BigEarthNet is a new large-scale Sentinel-2 benchmark archive, consisting of
590,326 Sentinel-2 image patches. To construct BigEarthNet, 125 Sentinel-2
tiles were acquired between June 2017 and May 2018 over the 10 countries
(Austria, Belgium, Finland, Ireland, Kosovo, Lithuania, Luxembourg,
Portugal, Serbia, Switzerland) of Europe. All the tiles were atmospherically
corrected by the Sentinel-2 Level 2A product generation and formatting tool
(sen2cor). Then, they were divided into 590,326 non-overlapping image
patches. Each image patch was annotated by the multiple land-cover classes
(i.e., multi-labels) that were provided from the CORINE Land Cover database
of the year 2018 (CLC 2018).
Bands
Bands
Name
Scale
Pixel Size
Wavelength
Description
B1
0.0001
60 meters
443.9nm (S2A) / 442.3nm (S2B)
Aerosols
B2
0.0001
10 meters
496.6nm (S2A) / 492.1nm (S2B)
Blue
B3
0.0001
10 meters
560nm (S2A) / 559nm (S2B)
Green
B4
0.0001
10 meters
664.5nm (S2A) / 665nm (S2B)
Red
B5
0.0001
20 meters
703.9nm (S2A) / 703.8nm (S2B)
Red Edge 1
B6
0.0001
20 meters
740.2nm (S2A) / 739.1nm (S2B)
Red Edge 2
B7
0.0001
20 meters
782.5nm (S2A) / 779.7nm (S2B)
Red Edge 3
B8
0.0001
10 meters
835.1nm (S2A) / 833nm (S2B)
NIR
B9
0.0001
60 meters
945nm (S2A) / 943.2nm (S2B)
Water vapor
B10
0.0001
60 meters
1373.5nm (S2A) / 1376.9nm (S2B)
Cirrus
B11
0.0001
20 meters
1613.7nm (S2A) / 1610.4nm (S2B)
SWIR 1
B12
0.0001
20 meters
2202.4nm (S2A) / 2185.7nm (S2B)
SWIR 2
B8A
0.0001
20 meters
864.8nm (S2A) / 864nm (S2B)
Red Edge 4
Image Properties
Image Properties
Name
Type
Description
labels
STRING_LIST
List of landcover types found in this image
source
STRING
Product ID of the corresponding Sentinel-2 1C image
tile_x
DOUBLE
X coordinate of tile in source image
tile_y
DOUBLE
Y coordinate of tile in source image
Terms of Use
Terms of Use
The BigEarthNet Archive is licensed under the Community Data License
Agreement - Permissive, Version 1.0. For more information,
please refer to
https://cdla.dev/permissive-1-0.
Citations
Citations:
G. Sumbul, M. Charfuelan, B. Demir, V. Markl, BigEarthNet: A Large-Scale
Benchmark Archive for Remote Sensing Image Understanding, IEEE International
Conference on Geoscience and Remote Sensing Symposium, pp. 5901-5904,
Yokohama, Japan, 2019.
BigEarthNet is a new large-scale Sentinel-2 benchmark archive, consisting of 590,326 Sentinel-2 image patches. To construct BigEarthNet, 125 Sentinel-2 tiles were acquired between June 2017 and May 2018 over the 10 countries (Austria, Belgium, Finland, Ireland, Kosovo, Lithuania, Luxembourg, Portugal, Serbia, Switzerland) of Europe. All the tiles were atmospherically corrected …
[null,null,[],[[["\u003cp\u003eBigEarthNet is a large-scale Sentinel-2 benchmark archive containing 590,326 image patches acquired between June 2017 and May 2018.\u003c/p\u003e\n"],["\u003cp\u003eIt covers 10 European countries (Austria, Belgium, Finland, Ireland, Kosovo, Lithuania, Luxembourg, Portugal, Serbia, Switzerland) and is based on 125 Sentinel-2 tiles.\u003c/p\u003e\n"],["\u003cp\u003eEach image patch is annotated with multiple land-cover classes from the CORINE Land Cover database of 2018 (CLC 2018).\u003c/p\u003e\n"],["\u003cp\u003eBigEarthNet provides Sentinel-2 Level 2A product image patches, atmospherically corrected by sen2cor.\u003c/p\u003e\n"],["\u003cp\u003eThe archive is licensed under the Community Data License Agreement - Permissive, Version 1.0.\u003c/p\u003e\n"]]],[],null,["# TUBerlin/BigEarthNet/v1\n\nDataset Availability\n: 2017-06-01T00:00:00Z--2018-05-31T00:00:00Z\n\nDataset Provider\n:\n\n\n [BigEarthNet](http://bigearth.net/)\n\nTags\n:\n [copernicus](/earth-engine/datasets/tags/copernicus) [landuse-landcover](/earth-engine/datasets/tags/landuse-landcover) [sentinel](/earth-engine/datasets/tags/sentinel) \n chip \n corine-derived \n label \n ml \ntile \n\n#### Description\n\nBigEarthNet is a new large-scale Sentinel-2 benchmark archive, consisting of\n590,326 Sentinel-2 image patches. To construct BigEarthNet, 125 Sentinel-2\ntiles were acquired between June 2017 and May 2018 over the 10 countries\n(Austria, Belgium, Finland, Ireland, Kosovo, Lithuania, Luxembourg,\nPortugal, Serbia, Switzerland) of Europe. All the tiles were atmospherically\ncorrected by the Sentinel-2 Level 2A product generation and formatting tool\n(sen2cor). Then, they were divided into 590,326 non-overlapping image\npatches. Each image patch was annotated by the multiple land-cover classes\n(i.e., multi-labels) that were provided from the CORINE Land Cover database\nof the year 2018 (CLC 2018).