Broadband spectrograms from surface ships are employed in convolutional neural networks (CNNs) to predict the seabed type, ship speed, and closest point of approach (CPA) range. Three CNN architectures of differing size and depth are trained on different representations of the spectrograms. Multitask learning is employed; the seabed type prediction comes from classification, and the ship speed and CPA range are estimated via regression. Due to the lack of labeled field data, the CNNs are trained on synthetic data generated using measured sound speed profiles, four seabed types, and a random distribution of source parameters. Additional synthetic datasets are used to evaluate the ability of the trained CNNs to interpolate and extrapolate source parameters. The trained models are then applied to a measured data sample from the 2017 Seabed Characterization Experiment (SBCEX 2017). While the largest network provides slightly more accurate predictions on tests with synthetic data, the smallest network generalized better to the measured data sample. With regard to the input data type, complex pressure spectral values gave the most accurate and consistent results for the ship speed and CPA predictions with the smallest network, whereas using absolute values of the pressure provided more accurate results compared to the expected seabed types.
REFERENCES
1.
S.
Mao
,
E.
Tu
,
G.
Zhang
,
L.
Rachmawati
,
E.
Rajabally
, and
G.-B.
Huang
, “
An automatic identification system (AIS) database for maritime trajectory prediction and data mining
,” in Proceedings of ELM-2016
(
Springer
,
New York
, 2018
), pp. 241
–257
.2.
S. C.
Wales
and
R. M.
Heitmeyer
, “
An ensemble source spectra model for merchant ship-radiated noise
,” J. Acoust. Soc. Am.
111
(3
), 1211
–1231
(2002
).3.
D. J.
Battle
,
P.
Gerstoft
,
W. A.
Kuperman
,
W. S.
Hodgkiss
, and
M.
Siderius
, “
Geoacoustic inversion of tow-ship noise via near-field-matched-field processing
,” IEEE J. Oceanic Eng.
28
(3
), 454
–467
(2003
).4.
M.
Nicholas
,
J. S.
Perkins
,
G. J.
Orris
,
L. T.
Fialkowski
, and
G. J.
Heard
, “
Environmental inversion and matched-field tracking with a surface ship and an l-shaped receiver array
,” J. Acoust. Soc. Am.
116
(5
), 2891
–2901
(2004
).5.
K. D.
Heaney
, “
Rapid geoacoustic characterization using a surface ship of opportunity
,” IEEE J. Oceanic Eng.
29
(1
), 88
–99
(2004
).6.
C.
Park
,
W.
Seong
, and
P.
Gerstoft
, “
Geoacoustic inversion in time domain using ship of opportunity noise recorded on a horizontal towed array
,” J. Acoust. Soc. Am.
117
(4
), 1933
–1941
(2005
).7.
R. A.
Koch
and
D. P.
Knobles
, “
Geoacoustic inversion with ships as sources
,” J. Acoust. Soc. Am.
117
(2
), 626
–637
(2005
).8.
D.
Tollefsen
and
S. E.
Dosso
, “
Bayesian geoacoustic inversion of ship noise on a horizontal array
,” J. Acoust. Soc. Am.
124
(2
), 788
–795
(2008
).9.
S. A.
Stotts
,
R. A.
Koch
,
S. M.
Joshi
,
V. T.
Nguyen
,
V. W.
Ferreri
, and
D. P.
Knobles
, “
Geoacoustic inversions of horizontal and vertical line array acoustic data from a surface ship source of opportunity
,” IEEE J. Oceanic Eng.
35
(1
), 79
–102
(2010
).10.
C.
Gervaise
,
B. G.
Kinda
,
J.
Bonnel
,
Y.
Stéphan
, and
S.
Vallez
, “
Passive geoacoustic inversion with a single hydrophone using broadband ship noise
,” J. Acoust. Soc. Am.
131
(3
), 1999
–2010
(2012
).11.
S. E.
Crocker
,
P. L.
Nielsen
,
J. H.
Miller
, and
M.
Siderius
, “
Geoacoustic inversion of ship radiated noise in shallow water using data from a single hydrophone
,” J. Acoust. Soc. Am.
136
(5
), EL362
–EL368
(2014
).12.
S.-H.
Byun
,
C. M.
Verlinden
, and
K. G.
Sabra
, “
Blind deconvolution of shipping sources in an ocean waveguide
,” J. Acoust. Soc. Am.
