Target Recognition Method of Sonar Image Based on Deep Learning

Renjie Gao; Yongsheng Yan; Xue Liu

doi:10.1109/ICSPCC59353.2023.10400377

Target Recognition Method of Sonar Image Based on Deep Learning

Renjie Gao
, Yongsheng Yan
, Xue Liu

School of Marine Science and Technology

Northwestern Polytechnical University Xian

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

The target detection and recognition technology of sonar images plays an important role in the field of marine environment monitoring. The traditional method mainly uses CNN to study sonar image recognition. However, the lack of data volume and the limitations of the CNN network itself reduce the accuracy of sonar image classification. To address the above issues, this paper first uses the DCGAN network for data augmentation. Data expansion generates fake images based on the collection of public data sets from the network and completes the expansion of the data set. Secondly, this paper uses ResNet network and DenseNet network to replace the traditional CNN network and uses focal loss to replace the traditional cross-entropy loss function. The model has a good classification effect for the data set in this paper. The classification accuracy of the improved ResNet network is 77%, and the classification accuracy of the improved DenseNet network is 84.1%.

Original language	English
Title of host publication	Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798350316728
DOIs	https://doi.org/10.1109/ICSPCC59353.2023.10400377
State	Published - 2023
Event	2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023 - Zhengzhou, Henan, China Duration: 14 Nov 2023 → 17 Nov 2023

Publication series

Name	Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023

Conference

Conference	2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023
Country/Territory	China
City	Zhengzhou, Henan
Period	14/11/23 → 17/11/23

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 14 Life Below Water

Keywords

deep learning
generative adversarial nets
residual neural network
sonar image

Access to Document

10.1109/ICSPCC59353.2023.10400377

Cite this

Gao, R., Yan, Y., & Liu, X. (2023). Target Recognition Method of Sonar Image Based on Deep Learning. In Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023 (Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICSPCC59353.2023.10400377

Gao, Renjie ; Yan, Yongsheng ; Liu, Xue. / Target Recognition Method of Sonar Image Based on Deep Learning. Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023. Institute of Electrical and Electronics Engineers Inc., 2023. (Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023).

@inproceedings{a3734dab430044fd9213dd785e909ddf,

title = "Target Recognition Method of Sonar Image Based on Deep Learning",

abstract = "The target detection and recognition technology of sonar images plays an important role in the field of marine environment monitoring. The traditional method mainly uses CNN to study sonar image recognition. However, the lack of data volume and the limitations of the CNN network itself reduce the accuracy of sonar image classification. To address the above issues, this paper first uses the DCGAN network for data augmentation. Data expansion generates fake images based on the collection of public data sets from the network and completes the expansion of the data set. Secondly, this paper uses ResNet network and DenseNet network to replace the traditional CNN network and uses focal loss to replace the traditional cross-entropy loss function. The model has a good classification effect for the data set in this paper. The classification accuracy of the improved ResNet network is 77\%, and the classification accuracy of the improved DenseNet network is 84.1\%.",

keywords = "deep learning, generative adversarial nets, residual neural network, sonar image",

author = "Renjie Gao and Yongsheng Yan and Xue Liu",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023 ; Conference date: 14-11-2023 Through 17-11-2023",

year = "2023",

doi = "10.1109/ICSPCC59353.2023.10400377",

language = "英语",

series = "Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023",

}

Gao, R, Yan, Y & Liu, X 2023, Target Recognition Method of Sonar Image Based on Deep Learning. in Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023. Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023, Institute of Electrical and Electronics Engineers Inc., 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023, Zhengzhou, Henan, China, 14/11/23. https://doi.org/10.1109/ICSPCC59353.2023.10400377

Target Recognition Method of Sonar Image Based on Deep Learning. / Gao, Renjie; Yan, Yongsheng; Liu, Xue.
Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023. Institute of Electrical and Electronics Engineers Inc., 2023. (Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Target Recognition Method of Sonar Image Based on Deep Learning

AU - Gao, Renjie

AU - Yan, Yongsheng

AU - Liu, Xue

PY - 2023

Y1 - 2023

N2 - The target detection and recognition technology of sonar images plays an important role in the field of marine environment monitoring. The traditional method mainly uses CNN to study sonar image recognition. However, the lack of data volume and the limitations of the CNN network itself reduce the accuracy of sonar image classification. To address the above issues, this paper first uses the DCGAN network for data augmentation. Data expansion generates fake images based on the collection of public data sets from the network and completes the expansion of the data set. Secondly, this paper uses ResNet network and DenseNet network to replace the traditional CNN network and uses focal loss to replace the traditional cross-entropy loss function. The model has a good classification effect for the data set in this paper. The classification accuracy of the improved ResNet network is 77%, and the classification accuracy of the improved DenseNet network is 84.1%.

AB - The target detection and recognition technology of sonar images plays an important role in the field of marine environment monitoring. The traditional method mainly uses CNN to study sonar image recognition. However, the lack of data volume and the limitations of the CNN network itself reduce the accuracy of sonar image classification. To address the above issues, this paper first uses the DCGAN network for data augmentation. Data expansion generates fake images based on the collection of public data sets from the network and completes the expansion of the data set. Secondly, this paper uses ResNet network and DenseNet network to replace the traditional CNN network and uses focal loss to replace the traditional cross-entropy loss function. The model has a good classification effect for the data set in this paper. The classification accuracy of the improved ResNet network is 77%, and the classification accuracy of the improved DenseNet network is 84.1%.

KW - deep learning

KW - generative adversarial nets

KW - residual neural network

KW - sonar image

UR - https://www.scopus.com/pages/publications/85184851150

U2 - 10.1109/ICSPCC59353.2023.10400377

DO - 10.1109/ICSPCC59353.2023.10400377

M3 - 会议稿件

AN - SCOPUS:85184851150

T3 - Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023

BT - Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023

Y2 - 14 November 2023 through 17 November 2023

ER -

Gao R, Yan Y, Liu X. Target Recognition Method of Sonar Image Based on Deep Learning. In Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023. Institute of Electrical and Electronics Engineers Inc. 2023. (Proceedings of 2023 IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2023). doi: 10.1109/ICSPCC59353.2023.10400377

Target Recognition Method of Sonar Image Based on Deep Learning

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this