Electric Transmission Infrastructure Satellite Imagery Dataset for Computer Vision


This dataset accompanies the paper, GridTracer: Automatic Mapping of Power Grids using Deep Learning and Overhead Imagery, found at https://arxiv.org/abs/2101.06390. Please see that link for more information (live link below in references).OverviewThis dataset contains fully annotated electric transmission and distribution infrastructure for approximately 264 km2 of high resolution satellite and aerial imagery, spanning 7 cities and 2 countries across 5 continents. This dataset was designed for training machine learning algorithms to automatically identify electricity infrastructure in satellite imagery; for those working on identifying the best pathways to electrification in low and middle income countries, and for researchers investigating domain adaptation for computer vision.Additional information on this dataset is available in the Documentation.pdf file included in this dataset.Data SourcesLINZ: Land Information New ZealandUSGS: United States Geological SurveySource of imagery tagged as from USGS: U.S. Geological Survey.

Data Access

Duke Authors

Cited Authors

  • Hu, W; Huang, B; Bradbury, K; Malof, J; Nair, V; Pathirathna, T; You, X; Han, Q; Yang, J; Streltsov, A; Collins, L

Published Date

  • July 21, 2021

Digital Object Identifier (DOI)

  • 10.6084/m9.figshare.14935434.v2