GLORIA

GEOMAR Library Ocean Research Information Access

X

Your email was sent successfully. Check your inbox.

An error occurred while sending the email. Please try again.

Proceed reservation?

Export
Filter
  • Comparative Studies. Non-European Languages/Literatures  (4)
Material
Publisher
Person/Organisation
Language
Years
FID
Subjects(RVK)
  • Comparative Studies. Non-European Languages/Literatures  (4)
RVK
  • 1
    Online Resource
    Online Resource
    Automated classification of Tursiops aduncus whistles based on a depth-wise separable convolutional neural network and data augmentation
    Acoustical Society of America (ASA) ; 2021
    In:  The Journal of the Acoustical Society of America Vol. 150, No. 5 ( 2021-11-01), p. 3861-3873
    Details
    In: The Journal of the Acoustical Society of America, Acoustical Society of America (ASA), Vol. 150, No. 5 ( 2021-11-01), p. 3861-3873
    Abstract: Whistle classification plays an essential role in studying the habitat and social behaviours of cetaceans. We obtained six categories of sweep whistles of two Tursiops aduncus individual signals using the passive acoustic mornitoring technique over a period of eight months in the Xiamen area. First, we propose a depthwise separable convolutional neural network for whistle classification. The proposed model adopts the depthwise convolution combined with the followed point-by-point convolution instead of the conventional convolution. As a result, it brings a better classification performance in sample sets with relatively independent features between different channels. Meanwhile, it leads to less computational complexity and fewer model parameters. Second, in order to solve the problem of an imbalance in the number of samples under each whistle category, we propose a random series method with five audio augmentation algorithms. The generalization ability of the trained model was improved by using an opening probability for each algorithm and the random selection of each augmentation factor within specific ranges. Finally, we explore the effect of the proposed augmentation method on the performance of our proposed architecture and find that it enhances the accuracy up to 98.53% for the classification of Tursiops aduncus whistles.
    Type of Medium: Online Resource
    ISSN: 0001-4966 , 1520-8524
    URL: Article
    DOI: 10.1121/10.0007291
    RVK:
    EQ 1000
    Language: English
    Publisher: Acoustical Society of America (ASA)
    Publication Date: 2021
    detail.hit.zdb_id: 1461063-2
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
    Online Resource
  • 2
    Online Resource
    Online Resource
    Robust unsupervised Tursiops aduncus whistle-event detection using gammatone multi-channel Savitzky–Golay based whistle enhancement
    Acoustical Society of America (ASA) ; 2022
    In:  The Journal of the Acoustical Society of America Vol. 151, No. 5 ( 2022-05-01), p. 3509-3521
    Details
    In: The Journal of the Acoustical Society of America, Acoustical Society of America (ASA), Vol. 151, No. 5 ( 2022-05-01), p. 3509-3521
    Abstract: Detecting whistle events is essential when studying the population density and behavior of cetaceans. After eight months of passive acoustic monitoring in Xiamen, we obtained long calls from two Tursiops aduncus individuals. In this paper, we propose an algorithm with an unbiased gammatone multi-channel Savitzky–Golay for smoothing dynamic continuous background noise and interference from long click trains. The algorithm uses the method of least squares to perform a local polynomial regression on the time–frequency representation of multi-frequency resolution call measurements, which can effectively retain the whistle profiles while filtering out noise and interference. We prove that it is better at separating out whistles and has lower computational complexity than other smoothing methods. In order to further extract whistle features in enhanced spectrograms, we also propose a set of multi-scale and multi-directional moving filter banks for various whistle durations and contour shapes. The final binary adaptive decisions at frame level for whistle events are obtained from the histograms of multi-scale and multi-directional spectrograms. Finally, we explore the entire data set and find that the proposed scheme achieves the highest frame-level F1-scores when detecting T. aduncus whistles than the baseline schemes, with an improvement of more than 6%.
    Type of Medium: Online Resource
    ISSN: 0001-4966 , 1520-8524
    URL: Article
    DOI: 10.1121/10.0011402
    RVK:
    EQ 1000
    Language: English
    Publisher: Acoustical Society of America (ASA)
    Publication Date: 2022
    detail.hit.zdb_id: 1461063-2
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
    Online Resource
  • 3
    Online Resource
    Online Resource
