CV4A Kenya Crop Type Competition Dataset: Sentinel-2 time series with PlantVillage field labels, Kenya 2019 (Radiant Earth Foundation)
- Provider
- Radiant Earth Foundation and PlantVillage
- License
- CC-BY-SA-4.0 (source)
- Format
- TIFF (in sampled files)
- Stored at
- Azure Blob Storage, https://radiantearth.blob.core.windows.net/mlhub/kenya-crop-challenge
- Access
- Free and public. No account with the data provider is needed.
- DOI
- 10.34911/rdnt.dw605x
- Category
- Earth Observation & Geospatial
- Source
- https://source.coop/radiantearth/african-crops-kenya-02
About this dataset
The crop type training dataset of the CV4A (Computer Vision for Agriculture) Kenya competition, published by Radiant Earth Foundation: Sentinel-2 multispectral image time series over smallholder farms in Kenya during the 2019 growing season, paired with per-pixel crop labels and field identifiers. Ground data were collected by PlantVillage with its mobile app, walking points around each field with basemap imagery as a guide; Radiant Earth then quality-controlled the polygons against Sentinel-2 and Google basemap imagery, removing overlapping fields with conflicting labels, fixing invalid shapes and dropping fields too small for a Sentinel-2 pixel. Seven classes follow the FAO crop list: maize, cassava, common bean, and the intercrops maize and common bean, maize and cassava, maize and soybean, and cassava and common bean. The data/ folder holds four tiles (0 to 3), each with _label.tif and _field_id.tif rasters and one subfolder per acquisition date (20190606 through the season) of Sentinel-2 band GeoTIFFs (B01 to B12, B8A, and a CLD cloud probability layer); FieldIds.csv lists the train/test fields and Documentation.pdf describes collection methods. Read with rasterio or rioxarray. Licensed CC BY-SA 4.0 (DataCite record for DOI 10.34911/rdnt.dw605x); cite Radiant Earth Foundation (2020), CV4A Competition Kenya Crop Type Dataset.
How to access this dataset in Python
import adlfs
fs = adlfs.AzureBlobFileSystem(account_name='radiantearth', anon=True)
fs.ls('mlhub/kenya-crop-challenge/')[:10]Lists the files with anonymous access. Runs on any machine, with no cloud account.
Data structure
File types in sampled files: TIFF
Files
Example file: data/0/0_field_id.tif
Analyze this dataset without downloading it
On Scigantic, CV4A Kenya Crop Type Competition Dataset: Sentinel-2 time series with PlantVillage field labels, Kenya 2019 (Radiant Earth Foundation) opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.
This page is also available as Markdown and as JSON.
Related Earth Observation & Geospatial datasets
- CanElevation - Canada Digital Elevation Models
- 3Dclouds
- EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification
- NOAA Geostationary Operational Environmental Satellites (GOES) 16, 17, 18 & 19
- Satellite - Sea surface temperature - Level 3 - Multi sensor - 3 day - Day and night time
- Planette C3S Seasonal Forecast Data
- State of Colorado Basemap and Foundational Geospatial Data (statewide vector tiles)
- STAC-GeoParquet Public Missions