\n\n### Bands\n\n**Bands**\n\n| Name | Scale | Pixel Size | Wavelength | Description |\n|-------|--------|------------|---------------------------------|-------------|\n| `B1` | 0.0001 | 60 meters | 443.9nm (S2A) / 442.3nm (S2B) | Aerosols |\n| `B2` | 0.0001 | 10 meters | 496.6nm (S2A) / 492.1nm (S2B) | Blue |\n| `B3` | 0.0001 | 10 meters | 560nm (S2A) / 559nm (S2B) | Green |\n| `B4` | 0.0001 | 10 meters | 664.5nm (S2A) / 665nm (S2B) | Red |\n| `B5` | 0.0001 | 20 meters | 703.9nm (S2A) / 703.8nm (S2B) | Red Edge 1 |\n| `B6` | 0.0001 | 20 meters | 740.2nm (S2A) / 739.1nm (S2B) | Red Edge 2 |\n| `B7` | 0.0001 | 20 meters | 782.5nm (S2A) / 779.7nm (S2B) | Red Edge 3 |\n| `B8` | 0.0001 | 10 meters | 835.1nm (S2A) / 833nm (S2B) | NIR |\n| `B9` | 0.0001 | 60 meters | 945nm (S2A) / 943.2nm (S2B) | Water vapor |\n| `B10` | 0.0001 | 60 meters | 1373.5nm (S2A) / 1376.9nm (S2B) | Cirrus |\n| `B11` | 0.0001 | 20 meters | 1613.7nm (S2A) / 1610.4nm (S2B) | SWIR 1 |\n| `B12` | 0.0001 | 20 meters | 2202.4nm (S2A) / 2185.7nm (S2B) | SWIR 2 |\n| `B8A` | 0.0001 | 20 meters | 864.8nm (S2A) / 864nm (S2B) | Red Edge 4 |\n\n### Image Properties\n\n**Image Properties**\n\n| Name | Type | Description |\n|--------|-------------|-----------------------------------------------------|\n| labels | STRING_LIST | List of landcover types found in this image |\n| source | STRING | Product ID of the corresponding Sentinel-2 1C image |\n| tile_x | DOUBLE | X coordinate of tile in source image |\n| tile_y | DOUBLE | Y coordinate of tile in source image |\n\n### Terms of Use\n\n**Terms of Use**\n\nThe BigEarthNet Archive is licensed under the Community Data License\nAgreement - Permissive, Version 1.0. For more information,\nplease refer to\n[https://cdla.dev/permissive-1-0](https://cdla.dev/permissive-1-0/).\n\n### Citations\n\nCitations:\n\n- G. Sumbul, M. Charfuelan, B. Demir, V. Markl, BigEarthNet: A Large-Scale\n Benchmark Archive for Remote Sensing Image Understanding, IEEE International\n Conference on Geoscience and Remote Sensing Symposium, pp. 5901-5904,\n Yokohama, Japan, 2019.\n\n### Explore with Earth Engine\n\n| **Important:** Earth Engine is a platform for petabyte-scale scientific analysis and visualization of geospatial datasets, both for public benefit and for business and government users. Earth Engine is free to use for research, education, and nonprofit use. To get started, please [register for Earth Engine access.](https://console.cloud.google.com/earth-engine)\n\n### Code Editor (JavaScript)\n\n```javascript\nvar geometry = ee.Geometry.Polygon(\n [[\n [16.656886757418057, 48.27086673747943],\n [16.656886757418057, 48.21359065567954],\n [16.733276070162198, 48.21359065567954],\n [16.733276070162198, 48.27086673747943]]]);\n\nvar ic = ee.ImageCollection('TUBerlin/BigEarthNet/v1');\n\nvar filtered = ic.filterBounds(geometry);\n\nvar tiles = filtered.map(function(image) {\n var labels = ee.List(image.get('labels'));\n\n var urban = labels.indexOf('Discontinuous urban fabric').gte(0);\n var highlight_urban = ee.Image(urban).toInt().multiply(1000);\n\n return image.addBands(\n {srcImg: image.select(['B4']).add(highlight_urban), overwrite: true});\n});\n\nvar image = tiles.mosaic().clip(geometry);\n\nvar visParams = {bands: ['B4', 'B3', 'B2'], min: 0, max: 3000};\n\nMap.addLayer(image, visParams);\nMap.centerObject(image, 13);\n```\n[Open in Code Editor](https://code.earthengine.google.com/?scriptPath=Examples:Datasets/TUBerlin/TUBerlin_BigEarthNet_v1) \n[TUBerlin/BigEarthNet/v1](/earth-engine/datasets/catalog/TUBerlin_BigEarthNet_v1) \nBigEarthNet is a new large-scale Sentinel-2 benchmark archive, consisting of 590,326 Sentinel-2 image patches. To construct BigEarthNet, 125 Sentinel-2 tiles were acquired between June 2017 and May 2018 over the 10 countries (Austria, Belgium, Finland, Ireland, Kosovo, Lithuania, Luxembourg, Portugal, Serbia, Switzerland) of Europe. All the tiles were atmospherically corrected ... \nTUBerlin/BigEarthNet/v1, copernicus,landuse-landcover,sentinel \n2017-06-01T00:00:00Z/2018-05-31T00:00:00Z \n36.9 -9 68.1 31.6 \nGoogle Earth Engine \nhttps://developers.google.com/earth-engine/datasets\n\n- [](https://doi.org/http://bigearth.net/)\n- [](https://doi.org/https://developers.google.com/earth-engine/datasets/catalog/TUBerlin_BigEarthNet_v1)"]]