Vegetation Index · RGB-based

NDGI – Normalized Difference Greenness Index

NDGI is a vegetation index based on visible green and red reflectance, designed to quantify greenness when NIR bands are not available. It works perfectly on RGB satellite or drone imagery.

1. Scientific Definition

The Normalized Difference Greenness Index (NDGI) measures vegetation greenness using the Red and Green bands. It is especially useful when only visible spectrum data is available (RGB cameras, drones, some satellite composites).

Formula

A commonly used formulation of NDGI is:

NDGI = (Green − Red) / (Green + Red) Dimensionless (–1 to +1)

Where:

  • Green: Green band reflectance
  • Red: Red band reflectance

Typical Interpretation

NDGI Range Interpretation
< 0.0 Water, shadows, non-vegetated dark surfaces
0.0 – 0.1 Very sparse vegetation / bare soil
0.1 – 0.3 Low to moderate vegetation greenness
> 0.3 Dense and healthy green vegetation

Key Applications

  • Vegetation mapping from RGB drone imagery
  • Crop monitoring where NIR is not available
  • Urban greenness assessment (parks, street vegetation)
  • Quick vegetation screening in high-resolution RGB imagery

2. Data & Bands for NDGI

Common Sensors & Bands

  • Sentinel-2 (ESA) – 10 m
    • Green: B3 (~560 nm)
    • Red: B4 (~665 nm)
  • Landsat 8/9 OLI – 30 m
    • Green: B3
    • Red: B4
  • UAV / RGB cameras
    • Use R & G reflectance from calibrated RGB images

Good Practice

  • Use surface reflectance when available.
  • Remove shadows and dark pixels to improve output quality.
  • Clip NDGI raster to AOI before export.

Palette Suggestion

Suggested NDGI palette: [ "#440154", "#3b528b", "#21908c", "#5dc963", "#fde725" ]

3. Google Earth Engine Code – NDGI for Any AOI

Steps: open code.earthengine.google.com → New Script → paste the code → draw your AOI as geometry → Run → Export NDGI as GeoTIFF. 

// NDGI for any AOI using Sentinel-2 SR
//-------------------------------------------------------

// Define AOI
var roi = geometry;

// Center map
Map.centerObject(roi, 11);

// Time range
var startDate = '2023-01-01';
var endDate   = '2023-12-31';

// Load Sentinel-2 SR
var s2 = ee.ImageCollection('COPERNICUS/S2_SR')
  .filterBounds(roi)
  .filterDate(startDate, endDate)
  .filter(ee.Filter.lt('CLOUDY_PIXEL_PERCENTAGE', 20))
  .select(['B3', 'B4']); // Green, Red

var image = s2.median().clip(roi);

// NDGI = (Green - Red) / (Green + Red)
var ndgi = image.expression(
  '(G - R) / (G + R)',
  {
    'G': image.select('B3'),
    'R': image.select('B4')
  }
).rename('NDGI');

// Visualization
var ndgiVis = {
  min: -1,
  max: 1,
  palette: [
    '#440154',
    '#3b528b',
    '#21908c',
    '#5dc963',
    '#fde725'
  ]
};

Map.addLayer(ndgi, ndgiVis, 'NDGI');

// True Color preview
var rgb = ee.ImageCollection('COPERNICUS/S2_SR')
  .filterBounds(roi)
  .filterDate(startDate, endDate)
  .select(['B4','B3','B2'])
  .median()
  .clip(roi);

Map.addLayer(rgb, {min:0, max:3000}, 'RGB', false);

// Export
Export.image.toDrive({
  image: ndgi,
  description: 'NDGI_Export',
  fileNamePrefix: 'NDGI_Export',
  region: roi,
  scale: 10,
  crs: 'EPSG:4326',
  maxPixels: 1e13
});