ee.Filter.maxDifference

建立一元或二元篩選器，如果左右運算元 (皆為數字) 之間的差異在指定上限內，就會通過篩選。如果做為聯結條件使用，這個數值差異會做為聯結指標。

用量	傳回
ee.Filter.maxDifference(difference, leftField, rightValue, rightField, leftValue)	篩選器

引數	類型	詳細資料
difference	浮點值	篩選器傳回 true 的最大差異。
leftField	字串，預設值為空值	左運算元的選取器。如果指定了 leftValue，則不應指定。
rightValue	物件，預設值：null	右側運算元的值。如果指定了 rightField，則不應指定這個欄位。
rightField	字串，預設值為空值	右運算元的選取器。如果指定了 rightValue，則不應指定這個值。
leftValue	物件，預設值：null	左側運算元的值。如果指定了 leftField，則不應指定此屬性。

範例

程式碼編輯器 (JavaScript)

// Field site vegetation characteristics from projects in western USA.
var fc = ee.FeatureCollection('BLM/AIM/v1/TerrADat/TerrestrialAIM')
  .filter('ProjectName == "Colorado NWDO Kremmling FO 2016"');

// Display field plots on the map.
Map.setCenter(-107.792, 39.871, 7);
Map.addLayer(fc);

// Compare the per-feature values of two properties and filter the collection
// based on the results of various relational expressions. The two properties
// to compare are invasive and non-invasive annual forb cover at each plot.
var leftProperty = 'InvAnnForbCover_AH';
var rightProperty = 'NonInvAnnForbCover_AH';

print('Plots where invasive forb cover is…');

print('…EQUAL to non-invasive cover',
      fc.filter(ee.Filter.equals(
        {leftField: leftProperty, rightField: rightProperty})));

print('…NOT EQUAL to non-invasive cover',
      fc.filter(ee.Filter.notEquals(
        {leftField: leftProperty, rightField: rightProperty})));

print('…LESS THAN non-invasive cover',
      fc.filter(ee.Filter.lessThan(
        {leftField: leftProperty, rightField: rightProperty})));

print('…LESS THAN OR EQUAL to non-invasive cover',
      fc.filter(ee.Filter.lessThanOrEquals(
        {leftField: leftProperty, rightField: rightProperty})));

print('…GREATER THAN non-invasive cover',
      fc.filter(ee.Filter.greaterThan(
        {leftField: leftProperty, rightField: rightProperty})));

print('…GREATER THAN OR EQUAL to non-invasive cover',
      fc.filter(ee.Filter.greaterThanOrEquals(
        {leftField: leftProperty, rightField: rightProperty})));

print('…is not greater than 10 percent different than non-invasive cover',
      fc.filter(ee.Filter.maxDifference(
        {difference: 10, leftField: leftProperty, rightField: rightProperty})));

// Instead of comparing values of two feature properties using the leftField
// and rightField parameters, you can compare a property value (leftProperty)
// against a constant value (rightValue).
print('Plots where invasive forb cover is greater than 20%',
      fc.filter(ee.Filter.greaterThan(
        {leftField: leftProperty, rightValue: 20})));

// You can also swap the operands to assign the constant to the left side of
// the relational expression (leftValue) and the feature property on the right
// (rightField). Here, we get the complement of the previous example.
print('Plots where 20% is greater than invasive forb cover.',
      fc.filter(ee.Filter.greaterThan(
        {leftValue: 20, rightField: leftProperty})));

// Binary filters are useful in joins. For example, group all same-site plots
// together using a saveAll join.
var groupingProp = 'SiteID';
var sitesFc = fc.distinct(groupingProp);

var joinFilter = ee.Filter.equals(
  {leftField: groupingProp, rightField: groupingProp});

var groupedPlots = ee.Join.saveAll('site_plots').apply(sitesFc, fc, joinFilter);
print('List of plots in first site', groupedPlots.first().get('site_plots'));

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)

# Field site vegetation characteristics from projects in western USA.
fc = ee.FeatureCollection('BLM/AIM/v1/TerrADat/TerrestrialAIM').filter(
    'ProjectName == "Colorado NWDO Kremmling FO 2016"'
)

# Display field plots on the map.
m = geemap.Map()
m.set_center(-107.792, 39.871, 7)
m.add_layer(fc)
display(m)

# Compare the per-feature values of two properties and filter the collection
# based on the results of various relational expressions. The two properties
# to compare are invasive and non-invasive annual forb cover at each plot.
left_property = 'InvAnnForbCover_AH'
right_property = 'NonInvAnnForbCover_AH'

display('Plots where invasive forb cover is…')

display(
    '…EQUAL to non-invasive cover',
    fc.filter(
        ee.Filter.equals(leftField=left_property, rightField=right_property)
    ),
)

display(
    '…NOT EQUAL to non-invasive cover',
    fc.filter(
        ee.Filter.notEquals(leftField=left_property, rightField=right_property)
    ),
)

display(
    '…LESS THAN non-invasive cover',
    fc.filter(
        ee.Filter.lessThan(leftField=left_property, rightField=right_property)
    ),
)

display(
    '…LESS THAN OR EQUAL to non-invasive cover',
    fc.filter(
        ee.Filter.lessThanOrEquals(
            leftField=left_property, rightField=right_property
        )
    ),
)

display(
    '…GREATER THAN non-invasive cover',
    fc.filter(
        ee.Filter.greaterThan(
            leftField=left_property, rightField=right_property
        )
    ),
)

display(
    '…GREATER THAN OR EQUAL to non-invasive cover',
    fc.filter(
        ee.Filter.greaterThanOrEquals(
            leftField=left_property, rightField=right_property
        )
    ),
)

display(
    '…is not greater than 10 percent different than non-invasive cover',
    fc.filter(
        ee.Filter.maxDifference(
            difference=10, leftField=left_property, rightField=right_property
        )
    ),
)

# Instead of comparing values of two feature properties using the leftField
# and rightField parameters, you can compare a property value (left_property)
# against a constant value (rightValue).
display(
    'Plots where invasive forb cover is greater than 20%',
    fc.filter(ee.Filter.greaterThan(leftField=left_property, rightValue=20)),
)

# You can also swap the operands to assign the constant to the left side of
# the relational expression (leftValue) and the feature property on the right
# (rightField). Here, we get the complement of the previous example.
display(
    'Plots where 20% is greater than invasive forb cover.',
    fc.filter(ee.Filter.greaterThan(leftValue=20, rightField=left_property)),
)

# Binary filters are useful in joins. For example, group all same-site plots
# together using a saveAll join.
grouping_prop = 'SiteID'
sites_fc = fc.distinct(grouping_prop)

join_filter = ee.Filter.equals(
    leftField=grouping_prop, rightField=grouping_prop
)

grouped_plots = ee.Join.saveAll('site_plots').apply(sites_fc, fc, join_filter)
display('List of plots in first site', grouped_plots.first().get('site_plots'))