ee.Image.distance

指定された距離カーネルを使用して、各バンド内のゼロ以外の最も近いピクセルまでの距離を計算します。

用途戻り値
Image.distance(kernel, skipMasked)画像
引数タイプ詳細
これ: image画像入力画像。
kernelカーネル、デフォルト: null距離カーネル。chebyshev、euclidean、manhattan のいずれか。
skipMaskedブール値。デフォルト値は true です。対応する入力ピクセルがマスクされている場合は、出力ピクセルをマスクします。

例

コードエディタ（JavaScript）

// The objective is to determine the per-pixel distance to a target
// feature (pixel value). In this example, the target feature is water in a
// land cover map.

// Import a Dynamic World land cover image and subset the 'label' band.
var lcImg = ee.Image(
  'GOOGLE/DYNAMICWORLD/V1/20210726T171859_20210726T172345_T14TQS')
  .select('label');

// Create a binary image where the target feature is value 1, all else 0.
// In the Dynamic World map, water is represented as value 0, so we use the
// ee.Image.eq() relational operator to set it to 1.
var targetImg = lcImg.eq(0);

// Set a max distance from target pixels to consider in the analysis. Pixels
// with distance greater than this value from target pixels will be masked out.
// Here, we are using units of meters, but the distance kernels also accept
// units of pixels.
var maxDistM = 10000;  // 10 km

// Calculate distance to target pixels. Several distance kernels are provided.

// Euclidean distance.
var euclideanKernel = ee.Kernel.euclidean(maxDistM, 'meters');
var euclideanDist = targetImg.distance(euclideanKernel);
var vis = {min: 0, max: maxDistM};
Map.setCenter(-95.68, 46.46, 9);
Map.addLayer(euclideanDist, vis, 'Euclidean distance to target pixels');

// Manhattan distance.
var manhattanKernel = ee.Kernel.manhattan(maxDistM, 'meters');
var manhattanDist = targetImg.distance(manhattanKernel);
Map.addLayer(manhattanDist, vis, 'Manhattan distance to target pixels', false);

// Chebyshev distance.
var chebyshevKernel = ee.Kernel.chebyshev(maxDistM, 'meters');
var chebyshevDist = targetImg.distance(chebyshevKernel);
Map.addLayer(chebyshevDist, vis, 'Chebyshev distance to target pixels', false);

// Add the target layer to the map; water is blue, all else masked out.
Map.addLayer(targetImg.mask(targetImg), {palette: 'blue'}, 'Target pixels');

Python の設定

Python API とインタラクティブな開発での geemap の使用については、 Python 環境のページをご覧ください。

import ee
import geemap.core as geemap

Colab（Python）

# The objective is to determine the per-pixel distance to a target
# feature (pixel value). In this example, the target feature is water in a
# land cover map.

# Import a Dynamic World land cover image and subset the 'label' band.
lc_img = ee.Image(
    'GOOGLE/DYNAMICWORLD/V1/20210726T171859_20210726T172345_T14TQS'
).select('label')

# Create a binary image where the target feature is value 1, all else 0.
# In the Dynamic World map, water is represented as value 0, so we use the
# ee.Image.eq() relational operator to set it to 1.
target_img = lc_img.eq(0)

# Set a max distance from target pixels to consider in the analysis. Pixels
# with distance greater than this value from target pixels will be masked out.
# Here, we are using units of meters, but the distance kernels also accept
# units of pixels.
max_dist_m = 10000  # 10 km

# Calculate distance to target pixels. Several distance kernels are provided.

# Euclidean distance.
euclidean_kernel = ee.Kernel.euclidean(max_dist_m, 'meters')
euclidean_dist = target_img.distance(euclidean_kernel)
vis = {'min': 0, 'max': max_dist_m}
m = geemap.Map()
m.set_center(-95.68, 46.46, 9)
m.add_layer(euclidean_dist, vis, 'Euclidean distance to target pixels')

# Manhattan distance.
manhattan_kernel = ee.Kernel.manhattan(max_dist_m, 'meters')
manhattan_dist = target_img.distance(manhattan_kernel)
m.add_layer(
    manhattan_dist, vis, 'Manhattan distance to target pixels', False
)

# Chebyshev distance.
chebyshev_kernel = ee.Kernel.chebyshev(max_dist_m, 'meters')
chebyshev_dist = target_img.distance(chebyshev_kernel)
m.add_layer(
    chebyshev_dist, vis, 'Chebyshev distance to target pixels', False
)

# Add the target layer to the map water is blue, all else masked out.
m.add_layer(
    target_img.mask(target_img), {'palette': 'blue'}, 'Target pixels'
)
m