Coupling rendering and generative adversarial networks for artificial SAS image generation

Albert Reed; Isaac D. Gerg; John D. McKay; Daniel C. Brown; David P. Williamsk; Suren Jayasuriya

doi:10.23919/OCEANS40490.2019.8962733

Coupling rendering and generative adversarial networks for artificial SAS image generation

Albert Reed, Isaac D. Gerg, John D. McKay, Daniel C. Brown, David P. Williamsk, Suren Jayasuriya

Applied Research Laboratory (ARL)

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

Abstract

There is a growing demand for large-scale Synthetic Aperture Sonar (SAS) datasets. This demand stems from data-driven applications such as Automatic Target Recognition (ATR) [1]-[3], segmentation [4] and oceanographic research of the seafloor, simulation for sensor prototype development and calibration [5], and even potential higher level tasks such as motion estimation [6] and micronavigation [7]. Unfortunately, the acquisition of SAS data is bottlenecked by the costly deployment of SAS imaging systems, and even when data acquisition is possible, the data is often skewed towards containing barren seafloor rather than objects of interest. This skew introduces a data imbalance problem wherein a dataset can have as much as a 1000-to-1 ratio of seafloor background to object-of-interest SAS image chips.

Original language	English (US)
Title of host publication	OCEANS 2019 MTS/IEEE Seattle, OCEANS 2019
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9780578576183
DOIs	https://doi.org/10.23919/OCEANS40490.2019.8962733
State	Published - Oct 2019
Event	2019 OCEANS MTS/IEEE Seattle, OCEANS 2019 - Seattle, United States Duration: Oct 27 2019 → Oct 31 2019

Publication series

Name	OCEANS 2019 MTS/IEEE Seattle, OCEANS 2019

Conference

Conference	2019 OCEANS MTS/IEEE Seattle, OCEANS 2019
Country	United States
City	Seattle
Period	10/27/19 → 10/31/19

All Science Journal Classification (ASJC) codes

Automotive Engineering
Ocean Engineering
Acoustics and Ultrasonics
Fluid Flow and Transfer Processes
Oceanography

Access to Document

10.23919/OCEANS40490.2019.8962733

Fingerprint Dive into the research topics of 'Coupling rendering and generative adversarial networks for artificial SAS image generation'. Together they form a unique fingerprint.

Cite this

Reed, A., Gerg, I. D., McKay, J. D., Brown, D. C., Williamsk, D. P., & Jayasuriya, S. (2019). Coupling rendering and generative adversarial networks for artificial SAS image generation. In OCEANS 2019 MTS/IEEE Seattle, OCEANS 2019 [8962733] (OCEANS 2019 MTS/IEEE Seattle, OCEANS 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.23919/OCEANS40490.2019.8962733

@inproceedings{e1eac3540c6043fca296c716176c734e,

title = "Coupling rendering and generative adversarial networks for artificial SAS image generation",

abstract = "There is a growing demand for large-scale Synthetic Aperture Sonar (SAS) datasets. This demand stems from data-driven applications such as Automatic Target Recognition (ATR) [1]-[3], segmentation [4] and oceanographic research of the seafloor, simulation for sensor prototype development and calibration [5], and even potential higher level tasks such as motion estimation [6] and micronavigation [7]. Unfortunately, the acquisition of SAS data is bottlenecked by the costly deployment of SAS imaging systems, and even when data acquisition is possible, the data is often skewed towards containing barren seafloor rather than objects of interest. This skew introduces a data imbalance problem wherein a dataset can have as much as a 1000-to-1 ratio of seafloor background to object-of-interest SAS image chips.",

author = "Albert Reed and Gerg, {Isaac D.} and McKay, {John D.} and Brown, {Daniel C.} and Williamsk, {David P.} and Suren Jayasuriya",

year = "2019",

month = oct,

doi = "10.23919/OCEANS40490.2019.8962733",

language = "English (US)",

series = "OCEANS 2019 MTS/IEEE Seattle, OCEANS 2019",

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

booktitle = "OCEANS 2019 MTS/IEEE Seattle, OCEANS 2019",

address = "United States",

note = "2019 OCEANS MTS/IEEE Seattle, OCEANS 2019 ; Conference date: 27-10-2019 Through 31-10-2019",

}

Reed, A, Gerg, ID, McKay, JD, Brown, DC, Williamsk, DP & Jayasuriya, S 2019, Coupling rendering and generative adversarial networks for artificial SAS image generation. in OCEANS 2019 MTS/IEEE Seattle, OCEANS 2019., 8962733, OCEANS 2019 MTS/IEEE Seattle, OCEANS 2019, Institute of Electrical and Electronics Engineers Inc., 2019 OCEANS MTS/IEEE Seattle, OCEANS 2019, Seattle, United States, 10/27/19. https://doi.org/10.23919/OCEANS40490.2019.8962733

Coupling rendering and generative adversarial networks for artificial SAS image generation. / Reed, Albert; Gerg, Isaac D.; McKay, John D.; Brown, Daniel C.; Williamsk, David P.; Jayasuriya, Suren.

OCEANS 2019 MTS/IEEE Seattle, OCEANS 2019. Institute of Electrical and Electronics Engineers Inc., 2019. 8962733 (OCEANS 2019 MTS/IEEE Seattle, OCEANS 2019).

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

TY - GEN

T1 - Coupling rendering and generative adversarial networks for artificial SAS image generation

AU - Reed, Albert

AU - Gerg, Isaac D.

AU - McKay, John D.

AU - Brown, Daniel C.

AU - Williamsk, David P.

AU - Jayasuriya, Suren

PY - 2019/10

Y1 - 2019/10

N2 - There is a growing demand for large-scale Synthetic Aperture Sonar (SAS) datasets. This demand stems from data-driven applications such as Automatic Target Recognition (ATR) [1]-[3], segmentation [4] and oceanographic research of the seafloor, simulation for sensor prototype development and calibration [5], and even potential higher level tasks such as motion estimation [6] and micronavigation [7]. Unfortunately, the acquisition of SAS data is bottlenecked by the costly deployment of SAS imaging systems, and even when data acquisition is possible, the data is often skewed towards containing barren seafloor rather than objects of interest. This skew introduces a data imbalance problem wherein a dataset can have as much as a 1000-to-1 ratio of seafloor background to object-of-interest SAS image chips.

AB - There is a growing demand for large-scale Synthetic Aperture Sonar (SAS) datasets. This demand stems from data-driven applications such as Automatic Target Recognition (ATR) [1]-[3], segmentation [4] and oceanographic research of the seafloor, simulation for sensor prototype development and calibration [5], and even potential higher level tasks such as motion estimation [6] and micronavigation [7]. Unfortunately, the acquisition of SAS data is bottlenecked by the costly deployment of SAS imaging systems, and even when data acquisition is possible, the data is often skewed towards containing barren seafloor rather than objects of interest. This skew introduces a data imbalance problem wherein a dataset can have as much as a 1000-to-1 ratio of seafloor background to object-of-interest SAS image chips.

UR - http://www.scopus.com/inward/record.url?scp=85079053724&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85079053724&partnerID=8YFLogxK

U2 - 10.23919/OCEANS40490.2019.8962733

DO - 10.23919/OCEANS40490.2019.8962733

M3 - Conference contribution

T3 - OCEANS 2019 MTS/IEEE Seattle, OCEANS 2019

BT - OCEANS 2019 MTS/IEEE Seattle, OCEANS 2019

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2019 OCEANS MTS/IEEE Seattle, OCEANS 2019

Y2 - 27 October 2019 through 31 October 2019

ER -