Humpback whale behavior, population distribution and structure can be inferred from long term underwater passive acoustic monitoring of their vocalizations. Here we develop automatic approaches for classifying humpback whale vocalizations into the two categories of song and non-song, employing machine learning techniques. The vocalization behavior of humpback whales was monitored over instantaneous vast areas of the Gulf of Maine using a large aperture coherent hydrophone array system via the passive ocean acoustic waveguide remote sensing technique over multiple diel cycles in Fall 2006. We use wavelet signal denoising and coherent array processing to enhance the signal-to-noise ratio. To build features vector for every time sequence of the beamformed signals, we employ Bag of Words approach to time-frequency features. Finally, we apply Support Vector Machine (SVM), Neural Networks, and Naive Bayes to classify the acoustic data and compare their performances. Best results are obtained using Mel Frequency Cepstrum Coefficient (MFCC) features and SVM which leads to 94% accuracy and 72.73% F1-score for humpback whale song versus non-song vocalization classification, showing effectiveness of the proposed approach for real-time classification at sea.
more »
« less
Comparing Performances of Five Distinct Automatic Classifiers for Fin Whale Vocalizations in Beamformed Spectrograms of Coherent Hydrophone Array
A large variety of sound sources in the ocean, including biological, geophysical, and man-made, can be simultaneously monitored over instantaneous continental-shelf scale regions via the passive ocean acoustic waveguide remote sensing (POAWRS) technique by employing a large-aperture densely-populated coherent hydrophone array system. Millions of acoustic signals received on the POAWRS system per day can make it challenging to identify individual sound sources. An automated classification system is necessary to enable sound sources to be recognized. Here, the objectives are to (i) gather a large training and test data set of fin whale vocalization and other acoustic signal detections; (ii) build multiple fin whale vocalization classifiers, including a logistic regression, support vector machine (SVM), decision tree, convolutional neural network (CNN), and long short-term memory (LSTM) network; (iii) evaluate and compare performance of these classifiers using multiple metrics including accuracy, precision, recall and F1-score; and (iv) integrate one of the classifiers into the existing POAWRS array and signal processing software. The findings presented here will (1) provide an automatic classifier for near real-time fin whale vocalization detection and recognition, useful in marine mammal monitoring applications; and (2) lay the foundation for building an automatic classifier applied for near real-time detection and recognition of a wide variety of biological, geophysical, and man-made sound sources typically detected by the POAWRS system in the ocean.
more »
« less
- Award ID(s):
- 1736749
- PAR ID:
- 10198592
- Date Published:
- Journal Name:
- Remote Sensing
- Volume:
- 12
- Issue:
- 2
- ISSN:
- 2072-4292
- Page Range / eLocation ID:
- 326
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
Distributed acoustic sensing (DAS) is a technique that measures strain changes along an optical fiber to distances of ∼100 km with a spatial sensitivity of tens of meters. In November 2021, 4 days of DAS data were collected on two cables of the Ocean Observatories Initiative Regional Cabled Array extending offshore central Oregon. Numerous 20 Hz fin whale calls, northeast Pacific blue whale A and B calls, and ship noises were recorded, highlighting the potential of DAS for monitoring the ocean. The data are publicly available to support studies to understand the sensitivity of submarine DAS for low-frequency acoustic monitoring.
-
Multiple mechanized ocean vessels, including both surface ships and submerged vehicles, can be simultaneously monitored over instantaneous continental-shelf scale regions >10,000 km 2 via passive ocean acoustic waveguide remote sensing. A large-aperture densely-sampled coherent hydrophone array system is employed in the Norwegian Sea in Spring 2014 to provide directional sensing in 360 degree horizontal azimuth and to significantly enhance the signal-to-noise ratio (SNR) of ship-radiated underwater sound, which improves ship detection ranges by roughly two orders of magnitude over that of a single hydrophone. Here, 30 mechanized ocean vessels spanning ranges from nearby to over 150 km from the coherent hydrophone array, are detected, localized and classified. The vessels are comprised of 20 identified commercial ships and 10 unidentified vehicles present in 8 h/day of Passive Ocean Acoustic Waveguide Remote Sensing (POAWRS) observation for two days. The underwater sounds from each of these ocean vessels received by the coherent hydrophone array are dominated by narrowband signals that are either constant frequency tonals or have frequencies that waver or oscillate slightly in time. The estimated bearing-time trajectory of a sequence of detections obtained from coherent beamforming are employed to determine the horizontal location of each vessel using the Moving Array Triangulation (MAT) technique. For commercial ships present in the region, the estimated horizontal positions obtained from passive acoustic sensing are verified by Global Positioning System (GPS) measurements of the ship locations found in a historical Automatic Identification System (AIS) database. We provide time-frequency characterizations of the underwater sounds radiated from the commercial ships and the unidentified vessels. The time-frequency features along with the bearing-time trajectory of the detected signals are applied to simultaneously track and distinguish these vessels.more » « less
-
The source levels, SL, of Antarctic blue and fin whale calls were estimated using acoustic recordings collected from directional sonobuoys deployed during an Antarctic voyage in 2019. Antarctic blue whale call types included stereotyped song and downswept frequency-modulated calls, often, respectively, referred to as Z-calls (comprising song units-A, B, and C) and D-calls. Fin whale calls included 20 Hz pulses and 40 Hz downswept calls. Source levels were obtained by measuring received levels (RL) and modelling transmission losses (TL) for each detection. Estimates of SL were sensitive to the parameters used in TL models, particularly the seafloor geoacoustic properties and depth of the calling whale. For our best estimate of TL and whale-depth, mean SL in dB re 1 μPa ± 1 standard deviation ranged between 188–191 ± 6–8 dB for blue whale call types and 189–192 ± 6 dB for fin whale call types. These estimates of SL are the first from the Southern Hemisphere for D-calls and 40 Hz downsweeps, and the largest sample size to-date for Antarctic blue whale song. Knowledge of source levels is essential for estimating the detection range and communication space of these calls and will enable more accurate comparisons of detections of these sounds from sonobuoy surveys and across international long-term monitoring networks.more » « less
-
Airgun source systems generate low frequency underwater sound used in reflection and refraction seismology for mapping ocean bottom stratigraphy with important applications in ocean geosciences, such as understanding plate tectonics, ascertaining ocean geological history and climate change, and offshore hydrocarbon prospecting. Seismo-acoustic airgun signals from geophysical surveying activity were recorded at very long ranges, spanning roughly 175-195 km, on a large-aperture densely-populated linear coherent hydrophone array in the Norwegian Sea during Spring 2014. Off the coast of Alesund, airgun signals were detected with 8 s inter-pulse intervals for 3 to 24 hour time periods per day over the 4 days of hydrophone array operation in that region. Here we provide a time-frequency characterization and bearing-time estimation of the received airgun pulses. By correcting for transmission losses in the range- and depth-dependent Norwegian Sea environment, we estimate the source level distribution back projected to a distance of 1 m from the airgun source system. This back-projected source level distribution is then applied to model the Probability of Detection (PoD) region for the airgun signals with the coherent hydrophone array as the receiver in the Norwegian Sea employing the passive ocean acoustic waveguide remote sensing (POAWRS) technique. The estimates of back-projected source level distribution and PoD region provide an understanding of the horizontal spatial propagation extent of the signals from the airgun source system in the shallow and deep water regions of the Norwegian Sea. These results can also be applied to studies of the potential impact of airgun signals on marine organisms.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.3390/rs12020326
