An automated approach to passive sonar classification using binary image features

Vahidpour, Vahid; Rastegarnia, Amir; Khalili, Azam

doi:10.1007/s11804-015-1312-z

Cited by 10 publications

(5 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The acoustic signal produced by a vessel moving in the sea is mainly composed of a broadband component (with a continuous spectrum), that is generated by the propeller and its hydrodynamic interactions; and a narrow band component (whose spectrum consists of line components at discrete frequencies), owing to the propulsion system and other mechanical parts [ 19 ]. The automatic classification of this type of signal is a challenging task, as the signal is also dependent on the vessel’s speed, the age and state of the propulsion system, the highly variable background noise and the diversity of sound propagation mechanisms in the ocean.…”

Section: Background Knowledgementioning

confidence: 99%

“…A fully connected neural network was applied to recognise ships in hydrophone data [ 19 ]. In order to do this, the original signal was treated by traditional image-processing methods (mainly a short-time Fourier transform) to generate a binary image, based on the frequency spectrum of signal segmentation.…”

Section: Machine Learning For the Classification Of Underwater Acoust...mentioning

confidence: 99%

See 1 more Smart Citation

A Survey of Underwater Acoustic Data Classification Methods Using Deep Learning for Shoreline Surveillance

Domingos

Santos

Skelton

et al. 2022

Sensors

View full text Add to dashboard Cite

This paper presents a comprehensive overview of current deep-learning methods for automatic object classification of underwater sonar data for shoreline surveillance, concentrating mostly on the classification of vessels from passive sonar data and the identification of objects of interest from active sonar (such as minelike objects, human figures or debris of wrecked ships). Not only is the contribution of this work to provide a systematic description of the state of the art of this field, but also to identify five main ingredients in its current development: the application of deep-learning methods using convolutional layers alone; deep-learning methods that apply biologically inspired feature-extraction filters as a preprocessing step; classification of data from frequency and time–frequency analysis; methods using machine learning to extract features from original signals; and transfer learning methods. This paper also describes some of the most important datasets cited in the literature and discusses data-augmentation techniques. The latter are used for coping with the scarcity of annotated sonar datasets from real maritime missions.

show abstract

Section: Background Knowledgementioning

confidence: 99%

Section: Machine Learning For the Classification Of Underwater Acoust...mentioning

confidence: 99%

A Survey of Underwater Acoustic Data Classification Methods Using Deep Learning for Shoreline Surveillance

Domingos

Santos

Skelton

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Usually, this sound is produced by the set of mechanical components in the vessel's propulsion system, such as its engine, as well as by hydrodynamic interactions of the propeller. The former typically produces a broadband continuous spectrum, while the latter generates narrow band components whose spectrum consists of power at discrete frequencies [26]. As there are different types of vessels, in diverse states of upkeep, the sound produced by them is fundamentally distinct from one another, depending on the vessel's speed, the state of its mechanical parts, and the hydrodynamics of its design.…”

Section: Related Workmentioning

confidence: 99%

An Investigation of Preprocessing Filters and Deep Learning Methods for Vessel Type Classification With Underwater Acoustic Data

et al. 2022

View full text Add to dashboard Cite

The illegal exploitation of protected marine environments has consistently threatened the biodiversity and economic development of coastal regions. Extensive monitoring in these -often remoteareas is challenging. Machine learning methods are useful in object detection and classification tasks and have the potential to underpin techniques for the development of robust monitoring systems to overcome this problem. However, development is hindered due to the limited number of publicly available labelled and curated datasets. Furthermore, there are relatively few open-source state-of-the-art methods to be used for evaluation. This paper presents an investigation of automated classification methods using underwater acoustic signals to infer the presence and type of vessels navigating in coastal regions. Various combinations of deep convolutional neural network architectures, and preprocessing filter layers, were evaluated using a new dataset based on a subset of the extensive open-source Ocean Networks Canada hydrophone data. Tests were conducted in which VGGNet and ResNet networks were applied to classify the input data. The data was preprocessed using either Constant Q Transform (CQT), Gammatone, Mel spectrogram, or a combination of these filters. With over 97% accuracy, using all three preprocessing representations simultaneously yielded the most reliable result. However, high accuracies of 94.95% were achieved using CQT as the preprocessing filter for a ResNet-based convolutional neural network, providing a trade-off between model complexity and accuracy; a result that is more than 10% higher than previously reported approaches. This more accurate classifier for underwater acoustics could be used as a reliable autonomous monitoring system in maritime environments.INDEX TERMS deep learning, hydrophones, marine environment, ship type, sound

