Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 May 16;11(1):502.
doi: 10.1038/s41597-024-03219-x.

A baseline structure inventory with critical attribution for the US and its territories

Hsiuhan Lexie Yang  1 , Melanie Laverdiere  2 , Taylor Hauser  2 , Benjamin Swan  2 , Erik Schmidt  2 , Jessica Moehl  2 , Andrew Reith  2 , Daniel Adams  2 , Bennett Morris  2 , Jacob McKee  2 , Matthew Whitehead  2 , Mark Tuttle  2
Affiliations

A baseline structure inventory with critical attribution for the US and its territories

Hsiuhan Lexie Yang et al. Sci Data. .

Abstract

Leveraging high performance computing, remote sensing, geographic data science, machine learning, and computer vision, Oak Ridge National Laboratory has partnered with Federal Emergency Management Agency (FEMA) to build a baseline structure inventory covering the US and its territories to support disaster preparedness, response, and recovery. The dataset contains more than 125 million structures with critical attribution, and is ready to be used by federal agencies, local government and first responders to accelerate on-the-ground response to disasters, further identify vulnerable areas, and develop strategies to enhance the resilience of critical structures and communities. Data can be freely and openly accessed through Figshare data repository, ESRI's Living Atlas or FEMA's Geodata platform.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
USA Structures workflow.
Fig. 2
Fig. 2
Data sampling process.(Adapted from).
Fig. 3
Fig. 3
Example of ISOSCELES sampling for the Upper Midwest states. (a) Full set of non-overlapping source imagery used for Upper Midwest States building extraction. (b) Exemplar scenes selected in first stage of ISOSCELES sampling. (c) Exemplar scenes and exemplar sample selected at the second stage of ISOSCELES sampling.
Fig. 4
Fig. 4
Spatial distribution of labelled samples across United States.
Fig. 5
Fig. 5
Example of unfavorable structure extraction outcomes due to poor image quality or overly complex patterns that are not able to be removed during automatic QA/QC. (a) Raw structure extraction results colored in purple. Note the omissions due to the clouds. (recreated from). (b) Additional false positives that require manual QA/QC. The false positives in the yellow boxes are particular difficult to filter out by VVM.
Fig. 6
Fig. 6
Kampville sample area, Validation of the VVM.
Fig. 7
Fig. 7
False positives (colored in purple in the left) over water bodies (ocean in this example).
See this image and copyright information in PMC

References

    1. Microsoft. Us building footprints. https://github.com/microsoft/USBuildingFootprints (2018).
    1. Sirko, W. et al. Continental-scale building detection from high resolution satellite imagery. CoRRabs/2107.12283 2017.12283 (2021).
    1. Yang HL, et al. Building extraction at scale using convolutional neural network: Mapping of the united states. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2018;11:2600–2614. doi: 10.1109/JSTARS.2018.2835377. - DOI
    1. Reith, A. et al. Providing geospatial intelligence through a scalable imagery pipeline. In Advances in Scalable and Intelligence Geospatial Analysis: New Trends, Challenges and Applications (CRC Press, 2023).
    1. Yuan J. Learning building extraction in aerial scenes with convolutional networks. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2018;40:2793–2798. doi: 10.1109/TPAMI.2017.2750680. - DOI - PubMed