141
(2
), 797
–807
(2017
).13.
K. L.
Gemba
,
J.
Sarkar
,
B.
Cornuelle
,
W. S.
Hodgkiss
, and
W.
Kuperman
, “
Estimating relative channel impulse responses from ships of opportunity in a shallow water environment
,” J. Acoust. Soc. Am.
144
(3
), 1231
–1244
(2018
).14.
L.
Muzi
,
M.
Siderius
, and
C. M.
Verlinden
, “
Passive bottom reflection-loss estimation using ship noise and a vertical line array
,” J. Acoust. Soc. Am.
141
(6
), 4372
–4379
(2017
).15.
L.
Xu
,
K.
Yang
, and
Q.
Yang
, “
Joint time-frequency inversion for seabed properties of ship noise on a vertical line array in south china sea
,” IEEE Access
6
, 62856
–62864
(2018
).16.
D.
Tollefsen
and
S. E.
Dosso
, “
Three-dimensional source tracking in an uncertain environment
,” J. Acoust. Soc. Am.
125
(5
), 2909
–2917
(2009
).17.
D.
Tollefsen
, “
Bayesian geoacoustic inversion and source tracking for horizontal line array data
,” J. Acoust. Soc. Am.
128
(1
), 506
(2010
).18.
D.
Tollefsen
and
S. E.
Dosso
, “
Three-dimensional localization of multiple sources in an uncertain ocean environment
,” Proc. Mtgs. Acoust.
19
(1
), 070071
(2013
).19.
D.
Tollefsen
,
P.
Gerstoft
, and
W. S.
Hodgkiss
, “
Multiple-array passive acoustic source localization in shallow water
,” J. Acoust. Soc. Am.
141
(3
), 1501
–1513
(2017
).20.
Z.
Huang
,
J.
Xu
,
Z.
Gong
,
H.
Wang
, and
Y.
Yan
, “
Source localization using deep neural networks in a shallow water environment
,” J. Acoust. Soc. Am.
143
(5
), 2922
–2932
(2018
).21.
H.
Niu
,
E.
Ozanich
, and
P.
Gerstoft
, “
Ship localization in santa barbara channel using machine learning classifiers
,” J. Acoust. Soc. Am.
142
(5
), EL455
–EL460
(2017
).22.
H.
Niu
,
E.
Reeves
, and
P.
Gerstoft
, “
Source localization in an ocean waveguide using supervised machine learning
,” J. Acoust. Soc. Am.
142
(3
), 1176
–1188
(2017
).23.
H.
Niu
,
Z.
Gong
,
E.
Ozanich
,
P.
Gerstoft
,
H.
Wang
, and
Z.
Li
, “
Deep learning for ocean acoustic source localization using one sensor
,” arXiv:1903.12319 (2019
).24.
E.
Ozanich
,
P.
Gerstoft
, and
H.
Niu
, “
A deep network for single-snapshot direction of arrival estimation
,” in 2019 IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP)
(
IEEE
,
New York
, 2019
), pp. 1
–6
.25.
E.
Ozanich
,
P.
Gerstoft
, and
H.
Niu
, “
A feedforward neural network for direction-of-arrival estimation
,” J. Acoust. Soc. Am.
147
(3
), 2035
–2048
(2020
).26.
E. L.
Ferguson
,
R.
Ramakrishnan
,
S. B.
Williams
, and
C. T.
Jin
, “
Convolutional neural networks for passive monitoring of a shallow water environment using a single sensor
,” in 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
(
IEEE
,
New York
, 2017
), pp. 2657
–2661
.27.
W.
Liu
,
Y.
Yang
,
M.
Xu
,
L.
Lü
,
Z.
Liu
, and
Y.
Shi
, “
Source localization in the deep ocean using a convolutional neural network
,” J. Acoust. Soc. Am.
147
(4
), EL314
–EL319
(2020
).28.
P. S.
Wilson
,
D. P.
Knobles
, and
T. B.
Neilsen
, “
Guest editorial an overview of the seabed characterization experiment
,” IEEE J. Oceanic Eng.
45
(1
), 1
–13
(2020
).29.
E. K.
Westwood
,
C. T.
Tindle
, and
N. R.
Chapman
, “
A normal mode model for acousto-elastic ocean environments
,” J. Acoust. Soc. Am.
100
(6
), 3631
–3645
(1996
).30.
