Automated Annotation of Satellite Imagery using Model-based Projections

Randy S. Roberts; John W. Goforth; George F. Weinert; Charles W. Grant; Will R. Ray; Brad J. Stinson; Andrew M. Duncan

doi:10.1109/AIPR.2018.8707425

Automated Annotation of Satellite Imagery using Model-based Projections

Randy S. Roberts, John W. Goforth, George F. Weinert, Charles W. Grant, Will R. Ray, Brad J. Stinson, Andrew M. Duncan

Autonomous Systems

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

GeoVisipedia is a new and novel approach to annotating satellite imagery. It uses wiki pages to annotate objects rather than simple labels. The use of wiki pages to contain annotations is particularly useful for annotating objects in imagery of complex geospatial configurations such as industrial facilities. GeoVisipedia uses the PRISM algorithm to project annotations applied to one image to other imagery, hence enabling ubiquitous annotation. This paper derives the PRISM algorithm, which uses image metadata and a 3D facility model to create a view matrix unique to each image. The view matrix is used to project model components onto a mask which aligns the components with the objects in the scene that they represent. Wiki pages are linked to model components, which are in turn linked to the image via the component mask. An illustration of the efficacy of the PRISM algorithm is provided, demonstrating the projection of model components onto an effluent stack. We conclude with a discussion of the efficiencies of GeoVisipedia over manual annotation, and the use of PRISM for creating training sets for machine learning algorithms.

Original language	English
Title of host publication	2018 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2018
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781538693063
DOIs	https://doi.org/10.1109/AIPR.2018.8707425
State	Published - Jul 2 2018
Event	2018 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2018 - Washington, United States Duration: Oct 9 2018 → Oct 11 2018

Publication series

Name	Proceedings - Applied Imagery Pattern Recognition Workshop
Volume	2018-October
ISSN (Print)	2164-2516

Conference

Conference	2018 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2018
Country/Territory	United States
City	Washington
Period	10/9/18 → 10/11/18

Funding

Figure 4 is reprinted from [2] with permission from the publisher. This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. Research supported by the Defense Nuclear Nonproliferation R&D, Office of Proliferation Detection (NA-221), National Nuclear Security Administration. Corresponding author: R. S. Roberts. LLNL-CONF-759499.

Keywords

3D geospatial models
Satellite imagery
geospatial image analysis
visual information

Access to Document

10.1109/AIPR.2018.8707425

Cite this

Roberts, R. S., Goforth, J. W., Weinert, G. F., Grant, C. W., Ray, W. R., Stinson, B. J., & Duncan, A. M. (2018). Automated Annotation of Satellite Imagery using Model-based Projections. In 2018 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2018 Article 8707425 (Proceedings - Applied Imagery Pattern Recognition Workshop; Vol. 2018-October). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/AIPR.2018.8707425

@inproceedings{affc0457470748bda5c9add257883e44,

title = "Automated Annotation of Satellite Imagery using Model-based Projections",

abstract = "GeoVisipedia is a new and novel approach to annotating satellite imagery. It uses wiki pages to annotate objects rather than simple labels. The use of wiki pages to contain annotations is particularly useful for annotating objects in imagery of complex geospatial configurations such as industrial facilities. GeoVisipedia uses the PRISM algorithm to project annotations applied to one image to other imagery, hence enabling ubiquitous annotation. This paper derives the PRISM algorithm, which uses image metadata and a 3D facility model to create a view matrix unique to each image. The view matrix is used to project model components onto a mask which aligns the components with the objects in the scene that they represent. Wiki pages are linked to model components, which are in turn linked to the image via the component mask. An illustration of the efficacy of the PRISM algorithm is provided, demonstrating the projection of model components onto an effluent stack. We conclude with a discussion of the efficiencies of GeoVisipedia over manual annotation, and the use of PRISM for creating training sets for machine learning algorithms.",

keywords = "3D geospatial models, Satellite imagery, geospatial image analysis, visual information",

author = "Roberts, {Randy S.} and Goforth, {John W.} and Weinert, {George F.} and Grant, {Charles W.} and Ray, {Will R.} and Stinson, {Brad J.} and Duncan, {Andrew M.}",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 2018 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2018 ; Conference date: 09-10-2018 Through 11-10-2018",

year = "2018",

month = jul,

day = "2",

doi = "10.1109/AIPR.2018.8707425",

language = "English",

series = "Proceedings - Applied Imagery Pattern Recognition Workshop",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2018 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2018",

