sarkarghya/power-plant-satellite-imagery-dataset-mirror

Power Plant Satellite Imagery Dataset — Figshare mirror

This repository is an unmodified file mirror of version 1 of the Figshare record Power Plant Satellite Imagery Dataset. Figshare remains the canonical source. This mirror is provided for easier access from Hugging Face tooling; it is not a new dataset release and is not affiliated with or endorsed by the original authors.

The original dataset contains satellite imagery for 4,454 United States power plants: 1 m four-band NAIP imagery, Landsat 8 imagery, and crowdsourced spatial-extent annotations. See the original overview PDF and README under figshare/ before use.

Citation and provenance

• Original DOI: 10.6084/m9.figshare.5307364.v1
• Figshare article ID: 5307364
• Version: 1
• Published: 2017-08-16T13:33:58Z
• Authors: Kyle Bradbury, Benjamin Brigman, Gouttham Chandrasekar, Leslie Collins, Shamikh Hossain, Marc Jeuland, Timothy Johnson, Boning Li, Trishul Nagenalli
• Original license: CC BY 4.0

Please cite the original Figshare record and authors rather than this mirror.

Mirrored files

All source files are stored byte-for-byte under figshare/. Their Figshare MD5 checksums are verified before upload. figshare_metadata.json is a snapshot of the original public API metadata, and checksums.md5 is a machine-readable integrity manifest.

• Power Plant Satellite Imagery Dataset Overview.pdf — 791,923 bytes — MD5 94cde642eacdbe1ec27239c341de1fc4
• README.md — 5,588 bytes — MD5 41373b43f95ba72c74fef59d8d9c2672
• uspp_naip.zip — 20,025,754,760 bytes — MD5 57a48a05b10cf0036ab983113d0cd8dd
• uspp_landsat.zip — 54,920,515 bytes — MD5 29189bc2830f9e1f69cd41717d0b5a4a
• annotations.zip — 26,936,191 bytes — MD5 c951077f44ee380d1a5e9ea150d0b624
• eGRID2014_Data_v2_US_EPA_with_imagery_list_for_power_plants.xlsx — 13,725,154 bytes — MD5 d4c929e4ef45e86b38fb8de35fab254c
• sample.zip — 6,493,760 bytes — MD5 20033b15c5432714e63f685b63c5ab3c

Integrity and limitations

The imagery and labels retain their original formats and vintage. No normalization, resampling, correction, or relabeling has been performed. Upstream source imagery may have additional attribution or use requirements; consult the original documentation.