    Three-dimensional finite element simulation of acoustic propagation in spiral bubble net of humpback whale
    Acoustical Society of America (ASA) ; 2019
    In:  The Journal of the Acoustical Society of America Vol. 146, No. 3 ( 2019-09-01), p. 1982-1995
    Details
    In: The Journal of the Acoustical Society of America, Acoustical Society of America (ASA), Vol. 146, No. 3 ( 2019-09-01), p. 1982-1995
    Abstract: In 2004, Leighton hypothesized that the acoustic calls emitted by humpback whales when feeding using bubble nets, may enhance the effectiveness of the net in confining prey (such as herring) by forming a “wall of sound” with a quiet zone within. Modelling of the acoustics of this phenomenon was previously restricted to 2D; this paper conducts a 3D model of the propagation of signals resembling those emitted by humpback whales when bubble netting, projected into an upward spiral bubble net which data to date suggest is the accurate form for the bubble net in 3D space. In this study, the feeding calls were analyzed in the time-frequency domain to extract acoustic information sufficient to allow modeling of the resulting spatial distribution of acoustic pressure and particle velocity, and how they vary over the duration of the call. Sound propagation in the bubble net was described by using a linear steady-state formulation for an effective medium of bubbly water. Using the predicted attenuation, phase velocity and density in bubbly water, a 3D finite element model was constructed to numerically simulate the upward-spiral bubble net which consists of a mixture of bubbles that exhibit a range of radii. The acoustic pressure field and particle motion field were both calculated within the bubble net. The simulation results show that the energy of the whale feeding call could be effectively focused in the bubble net, generating intensive sound pressure and particle motion fields in the bubbly arm of the net, but with some “quiet” regions closer to the center of the net, as Leighton hypothesized. Furthermore, when the hearing ability of herring is taken into consideration, the results suggest that this acoustic focusing effect could be a plausible factor in trapping them in the bubble net. It also allows speculation on the possible enhancements that the time-varying nature of the call during feeding could give to the whale in this mechanism for the bubble net feeding by humpback whales.
    Type of Medium: Online Resource
    ISSN: 0001-4966 , 1520-8524
    URL: Article
    DOI: 10.1121/1.5126003
    RVK:
    EQ 1000
    Language: English
    Publisher: Acoustical Society of America (ASA)
    Publication Date: 2019
    detail.hit.zdb_id: 1461063-2
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
    Online Resource
  • 4
    Online Resource
    Online Resource
    Elastic feature of cylindrical shells extraction in time-frequency domain using biomimetic dolphin click
    Acoustical Society of America (ASA) ; 2017
    In:  The Journal of the Acoustical Society of America Vol. 142, No. 6 ( 2017-12-01), p. 3787-3795
    Details
    In: The Journal of the Acoustical Society of America, Acoustical Society of America (ASA), Vol. 142, No. 6 ( 2017-12-01), p. 3787-3795
    Abstract: A dolphin's biosonar may effectively discriminate subtle differences among targets. In order to investigate the possible physical mechanism of target discrimination, in this study, a finite element model excited by a biomimetic click pulse was proposed. The acoustic scattering field and stress distribution of a stainless steel shell were simulated. The biomimetic click experiments were then conducted to verify the theoretical predictions in an anechoic tank. The experimental results showed a good agreement with the model simulations. Furthermore, the elastic time-frequency features of three cylindrical shells with different wall thickness were obtained using a fractional Fourier transform filter to eliminate specular reflection and cross-term interference. To compare discrimination capacity of the time-frequency features with and without the specular reflection, a time-frequency correlator was applied to calculate the correlation coefficient between different shells. The results indicated that the time-frequency features can be represented in high resolution with less cross-term interference, and these features without specular reflection showed a good capacity to discriminate the shells with different wall thickness.
    Type of Medium: Online Resource
    ISSN: 0001-4966 , 1520-8524
    URL: Article
    DOI: 10.1121/1.5017835
    RVK:
    EQ 1000
    Language: English
    Publisher: Acoustical Society of America (ASA)
    Publication Date: 2017
    detail.hit.zdb_id: 1461063-2
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
    Online Resource
Close ⊗
This website uses cookies and the analysis tool Matomo. More information can be found here...