.g. mean and standard deviation
# simultaneously. The resulting property names are the concatenation of the
# band names and statistic names, separated by an underscore.
reducer = ee.Reducer.mean().combine(
    reducer2=ee.Reducer.stdDev(), sharedInputs=True
)
multi_stats = img.reduceRegions(
    collection=region_col,
    reducer=reducer,
    scale=30,
    crs='EPSG:3310',
)
display('Mean & SD band values, Santa Cruz Mountains ecoregions', multi_stats)