Deep Learning Based Underwater Acoustic Target Recognition: Introduce a Recent Temporal 2D Modeling Method
In recent years, the application of deep learning models for underwater target recognition has become a popular trend. Most of these are pure 1D models used for processing time-domain signals or pure 2D models used for processing time-frequency spectra. In this paper, a recent temporal 2D modeling m...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/24/5/1633 |
| _version_ | 1797263795854245888 |
|---|---|
| author | Jun Tang Wenbo Gao Enxue Ma Xinmiao Sun Jinying Ma |
| author_facet | Jun Tang Wenbo Gao Enxue Ma Xinmiao Sun Jinying Ma |
| author_sort | Jun Tang |
| collection | DOAJ |
| description | In recent years, the application of deep learning models for underwater target recognition has become a popular trend. Most of these are pure 1D models used for processing time-domain signals or pure 2D models used for processing time-frequency spectra. In this paper, a recent temporal 2D modeling method is introduced into the construction of ship radiation noise classification models, combining 1D and 2D. This method is based on the periodic characteristics of time-domain signals, shaping them into 2D signals and discovering long-term correlations between sampling points through 2D convolution to compensate for the limitations of 1D convolution. Integrating this method with the current state-of-the-art model structure and using samples from the Deepship database for network training and testing, it was found that this method could further improve the accuracy (0.9%) and reduce the parameter count (30%), providing a new option for model construction and optimization. Meanwhile, the effectiveness of training models using time-domain signals or time-frequency representations has been compared, finding that the model based on time-domain signals is more sensitive and has a smaller storage footprint (reduced to 30%), whereas the model based on time-frequency representation can achieve higher accuracy (1–2%). |
| first_indexed | 2024-04-25T00:18:41Z |
| format | Article |
| id | doaj.art-3dae04b9896b40988b1acaf0d078e526 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-25T00:18:41Z |
| publishDate | 2024-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-3dae04b9896b40988b1acaf0d078e5262024-03-12T16:55:27ZengMDPI AGSensors1424-82202024-03-01245163310.3390/s24051633Deep Learning Based Underwater Acoustic Target Recognition: Introduce a Recent Temporal 2D Modeling MethodJun Tang0Wenbo Gao1Enxue Ma2Xinmiao Sun3Jinying Ma4School of Civil Engineering, Tianjin University, Tianjin 300072, ChinaSchool of Civil Engineering, Tianjin University, Tianjin 300072, ChinaSchool of Civil Engineering, Tianjin University, Tianjin 300072, ChinaSchool of Electrical and Information Engineering, Tianjin University, Tianjin 300072, ChinaSchool of Electronic Engineering, Tianjin University of Technology and Education, Tianjin 300222, ChinaIn recent years, the application of deep learning models for underwater target recognition has become a popular trend. Most of these are pure 1D models used for processing time-domain signals or pure 2D models used for processing time-frequency spectra. In this paper, a recent temporal 2D modeling method is introduced into the construction of ship radiation noise classification models, combining 1D and 2D. This method is based on the periodic characteristics of time-domain signals, shaping them into 2D signals and discovering long-term correlations between sampling points through 2D convolution to compensate for the limitations of 1D convolution. Integrating this method with the current state-of-the-art model structure and using samples from the Deepship database for network training and testing, it was found that this method could further improve the accuracy (0.9%) and reduce the parameter count (30%), providing a new option for model construction and optimization. Meanwhile, the effectiveness of training models using time-domain signals or time-frequency representations has been compared, finding that the model based on time-domain signals is more sensitive and has a smaller storage footprint (reduced to 30%), whereas the model based on time-frequency representation can achieve higher accuracy (1–2%).https://www.mdpi.com/1424-8220/24/5/1633underwater acoustic target recognitiondeep learningtemporal 2D modelingshort-time Fourier transform |
| spellingShingle | Jun Tang Wenbo Gao Enxue Ma Xinmiao Sun Jinying Ma Deep Learning Based Underwater Acoustic Target Recognition: Introduce a Recent Temporal 2D Modeling Method Sensors underwater acoustic target recognition deep learning temporal 2D modeling short-time Fourier transform |
| title | Deep Learning Based Underwater Acoustic Target Recognition: Introduce a Recent Temporal 2D Modeling Method |
| title_full | Deep Learning Based Underwater Acoustic Target Recognition: Introduce a Recent Temporal 2D Modeling Method |
| title_fullStr | Deep Learning Based Underwater Acoustic Target Recognition: Introduce a Recent Temporal 2D Modeling Method |
| title_full_unstemmed | Deep Learning Based Underwater Acoustic Target Recognition: Introduce a Recent Temporal 2D Modeling Method |
| title_short | Deep Learning Based Underwater Acoustic Target Recognition: Introduce a Recent Temporal 2D Modeling Method |
| title_sort | deep learning based underwater acoustic target recognition introduce a recent temporal 2d modeling method |
| topic | underwater acoustic target recognition deep learning temporal 2D modeling short-time Fourier transform |
| url | https://www.mdpi.com/1424-8220/24/5/1633 |
| work_keys_str_mv | AT juntang deeplearningbasedunderwateracoustictargetrecognitionintroducearecenttemporal2dmodelingmethod AT wenbogao deeplearningbasedunderwateracoustictargetrecognitionintroducearecenttemporal2dmodelingmethod AT enxuema deeplearningbasedunderwateracoustictargetrecognitionintroducearecenttemporal2dmodelingmethod AT xinmiaosun deeplearningbasedunderwateracoustictargetrecognitionintroducearecenttemporal2dmodelingmethod AT jinyingma deeplearningbasedunderwateracoustictargetrecognitionintroducearecenttemporal2dmodelingmethod |