Temporal segmentation algorithms using Google Earth Engine

by Inacio Bueno   Last Updated September 05, 2018 17:22 PM

I'm using the temporal segmentation algorithm Vegetation Regeneration and Disturbance Estimates - Verdet - for land cover change detection. The output layer is created but I can't display or export it (GEE returns processing errors). Does anyone know how can I work with this tool?

// NDVI image collection from Landsat TM
var collection5 = ee.ImageCollection('LANDSAT/LT05/C01/T1_SR')
    .filter(ee.Filter.eq('WRS_PATH', 219))
    .filter(ee.Filter.eq('WRS_ROW', 72))
    .filterDate('2008-01-01', '2011-12-31')
    .filterMetadata('CLOUD_COVER', 'equals', 0);

  var segIndex = function(img){
      var index = img.normalizedDifference(['B4', 'B5']);
      return index;
  var collection5 = collection5.map(segIndex);

// NDVI image collection from Landsat OLI
var collection8 = ee.ImageCollection('LANDSAT/LC08/C01/T1_SR')
    .filter(ee.Filter.eq('WRS_PATH', 219))
    .filter(ee.Filter.eq('WRS_ROW', 72))
    .filterDate('2013-01-01', '2017-12-31')
    .filterMetadata('CLOUD_COVER', 'equals', 0);

  var collection8 = collection8.map(segIndex);

// Merge collections
var merge = ee.ImageCollection(collection5.merge(collection8));

// Run VerDET
var verdet = ee.Algorithms.TemporalSegmentation.Verdet(merge);

Related Questions

How to calculate forest loss in google earth engine

Updated March 07, 2017 14:22 PM

Loop inside ImageCollection

Updated October 28, 2018 14:22 PM