show abstract

“…The acoustic signal produced by a vessel moving in the sea is mainly composed of a broadband component, which has a continuous spectrum, generated by the propeller and its hydrodynamic interactions. It is also comprised of a narrow band component, whose spectrum consists of line components at discrete frequencies, owing to the propulsion system and other mechanical parts (VAHIDPOUR; RASTEGARNIA; KHALILI, 2015). The automatic classification of this type of signal is a challenging task as the acoustic signal is also dependent on the vessel's speed, the age and state of the propulsion system, the highly variable background noise and the diversity of sound propagation mechanisms in the ocean.…”

Section: Fundamentals Of Underwater Acousticsmentioning

confidence: 99%

“…A fully-connected neural network was applied to recognise ships in hydrophone data (VAHIDPOUR; RASTEGARNIA; KHALILI, 2015). In order to do this, the origi- 2).…”

Section: Classification Using Frequency and Time-frequency Analysismentioning

confidence: 99%

Machine learning methods for vessel type classification with underwater acoustic data

Domingos¹

View full text Add to dashboard Cite

RESUMOA identificação de embarcações em ambientes de tráfego controlado pode ser benéfica para manutenção da biodiversidade e proteção dos ambientes costeiros de regiões protegidas, gerando contribuições para a comunidade local e para o ecossistema. Nesse âmbito, vê-se latente a necessidade de melhores técnicas de identificação e classificação de embarcações, proporcionando mecanismos para melhora destes sistemas. Sinais sonoros subaquáticos são mais difíceis de serem mascarados ou omitidos durante a navegação de uma embarcação quando comparados com outras fontes de dados, proporcionando uma fonte confiável e resistente a fraudes para sistemas de classificação, porém, estes sofrem interferências das condições do meio em que se encontram. Neste trabalho, uma metodologia foi proposta para realizar a classificação de sinais sonoros subaquáticos provenientes de embarcações utilizando técnicas de aprendizado de máquina, considerando também as variáveis ambientais, como a distância entre os hidrofones e as embarcações. Uma comparação relativa à performance das redes neurais convolucionais mais comuns foi realizada utilizando a arquitetura da VGG e da ResNet 18. Também foram realizadas comparações entre os três filtros de pré-processamentos comumente presentes na literatura, os espectrogramas Mel, os filtros Gamma, e a transformada de constante Q, proporcionando um estudo sobre o impacto de tais variáveis na classificação final. Devido a escassez de conjuntos de dados anotados para estudo deste problema, um conjunto de dados anotados foi proposto utilizando como base os sinais sonoros da iniciativa Ocean Canada Network. Os resultados obtidos atingiram a acurácia de 94.95% no conjunto de dados proposto usando CQT como filtro de pré-processamento para uma rede neural convolucional baseada na ResNet. Os códigos fontes para reprodução dos testes, assim como para obtenção do dataset, estão disponibilizados de maneira gratuita e pública para fins acadêmicos.

show abstract

An automated approach to passive sonar classification using binary image features

Cited by 10 publications

References 21 publications

A Survey of Underwater Acoustic Data Classification Methods Using Deep Learning for Shoreline Surveillance

A Survey of Underwater Acoustic Data Classification Methods Using Deep Learning for Shoreline Surveillance

An Investigation of Preprocessing Filters and Deep Learning Methods for Vessel Type Classification With Underwater Acoustic Data

Machine learning methods for vessel type classification with underwater acoustic data

Contact Info

Product

Resources

About