{"success":true,"data":{"author":"Rustowicz, Cheong, Wang, Ermon, Burke and Lobell (Stanford University)","bucketAccessType":"public","bucketListable":true,"bucketName":"https://radiantearth.blob.core.windows.net/mlhub","bucketPrefix":"su-african-crops-south-sudan","category":"Earth Observation & Geospatial","cloudProvider":"azure","createdAt":"2026-11-16T02:38:26.666Z","description":"Pixel-level crop type labels for smallholder farms in South Sudan with time series from three satellites, from Stanford University's Sustainability and AI Lab. Small irregular plots, frequent cloud and scarce ground truth make these regions hard for crop mapping, so each chip carries Sentinel-1 SAR backscatter, Sentinel-2 multispectral reflectance and high-frequency 3 m PlanetScope imagery across 2016 and 2017, with one crop label per pixel; pixels without ground truth are 0. The crop is assumed constant over each chip's time series. Measured 2026-11-15: more than 40,000 files (489 GB listed and more beyond). original/ holds NumPy arrays (s1_npy/, s2_npy/, planet_npy/, truth_npy/) as used in the paper; planet/, s1/, s2/ and truth/ hold the same data as GeoTIFFs with STAC JSON. Load the arrays with numpy or the GeoTIFFs with rasterio; the authors' 2D and 3D U-Net and CLSTM code is on GitHub (roserustowicz/crop-type-mapping). Released under CC BY-SA 4.0; cite Rustowicz et al. (2019), CVPR Workshops, and the Radiant MLHub DOI 10.34911/rdnt.v6kx6n.","displayTitle":"Semantic Segmentation of Crop Type in South Sudan (Sentinel-1, Sentinel-2 and PlanetScope, 2016-2017)","doi":"10.34911/rdnt.v6kx6n","fuseMountBrowsable":true,"id":"f4bc715b-0cf6-5034-8ca9-9fe2ce0055df","license":"CC-BY-SA-4.0","licenseEvidenceUrl":"https://radiantearth.blob.core.windows.net/mlhub/su-african-crops-south-sudan/README.md","objectCountEstimate":6,"probeStatus":"ok","referenceUrl":"https://doi.org/10.34911/rdnt.v6kx6n","schemaCard":{"accessHints":[{"language":"python-local","snippet":"import os\nsorted(os.listdir('/mnt/azure-archive/data'))[:20]"},{"language":"python-adlfs","snippet":"import adlfs\nfs = adlfs.AzureBlobFileSystem(account_name='radiantearth', anon=True)\nfs.ls('mlhub/su-african-crops-south-sudan/')[:10]"},{"language":"http","snippet":"curl -s 'https://radiantearth.blob.core.windows.net/mlhub?restype=container&comp=list&prefix=su-african-crops-south-sudan/&delimiter=/&maxresults=20'"}],"format":"binary","generatedAt":"2026-11-16T02:38:34.708Z","layout":{"example":".source/metadata.json","pattern":"{prefix}/{filename}"},"mountPath":"/mnt/azure-archive/data","sampleHeaders":[{"bytes":"{\n    \"description\": \"Automatic, accurate crop type maps can provide unprecedented information for understanding food systems, especially in developing countries where ground surveys are infrequent. However, little work has applied existing methods to these data scarce environments, which also have unique challenges of irregularly shaped fields, frequent cloud coverage, small plots, and a severe lack of training data. To address this gap in the literature, we provide the first crop type semantic segmentatio","encoding":"utf8","key":".source/metadata.json","truncatedAt":512}],"sampleObjects":[".source/metadata.json"],"sidecars":[{"bytes":"# Semantic Segmentation of Crop Type in South Sudan\nAutomatic, accurate crop type maps can provide unprecedented information for understanding food systems, especially in developing countries where ground surveys are infrequent. However, little work has applied existing methods to these data scarce environments, which also have unique challenges of irregularly shaped fields, frequent cloud coverage, small plots, and a severe lack of training data. To address this gap in the literature, we provide the first crop type semantic segmentation dataset of small holder farms, specifically in Ghana and South Sudan. We are also the first to utilize high resolution, high frequency satellite data in segmenting small holder farms.\n\n\nThe dataset includes time series of satellite imagery from Sentinel-1, Sentinel-2, and PlanetScope satellites throughout 2016 and 2017. For each tile/chip in the dataset, there are time series of imagery from each of the satellites, as well as a corresponding label that defines the crop type at each pixel. The label has only one value at each pixel location, and assumes that the crop type remains the same across the full time span of the satellite image time series. In many cases where ground truth was not available, pixels have no label and are set to a value of 0.\n\n## Documentation\n* [Link](https://radiantearth.blob.core.windows.net/mlhub/su-african-crops-south-sudan/documentation.pdf)\n\n## Tools & Applications\n* [Crop type mapping of small holder farms in Ghana and South Sudan](https://github.com/roserustowicz/crop-type-mapping)\n\n## Publications\n* [Semantic Segmentation of Crop Type in Africa: A Novel Dataset and Analysis of Deep Learning Methods](https://openaccess.thecvf.com/content_CVPRW_2019/papers/cv4gc/Rustowicz_Semantic_Segmentation_of_Crop_Type_in_Africa_A_Novel_Dataset_CVPRW_2019_paper.pdf) by [Rose Rustowicz](https://www.linkedin.com/in/rose-rustowicz-ba0ba958/)\n\n## Creator & Contact\n* [Stanford University](https://www.stanford.edu/)\n* roserustowicz@gmail.com\n\n## License\n* [CC-BY-SA-4.0](https://creativecommons.org/licenses/by-sa/4.0/)\n\n## Citation & DOI\nRustowicz R., Cheong R., Wang L., Ermon S., Burke M., Lobell D. (2020) \"Semantic Segmentation of Crop Type in South Sudan Dataset\", Version 1.0, Radiant MLHub. [Date Accessed] https://doi.org/10.34911/rdnt.v6kx6n\n","encoding":"utf8","key":"README.md","name":"README.md","size":2332,"truncated":false},{"bytes":"{\n    \"description\": \"Automatic, accurate crop type maps can provide unprecedented information for understanding food systems, especially in developing countries where ground surveys are infrequent. However, little work has applied existing methods to these data scarce environments, which also have unique challenges of irregularly shaped fields, frequent cloud coverage, small plots, and a severe lack of training data. To address this gap in the literature, we provide the first crop type semantic segmentation dataset \",\n    \"tags\": [\n        \"sentinel-1\",\n        \"sentinel-2\",\n        \"planetscope\",\n        \"sar\",\n        \"crop type\",\n        \"agriculture\",\n        \"segmentation\"\n    ]\n}","encoding":"utf8","key":".source/metadata.json","name":"metadata.json","size":695,"truncated":false}],"sourceObjectKey":"su-african-crops-south-sudan/.source/metadata.json","starterCell":"path = '/mnt/azure-archive/data/.source/metadata.json'\nwith open(path, 'rb') as fh:\n    print(fh.read(64))","starterSamplePath":"/mnt/azure-archive/data/.source/metadata.json","totalObjects":"6 top-level prefixes"},"summary":"Pixel-level crop type labels for smallholder farms in South Sudan with time series from three satellites, from Stanford University's Sustainability and AI Lab.","title":"Semantic Segmentation of Crop Type in South Sudan (Sentinel-1, Sentinel-2 and PlanetScope, 2016-2017)","updatedAt":"2026-11-16T16:58:40.523Z","visibility":"public","websiteUrl":"https://radiantearth.blob.core.windows.net/mlhub/su-african-crops-south-sudan/README.md"}}