LandCoverNet is a global annual land cover classification training dataset with labels for the multi-spectral satellite imagery from Sentinel-1, Sentinel-2 and Landsat-8 missions in 2018. LandCoverNet Australia contains data across Australia, which accounts for ~7% of the global dataset. Each pixel is identified as one of the seven land cover classes based on its annual time series. These classes are water, natural bare ground, artificial bare ground, woody vegetation, cultivated vegetation, (semi) natural vegetation, and permanent snow/ice.
There are a total of 600 image chips of 256 x 256 pixels in LandCoverNet Australia V1.0 spanning 20 tiles. Each image chip contains temporal observations from the following satellite products with an annual class label, all stored in raster format (GeoTIFF files):
* Sentinel-1 ground range distance (GRD) with radiometric calibration and orthorectification at 10m spatial resolution
* Sentinel-2 surface reflectance product (L2A) at 10m spatial resolution
* Landsat-8 surface reflectance product from Collection 2 Level-2

Radiant Earth Foundation designed and generated this dataset with a grant from Schmidt Futures with additional support from NASA ACCESS, Microsoft AI for Earth and in kind technology support from Sinergise.

LandCoverNet is a global annual land cover classification training dataset with labels for the multi-spectral satellite imagery from Sentinel-1, Sentinel-2 and Landsat-8 missions in 2018. LandCoverNet Asia contains data across Asia, which accounts for ~31% of the global dataset. Each pixel is identified as one of the seven land cover classes based on its annual time series. These classes are water, natural bare ground, artificial bare ground, woody vegetation, cultivated vegetation, (semi) natural vegetation, and permanent snow/ice.
There are a total of 2753 image chips of 256 x 256 pixels in LandCoverNet South America V1.0 spanning 92 tiles. Each image chip contains temporal observations from the following satellite products with an annual class label, all stored in raster format (GeoTIFF files):
* Sentinel-1 ground range distance (GRD) with radiometric calibration and orthorectification at 10m spatial resolution
* Sentinel-2 surface reflectance product (L2A) at 10m spatial resolution
* Landsat-8 surface reflectance product from Collection 2 Level-2

Radiant Earth Foundation designed and generated this dataset with a grant from Schmidt Futures with additional support from NASA ACCESS, Microsoft AI for Earth and in kind technology support from Sinergise.

LandCoverNet is a global annual land cover classification training dataset with labels for the multi-spectral satellite imagery from Sentinel-1, Sentinel-2 and Landsat-8 missions in 2018. LandCoverNet South America contains data across South America, which accounts for ~13% of the global dataset. Each pixel is identified as one of the seven land cover classes based on its annual time series. These classes are water, natural bare ground, artificial bare ground, woody vegetation, cultivated vegetation, (semi) natural vegetation, and permanent snow/ice.
There are a total of 1200 image chips of 256 x 256 pixels in LandCoverNet South America V1.0 spanning 40 tiles. Each image chip contains temporal observations from the following satellite products with an annual class label, all stored in raster format (GeoTIFF files): * Sentinel-1 ground range distance (GRD) with radiometric calibration and orthorectification at 10m spatial resolution * Sentinel-2 surface reflectance product (L2A) at 10m spatial resolution * Landsat-8 surface reflectance product from Collection 2 Level-2
Radiant Earth Foundation designed and generated this dataset with a grant from Schmidt Futures with additional support from NASA ACCESS, Microsoft AI for Earth and in kind technology support from Sinergise.

This dataset consists of chips of Sentinel-1 and Sentinel-2 satellite data. Each Sentinel-1 chip contains a corresponding label for water and each Sentinel-2 chip contains a corresponding label for water and clouds. Data is stored in folders by a unique event identifier as the folder name. Within each event folder there are subfolders for Sentinel-1 (s1) and Sentinel-2 (s2) data. Each chip is contained in its own sub-folder with the folder name being the source image id, followed by a unique chip identifier consisting of a hyphenated set of 5 numbers. All bands of the satellite data, as well as the labels, and overview images are contained within the chip folder.

GEO-Bench: m-SA-crop-type dataset

This dataset has been modified to be included in the GEO-Bench dataset. All changes with respect to the original version are documented at https://github.com/ServiceNow/geo-bench.

The original version of this dataset is due to the Western Cape Department of Agriculture and the Radiant Earth Foundation (2021) and is available at: https://mlhub.earth/data/ref_south_africa_crops_competition_v1.

See the LICENSE file provided alongside this dataset for applicable licensing information.

The South Africa Crop Type Competition is a dataset produced as part of the Radiant Earth Spot the Crop Challenge. This collection contains the field identification label which represents the area of crop fields and the corresponding crop type label collected via aerial and vehicle surveys.