}

Roberts, RS, Goforth, JW, Weinert, GF, Grant, CW, Ray, WR, Stinson, BJ & Duncan, AM 2018, Automated Annotation of Satellite Imagery using Model-based Projections. in 2018 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2018., 8707425, Proceedings - Applied Imagery Pattern Recognition Workshop, vol. 2018-October, Institute of Electrical and Electronics Engineers Inc., 2018 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2018, Washington, United States, 10/9/18. https://doi.org/10.1109/AIPR.2018.8707425

Automated Annotation of Satellite Imagery using Model-based Projections. / Roberts, Randy S.; Goforth, John W.; Weinert, George F. et al.
2018 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2018. Institute of Electrical and Electronics Engineers Inc., 2018. 8707425 (Proceedings - Applied Imagery Pattern Recognition Workshop; Vol. 2018-October).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Automated Annotation of Satellite Imagery using Model-based Projections

AU - Roberts, Randy S.

AU - Goforth, John W.

AU - Weinert, George F.

AU - Grant, Charles W.

AU - Ray, Will R.

AU - Stinson, Brad J.

AU - Duncan, Andrew M.

PY - 2018/7/2

Y1 - 2018/7/2

N2 - GeoVisipedia is a new and novel approach to annotating satellite imagery. It uses wiki pages to annotate objects rather than simple labels. The use of wiki pages to contain annotations is particularly useful for annotating objects in imagery of complex geospatial configurations such as industrial facilities. GeoVisipedia uses the PRISM algorithm to project annotations applied to one image to other imagery, hence enabling ubiquitous annotation. This paper derives the PRISM algorithm, which uses image metadata and a 3D facility model to create a view matrix unique to each image. The view matrix is used to project model components onto a mask which aligns the components with the objects in the scene that they represent. Wiki pages are linked to model components, which are in turn linked to the image via the component mask. An illustration of the efficacy of the PRISM algorithm is provided, demonstrating the projection of model components onto an effluent stack. We conclude with a discussion of the efficiencies of GeoVisipedia over manual annotation, and the use of PRISM for creating training sets for machine learning algorithms.

AB - GeoVisipedia is a new and novel approach to annotating satellite imagery. It uses wiki pages to annotate objects rather than simple labels. The use of wiki pages to contain annotations is particularly useful for annotating objects in imagery of complex geospatial configurations such as industrial facilities. GeoVisipedia uses the PRISM algorithm to project annotations applied to one image to other imagery, hence enabling ubiquitous annotation. This paper derives the PRISM algorithm, which uses image metadata and a 3D facility model to create a view matrix unique to each image. The view matrix is used to project model components onto a mask which aligns the components with the objects in the scene that they represent. Wiki pages are linked to model components, which are in turn linked to the image via the component mask. An illustration of the efficacy of the PRISM algorithm is provided, demonstrating the projection of model components onto an effluent stack. We conclude with a discussion of the efficiencies of GeoVisipedia over manual annotation, and the use of PRISM for creating training sets for machine learning algorithms.

KW - 3D geospatial models

KW - Satellite imagery

KW - geospatial image analysis

KW - visual information

UR - http://www.scopus.com/inward/record.url?scp=85065996694&partnerID=8YFLogxK

U2 - 10.1109/AIPR.2018.8707425

DO - 10.1109/AIPR.2018.8707425

M3 - Conference contribution

AN - SCOPUS:85065996694

T3 - Proceedings - Applied Imagery Pattern Recognition Workshop

BT - 2018 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2018

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2018 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2018

Y2 - 9 October 2018 through 11 October 2018

ER -

Roberts RS, Goforth JW, Weinert GF, Grant CW, Ray WR, Stinson BJ et al. Automated Annotation of Satellite Imagery using Model-based Projections. In 2018 IEEE Applied Imagery Pattern Recognition Workshop, AIPR 2018. Institute of Electrical and Electronics Engineers Inc. 2018. 8707425. (Proceedings - Applied Imagery Pattern Recognition Workshop). doi: 10.1109/AIPR.2018.8707425

Automated Annotation of Satellite Imagery using Model-based Projections

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this