Google Earth Engine: Get the ID of images used for median calculation

Question

In Earth Engine it is quite easy to create a median image out of several images from an image collection. But is there any way to get the information which particular images were used to create the median image?

Answer 1

You can do that by applying the aggregate_array function over the filtered collection used to build the median image. In this function you need to specify the image property that corresponds to its name, which corresponds to the system:index property. As you did not provide any example, here is a simple one.

In addition, if you wish to export the names of the images as a csv file, you can also do that by transforming the ee.List (names of the images) into ee.Features, then to a ee.FeatureCollection and finally, export it using Export.table.toDrive.

// Filter image collection 
var imCol = ee.ImageCollection("LANDSAT/LE07/C01/T1_RT")
            // Filter by spatial extent of a polygon
            .filterBounds(ee.Geometry.Polygon(
        [[[-102.10208697216933, 20.235359993650682],
          [-102.10208697216933, 18.82723807089423],
          [-100.52005572216933, 18.82723807089423],
          [-100.52005572216933, 20.235359993650682]]], null, false))
            // Filter by date
            .filterDate('2008-01-01','2009-01-01')
            // Filter by a property of the images, e.g., cloud cover land
            .filter(ee.Filter.lte('CLOUD_COVER_LAND',20));

// Obtain median image
var reduced = imCol.reduce(ee.Reducer.median());

// Aggregate over a property of the images in the filtered image collection
// and obtain a list. In this case get the system:index property, which corresponds
// to the image name
var idList = imCol.aggregate_array('system:index');

// Print the list
print(idList);

// If you want to export it

// First transform list into Features
var exportFeatCol = idList.map(function(entry){
  return ee.Feature(null, ee.Dictionary({id: entry}));
});

// Export and cast object as FeatureCollection
Export.table.toDrive({
  collection: ee.FeatureCollection(exportFeatCol),
  description: 'IdImages',
  fileFormat: 'CSV'
});