LandCoverNet is a global annual land cover classification training dataset with labels for the multi-spectral satellite imagery from Sentinel-1, Sentinel-2 and Landsat-8 missions in 2018. LandCoverNet Europe contains data across Europe, which accounts for ~9.5% of the global dataset. Each pixel is identified as one of the seven land cover classes based on its annual time series. These classes are water, natural bare ground, artificial bare ground, woody vegetation, cultivated vegetation, (semi) natural vegetation, and permanent snow/ice.
There are a total of 840 image chips of 256 x 256 pixels in LandCoverNet Europe V1.0 spanning 28 tiles. Each image chip contains temporal observations from the following satellite products with an annual class label, all stored in raster format (GeoTIFF files):
* Sentinel-1 ground range distance (GRD) with radiometric calibration and orthorectification at 10m spatial resolution
* Sentinel-2 surface reflectance product (L2A) at 10m spatial resolution
* Landsat-8 surface reflectance product from Collection 2 Level-2

Radiant Earth Foundation designed and generated this dataset with a grant from Schmidt Futures with additional support from NASA ACCESS, Microsoft AI for Earth and in kind technology support from Sinergise.

This dataset was produced as part of the Crop Type Detection competition at the Computer Vision for Agriculture (CV4A) Workshop at the ICLR 2020 conference. The objective of the competition was to create a machine learning model to classify fields by crop type from images collected during the growing season by the Sentinel-2 satellites.

The ground reference data were collected by the PlantVillage team, and Radiant Earth Foundation curated the training dataset after inspecting and selecting more than 4,000 fields from the original ground reference data. The dataset has been split into training and test sets (3,286 in the train and 1,402 in the test).

The dataset is cataloged in four tiles. These tiles are smaller than the original Sentinel-2 tile that has been clipped and chipped to the geographical area that labels have been collected.

Each tile has a) 13 multi-band observations throughout the growing season. Each observation includes 12 bands from Sentinel-2 L2A product, and a cloud probability layer. The twelve bands are [B01, B02, B03, B04, B05, B06, B07, B08, B8A, B09, B11, B12]. The cloud probability layer is a product of the Sentinel-2 atmospheric correction algorithm (Sen2Cor) and provides an estimated cloud probability (0-100%) per pixel. All of the bands are mapped to a common 10 m spatial resolution grid.; b) A raster layer indicating the crop ID for the fields in the training set; and c) A raster layer indicating field IDs for the fields (both training and test sets). Fields with a crop ID of 0 are the test fields.

This dataset contains field boundaries and crop type information for fields in Kenya. PlantVillage app is used to collect multiple points around each field and collectors have access to basemap imagery in the app during data collection. They use the basemap as a guide in collecting and verifying the points.

Post ground data collection, Radiant Earth Foundation conducted a quality control of the polygons using Sentinel-2 imagery of the growing season as well as Google basemap imagery. Two actions were taken on the data 1)several polygons that had overlapping areas with different crop labels were removed, 2) invalid polygons where multiple points were collected in corners of the field (within a distance of less than 0.5m) and the overall shape was not convex, were corrected. Finally, ground reference polygons were matched with corresponding time series data from Sentinel-2 satellites (listed in the source imagery property of each label item).

This dataset contains crop types of agricultural fields in four states of Uttar Pradesh, Rajasthan, Odisha and Bihar in northern India. There are 13 different classes in the dataset including Fallow land and 12 crop types of Wheat, Mustard, Lentil, Green pea, Sugarcane, Garlic, Maize, Gram, Coriander, Potato, Bersem, and Rice. The dataset is split to train and test collections as part of the AgriFieldNet India Competition. Ground reference data for this dataset is collected by IDinsight’s Data on Demand team. Radiant Earth Foundation carried out the training dataset curation and publication. This training dataset is generated through a grant from the Enabling Crop Analytics at Scale (ECAAS) Initiative funded by The Bill & Melinda Gates Foundation and implemented by Tetra Tech.

This dataset contains field boundaries and crop types from farms in Tanzania. Great African Food Company used Farmforce app to collect a point within each field, and recorded other properties including area of the field.

Radiant Earth Foundation team used the point measurements from the ground data collection and the area of each field overlaid on satellite imagery (multiple Sentinel-2 scenes during the growing season, and Google basemap) to draw the polygons for each field. These polygons do not cover the entirety of the field, and are always enclosed within the field. Therefore, they should not be used for field boundary detection, rather as reference polygons for crop type classification. Data points that were not clear if they belong to a neighboring farm (e.g. the point was on the edge of two farms)were removed from the dataset. Finally, ground reference polygons were matched with corresponding time series data from Sentinel-2 satellites (listed in the source imagery property of each label item).

This dataset contains ground reference crop type labels and multispectral and synthetic aperture radar (SAR) imagery from multiple satellites in an area located in Western Cape, South Africa. There are five crop types from the year 2017: Wheat, Barely, Canola, Lucerne/Medics, Small grain grazing. The AOI is split to three tiles. Two tiles are provided as training labels, and one tile will be used for scoring in the competition.

Input imagery consist of time series of Sentinel-2, Sentinel-1 and Planet Fusion (daily and 5-day composite) data. You can access each source from a different collection.

The Planet fusion data are made available under a CC-BY-SA license. As an exception to the AI4EO Terms and Conditions published on the competition website, you confirm, by participating in it, that you agree that your results will be made public under the same, open-source license.

The Western Cape Department of Agriculture (WCDoA) vector data are supplied via Radiant Earth Foundation with limited distribution rights. Data supplied by the WCDoA may not be distributed further or used for commercial purposes. The vector data supplied are intended strictly for use within the scope of this remote sensing competition - for the purpose of academic research to our mutual benefit. The data is intended for research purposes only and the WCDoA cannot be held responsible for any errors or omissions which may occur in the data.