Cargando…

Deep Learning Methods for Underwater Target Feature Extraction and Recognition

The classification and recognition technology of underwater acoustic signal were always an important research content in the field of underwater acoustic signal processing. Currently, wavelet transform, Hilbert-Huang transform, and Mel frequency cepstral coefficients are used as a method of underwat...

Descripción completa

Detalles Bibliográficos
Autores principales: Hu, Gang, Wang, Kejun, Peng, Yuan, Qiu, Mengran, Shi, Jianfei, Liu, Liangliang
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Hindawi 2018
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5892262/
https://www.ncbi.nlm.nih.gov/pubmed/29780407
http://dx.doi.org/10.1155/2018/1214301
_version_ 1783313140387151872
author Hu, Gang
Wang, Kejun
Peng, Yuan
Qiu, Mengran
Shi, Jianfei
Liu, Liangliang
author_facet Hu, Gang
Wang, Kejun
Peng, Yuan
Qiu, Mengran
Shi, Jianfei
Liu, Liangliang
author_sort Hu, Gang
collection PubMed
description The classification and recognition technology of underwater acoustic signal were always an important research content in the field of underwater acoustic signal processing. Currently, wavelet transform, Hilbert-Huang transform, and Mel frequency cepstral coefficients are used as a method of underwater acoustic signal feature extraction. In this paper, a method for feature extraction and identification of underwater noise data based on CNN and ELM is proposed. An automatic feature extraction method of underwater acoustic signals is proposed using depth convolution network. An underwater target recognition classifier is based on extreme learning machine. Although convolution neural networks can execute both feature extraction and classification, their function mainly relies on a full connection layer, which is trained by gradient descent-based; the generalization ability is limited and suboptimal, so an extreme learning machine (ELM) was used in classification stage. Firstly, CNN learns deep and robust features, followed by the removing of the fully connected layers. Then ELM fed with the CNN features is used as the classifier to conduct an excellent classification. Experiments on the actual data set of civil ships obtained 93.04% recognition rate; compared to the traditional Mel frequency cepstral coefficients and Hilbert-Huang feature, recognition rate greatly improved.
format Online
Article
Text
id pubmed-5892262
institution National Center for Biotechnology Information
language English
publishDate 2018
publisher Hindawi
record_format MEDLINE/PubMed
spelling pubmed-58922622018-05-20 Deep Learning Methods for Underwater Target Feature Extraction and Recognition Hu, Gang Wang, Kejun Peng, Yuan Qiu, Mengran Shi, Jianfei Liu, Liangliang Comput Intell Neurosci Research Article The classification and recognition technology of underwater acoustic signal were always an important research content in the field of underwater acoustic signal processing. Currently, wavelet transform, Hilbert-Huang transform, and Mel frequency cepstral coefficients are used as a method of underwater acoustic signal feature extraction. In this paper, a method for feature extraction and identification of underwater noise data based on CNN and ELM is proposed. An automatic feature extraction method of underwater acoustic signals is proposed using depth convolution network. An underwater target recognition classifier is based on extreme learning machine. Although convolution neural networks can execute both feature extraction and classification, their function mainly relies on a full connection layer, which is trained by gradient descent-based; the generalization ability is limited and suboptimal, so an extreme learning machine (ELM) was used in classification stage. Firstly, CNN learns deep and robust features, followed by the removing of the fully connected layers. Then ELM fed with the CNN features is used as the classifier to conduct an excellent classification. Experiments on the actual data set of civil ships obtained 93.04% recognition rate; compared to the traditional Mel frequency cepstral coefficients and Hilbert-Huang feature, recognition rate greatly improved. Hindawi 2018-03-27 /pmc/articles/PMC5892262/ /pubmed/29780407 http://dx.doi.org/10.1155/2018/1214301 Text en Copyright © 2018 Gang Hu et al. https://creativecommons.org/licenses/by/4.0/ This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
spellingShingle Research Article
Hu, Gang
Wang, Kejun
Peng, Yuan
Qiu, Mengran
Shi, Jianfei
Liu, Liangliang
Deep Learning Methods for Underwater Target Feature Extraction and Recognition
title Deep Learning Methods for Underwater Target Feature Extraction and Recognition
title_full Deep Learning Methods for Underwater Target Feature Extraction and Recognition
title_fullStr Deep Learning Methods for Underwater Target Feature Extraction and Recognition
title_full_unstemmed Deep Learning Methods for Underwater Target Feature Extraction and Recognition
title_short Deep Learning Methods for Underwater Target Feature Extraction and Recognition
title_sort deep learning methods for underwater target feature extraction and recognition
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5892262/
https://www.ncbi.nlm.nih.gov/pubmed/29780407
http://dx.doi.org/10.1155/2018/1214301
work_keys_str_mv AT hugang deeplearningmethodsforunderwatertargetfeatureextractionandrecognition
AT wangkejun deeplearningmethodsforunderwatertargetfeatureextractionandrecognition
AT pengyuan deeplearningmethodsforunderwatertargetfeatureextractionandrecognition
AT qiumengran deeplearningmethodsforunderwatertargetfeatureextractionandrecognition
AT shijianfei deeplearningmethodsforunderwatertargetfeatureextractionandrecognition
AT liuliangliang deeplearningmethodsforunderwatertargetfeatureextractionandrecognition