D. F.
Van Komen
,
T. B.
Neilsen
,
K.
Howarth
,
D. P.
Knobles
, and
P. H.
Dahl
, “
Seabed and range estimation of impulsive time series using a convolutional neural network
,” J. Acoust. Soc. Am.
147
(5
), EL403
–EL408
(2020
).31.
D. P.
Knobles
,
R. A.
Koch
,
L. A.
Thompson
,
K. C.
Focke
, and
P. E.
Eisman
, “
Broadband sound propagation in shallow water and geoacoustic inversion
,” J. Acoust. Soc. Am.
113
(1
), 205
–222
(2003
).32.
D. P.
Knobles
,
P. S.
Wilson
,
J. A.
Goff
,
L.
Wan
,
M. J.
Buckingham
,
J. D.
Chaytor
, and
M.
Badiey
, “
Maximum entropy derived statistics of sound-speed structure in a fine-grained sediment inferred from sparse broadband acoustic measurements on the new england continental shelf
,” IEEE J. Oceanic Eng.
45
, 161
–173
(2020
).33.
G. R.
Potty
,
J. H.
Miller
, and
J. F.
Lynch
, “
Inversion for sediment geoacoustic properties at the new england bight
,” J. Acoust. Soc. Am.
114
(4
), 1874
–1887
(2003
).34.
J.-X.
Zhou
,
X.-Z.
Zhang
, and
D. P.
Knobles
, “
Low-frequency geoacoustic model for the effective properties of sandy seabottoms
,” J. Acoust. Soc. Am.
125
(5
), 2847
–2866
(2009
).35.
C. M.
Bishop
, Neural Networks for Pattern Recognition
(
Oxford University Press
,
Oxford, UK
, 1995
).36.
A.
Krizhevsky
,
I.
Sutskever
, and
G. E.
Hinton
, “
Imagenet classification with deep convolutional neural networks
,” in Advances in Neural Information Processing Systems
(2012
), pp. 1097
–1105
).37.
J.
Deng
,
W.
Dong
,
R.
Socher
,
L.-J.
Li
,
K.
Li
, and
L.
Fei-Fei
, “
Imagenet: A large-scale hierarchical image database
,” in 2009 IEEE Conference on Computer Vision and Pattern Recognition
IEEE, New York
, 2009
), pp. 248
–255
.38.
A.
Paszke
,
S.
Gross
,
F.
Massa
,
A.
Lerer
,
J.
Bradbury
,
G.
Chanan
,
T.
Killeen
,
Z.
Lin
,
N.
Gimelshein
, and
L.
Antiga
, “
Pytorch: An imperative style, high-performance deep learning library
,” in Advances in Neural Information Processing Systems
(2019
), pp. 8024
–8035
.39.
40.
D. F.
Van Komen
,
T. B.
Neilsen
,
D. P.
Knobles
, and
M.
Badiey
, “
A convolutional neural network for source range and ocean seabed classification using pressure time-series
,” Proc. Mtgs. Acoust.
36
(1
), 070004
(2019
).41.
I.
Loshchilov
and
F.
Hutter
, “
Sgdr: Stochastic gradient descent with warm restarts
,” arXiv:1608.03983 (2016
).42.
T. B.
Neilsen
,
C. D.
Escobar
,
M. C.
Acree
,
W. S.
Hodgkiss
,
D. F.
Van Komen
,
D. P.
Knobles
,
M.
Badiey
, and
J.
Castro
, “
Learning source location and seabed type from towed mid-frequency tonals on a vertical line array
,” J. Acoust. Soc. Am.
149
(1
), 692
–705
(2021
).43.
A.
Kendall
,
Y.
Gal
, and
R.
Cipolla
, “
Multi-task learning using uncertainty to weigh losses for scene geometry and semantics
,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition
(2018
), pp. 7482
–7491
.44.
This suggestion was given by Tony-Y on the
https://discuss.pytorch.org/t/how-to-learn-the-weights-between-two-losses/39681/12 PyTorch online forums (Last viewed May 19, 2020).45.
R.
Kohavi
, “
A study of cross-validation and bootstrap for accuracy estimation and model selection
,” in IJCAI
, Montreal, Canada
(1995
), Vol.
14
, pp. 1137
–1145
.© 2021 Acoustical Society of America.
2021
Acoustical Society of America
You do not currently have access to this content.
