Classification of odontocete echolocation clicks using convolutional neural network

Luo, Wenyu

doi:10.1121/10.0000514

Cited by 12 publications

(7 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In contrast to our work which encompasses a large spectral input for the soundscape as a whole, Ibrahim et al (2020) focuses on the frequency range 10 -400 Hz, narrowing the focus of the spectral input to the signals of interest. As the bandwidth of interest (Liu et al, 2018) for echolocation clicks, 93% for both whistles and clicks of a single species (Bergler et al, 2019) and outperform existing general mixed model efforts (Roch et al, 2011a) achieving high accuracies at multi-species click classification (Yang et al, 2020), making use of larger architectures and labelled training sets. Using an open-source 'light-weight' architecture and a small annotated training set we demonstrate similar overall accuracies of our model and in-depth exploration of seasonal variation on model performance to present researchers with an insight into the reliability of CNNs across annual cycles.…”

Section: Discussionmentioning

confidence: 99%

“…CNNs learn to discriminate spectrotemporal information directly from a labelled spectrogram used as an image input, removing the dependence on human experts for manual feature extraction, and improving the robustness to variation in signal structure, caller distance and signal-to-noiseratio (SNR) conditions (Gibb et al, 2019). The success of CNNs has been demonstrated by many studies in the marine domain for binary species detection and multi-class species classification (Belgith et al, 2018;Harvey, 2018;Liu et al, 2018;Bergler et al, 2019;Bermant et al, 2019;Shiu et al, 2020;Yang et al, 2020;Zhong et al, 2020;Allen et al, 2021) advancing the capabilities of mining large PAM datasets for detecting species of interest. Existing work tends to make use of spectrogram representations across a limited bandwidth, which is selected according to the species (or signal) of interest.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

More than a whistle: Automated detection of marine sound sources with a convolutional neural network

White

Bull

et al. 2022

Front. Mar. Sci.

View full text Add to dashboard Cite

The effective analysis of Passive Acoustic Monitoring (PAM) data has the potential to determine spatial and temporal variations in ecosystem health and species presence if automated detection and classification algorithms are capable of discrimination between marine species and the presence of anthropogenic and environmental noise. Extracting more than a single sound source or call type will enrich our understanding of the interaction between biological, anthropogenic and geophonic soundscape components in the marine environment. Advances in extracting ecologically valuable cues from the marine environment, embedded within the soundscape, are limited by the time required for manual analyses and the accuracy of existing algorithms when applied to large PAM datasets. In this work, a deep learning model is trained for multi-class marine sound source detection using cloud computing to explore its utility for extracting sound sources for use in marine mammal conservation and ecosystem monitoring. A training set is developed comprising existing datasets amalgamated across geographic, temporal and spatial scales, collected across a range of acoustic platforms. Transfer learning is used to fine-tune an open-source state-of-the-art ‘small-scale’ convolutional neural network (CNN) to detect odontocete tonal and broadband call types and vessel noise (from 0 to 48 kHz). The developed CNN architecture uses a custom image input to exploit the differences in temporal and frequency characteristics between each sound source. Each sound source is identified with high accuracy across various test conditions, including variable signal-to-noise-ratio. We evaluate the effect of ambient noise on detector performance, outlining the importance of understanding the variability of the regional soundscape for which it will be deployed. Our work provides a computationally low-cost, efficient framework for mining big marine acoustic data, for information on temporal scales relevant to the management of marine protected areas and the conservation of vulnerable species.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

More than a whistle: Automated detection of marine sound sources with a convolutional neural network

White

Bull

et al. 2022

Front. Mar. Sci.

View full text Add to dashboard Cite

show abstract

“…Recent studies in acoustics also showed that CNNs work well with acoustic data to identify and classify sounds in both time-(1D) and time-frequency domains (2D). In these works, the marine mammal species were classified using the sounds produced by animals using CNNs [21,22]. Our goal here is to solve a related regression problem by training CNNs directly on the time-domain echo data to learn the edge-diffraction induced patterns of temporal echoes (see Figure 1B) and map them to the material parameter tuple (ρ, K, G).…”

Section: Materials Parameter Retrieval Using Cnnsmentioning

confidence: 99%

Metamaterial characterization from far-field acoustic wave measurements using convolutional neural network

Cheong

Kwon²,

Popa³

2022

Front. Phys.

View full text Add to dashboard Cite

Identifying the material properties of unknown media is an important scientific/engineering challenge in areas as varied as in-vivo tissue health diagnostics and metamaterial characterization. Currently, techniques exist to retrieve the material parameters of large unknown media from elastic wave scattering in free-space using analytical or numerical methods. However, applying these methods to small samples on the order of few wavelengths in diameter is challenging, as the fields scattered by these samples become significantly contaminated by diffraction from the sample edges. Here, we propose a method to extract the material parameters of small samples using convolutional neural networks trained to learn the mapping between far-field echoes and the material parameters. Networks were trained with synthetic time domain echo data obtained by simulating the free-space scattering of sound from unknown media underwater. Results show that neural networks can accurately predict effective material parameters such as mass density, bulk modulus, and shear modulus even when small training sets are used. Furthermore, we demonstrate in experiments executed in a water tank that the networks trained with synthetic data can accurately estimate the material properties of fabricated metamaterial samples from single-point echo measurements performed in the far-field. This work highlights the effectiveness of our approach to identify unknown media using far-field acoustic reflection dominated by diffraction fields and will open a new avenue toward acoustic sensing techniques.

show abstract

“…In classifying animal sounds, deep neural network (DNN) methods have progressed tremendously with accessibility to large training data and increasing computational power. Using spectrograms generated from raw audio recordings as input, researchers have applied convolutional neural networks (CNN), either by training the model from scratch or using transfer learning with pre-trained model weights, to classify calls from different species (Bergler et al, 2019;Yang et al, 2020, Zhong et al, 2020, Kirsebom et al, 2020. Another approach is the use of recurrent neural networks (RNN), which utilize temporal information of animal calls for classification tasks (Ibrahim et al, 2018;Shiu et al, 2020).…”

Section: B Motivation For the Workmentioning

confidence: 99%

Detecting, classifying, and counting blue whale calls with Siamese neural networks

Zhong

Torterotot²,

Branch

et al. 2021

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

The goal of this project is to use acoustic signatures to detect, classify, and count the calls of four acoustic populations of blue whales so that, ultimately, the conservation status of each population can be better assessed. We used manual annotations from 350 h of audio recordings from the underwater hydrophones in the Indian Ocean to build a deep learning model to detect, classify, and count the calls from four acoustic song types. The method we used was Siamese neural networks (SNN), a class of neural network architectures that are used to find the similarity of the inputs by comparing their feature vectors, finding that they outperformed the more widely used convolutional neural networks (CNN). Specifically, the SNN outperform a CNN with 2% accuracy improvement in population classification and 1.7%-6.4% accuracy improvement in call count estimation for each blue whale population. In addition, even though we treat the call count estimation problem as a classification task and encode the number of calls in each spectrogram as a categorical variable, SNN surprisingly learned the ordinal relationship among them. SNN are robust and are shown here to be an effective way to automatically mine large acoustic datasets for blue whale calls. V

show abstract

Classification of odontocete echolocation clicks using convolutional neural network

Cited by 12 publications

References 28 publications

More than a whistle: Automated detection of marine sound sources with a convolutional neural network

More than a whistle: Automated detection of marine sound sources with a convolutional neural network

Metamaterial characterization from far-field acoustic wave measurements using convolutional neural network

Detecting, classifying, and counting blue whale calls with Siamese neural networks

Contact Info

Product

Resources

About