An Underwater Side-Scan Sonar Transfer Recognition Method Based on Crossed Point-to-Point Second-Order Self-Attention Mechanism
Recognizing targets through side-scan sonar (SSS) data by deep learning-based techniques has been particularly challenging. The primary challenge stems from the difficulty and time consumption associated with underwater acoustic data acquisition, which demands systematic explorations to obtain suffi...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-09-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/18/4517 |
| _version_ | 1797577255827800064 |
|---|---|
| author | Jian Wang Haisen Li Chao Dong Jing Wang Bing Zheng Tianyao Xing |
| author_facet | Jian Wang Haisen Li Chao Dong Jing Wang Bing Zheng Tianyao Xing |
| author_sort | Jian Wang |
| collection | DOAJ |
| description | Recognizing targets through side-scan sonar (SSS) data by deep learning-based techniques has been particularly challenging. The primary challenge stems from the difficulty and time consumption associated with underwater acoustic data acquisition, which demands systematic explorations to obtain sufficient training samples for accurate deep learning-based models. Moreover, if the sample size of the available data is small, the design of effective target recognition models becomes complex. These challenges have posed significant obstacles to developing accurate SSS-based target recognition methods via deep learning models. However, utilizing multi-modal datasets to enhance the recognition performance of sonar images through knowledge transfer in deep networks appears promising. Owing to the unique statistical properties of various modal images, transitioning between different modalities can significantly increase the complexity of network training. This issue remains unresolved, directly impacting the target transfer recognition performance. To enhance the precision of categorizing underwater sonar images when faced with a limited number of mode types and data samples, this study introduces a crossed point-to-point second-order self-attention (PPCSSA) method based on double-mode sample transfer recognition. In the PPCSSA method, first-order importance features are derived by extracting key horizontal and longitudinal point-to-point features. Based on these features, the self-supervised attention strategy effectively removes redundant features, securing the second-order significant features of SSS images. This strategy introduces a potent low-mode-type small-sample learning method for transfer learning. Classification experiment results indicate that the proposed method excels in extracting key features with minimal training complexity. Moreover, experimental outcomes underscore that the proposed technique enhances recognition stability and accuracy, achieving a remarkable overall accuracy rate of 99.28%. Finally, the proposed method maintains high recognition accuracy even in noisy environments. |
| first_indexed | 2024-03-10T22:05:38Z |
| format | Article |
| id | doaj.art-a3bd042b74b0430fa5de1322c30f4002 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T22:05:38Z |
| publishDate | 2023-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-a3bd042b74b0430fa5de1322c30f40022023-11-19T12:48:49ZengMDPI AGRemote Sensing2072-42922023-09-011518451710.3390/rs15184517An Underwater Side-Scan Sonar Transfer Recognition Method Based on Crossed Point-to-Point Second-Order Self-Attention MechanismJian Wang0Haisen Li1Chao Dong2Jing Wang3Bing Zheng4Tianyao Xing5National Key Laboratory of Underwater Acoustic Technology, Harbin Engineering University, Harbin 150001, ChinaNational Key Laboratory of Underwater Acoustic Technology, Harbin Engineering University, Harbin 150001, ChinaSouth China Sea Marine Survey Center, Ministry of Natural Resources, Guangzhou 510300, ChinaInstitute for Advanced Study, University of Electronic Science and Technology of China, UESTC, Shenzhen 518000, ChinaSouth China Sea Marine Survey Center, Ministry of Natural Resources, Guangzhou 510300, ChinaNational Key Laboratory of Underwater Acoustic Technology, Harbin Engineering University, Harbin 150001, ChinaRecognizing targets through side-scan sonar (SSS) data by deep learning-based techniques has been particularly challenging. The primary challenge stems from the difficulty and time consumption associated with underwater acoustic data acquisition, which demands systematic explorations to obtain sufficient training samples for accurate deep learning-based models. Moreover, if the sample size of the available data is small, the design of effective target recognition models becomes complex. These challenges have posed significant obstacles to developing accurate SSS-based target recognition methods via deep learning models. However, utilizing multi-modal datasets to enhance the recognition performance of sonar images through knowledge transfer in deep networks appears promising. Owing to the unique statistical properties of various modal images, transitioning between different modalities can significantly increase the complexity of network training. This issue remains unresolved, directly impacting the target transfer recognition performance. To enhance the precision of categorizing underwater sonar images when faced with a limited number of mode types and data samples, this study introduces a crossed point-to-point second-order self-attention (PPCSSA) method based on double-mode sample transfer recognition. In the PPCSSA method, first-order importance features are derived by extracting key horizontal and longitudinal point-to-point features. Based on these features, the self-supervised attention strategy effectively removes redundant features, securing the second-order significant features of SSS images. This strategy introduces a potent low-mode-type small-sample learning method for transfer learning. Classification experiment results indicate that the proposed method excels in extracting key features with minimal training complexity. Moreover, experimental outcomes underscore that the proposed technique enhances recognition stability and accuracy, achieving a remarkable overall accuracy rate of 99.28%. Finally, the proposed method maintains high recognition accuracy even in noisy environments.https://www.mdpi.com/2072-4292/15/18/4517attention mechanismside-scan sonar image classificationcrossed point-to-pointmulti-modal transfer learningself-supervision |
| spellingShingle | Jian Wang Haisen Li Chao Dong Jing Wang Bing Zheng Tianyao Xing An Underwater Side-Scan Sonar Transfer Recognition Method Based on Crossed Point-to-Point Second-Order Self-Attention Mechanism Remote Sensing attention mechanism side-scan sonar image classification crossed point-to-point multi-modal transfer learning self-supervision |
| title | An Underwater Side-Scan Sonar Transfer Recognition Method Based on Crossed Point-to-Point Second-Order Self-Attention Mechanism |
| title_full | An Underwater Side-Scan Sonar Transfer Recognition Method Based on Crossed Point-to-Point Second-Order Self-Attention Mechanism |
| title_fullStr | An Underwater Side-Scan Sonar Transfer Recognition Method Based on Crossed Point-to-Point Second-Order Self-Attention Mechanism |
| title_full_unstemmed | An Underwater Side-Scan Sonar Transfer Recognition Method Based on Crossed Point-to-Point Second-Order Self-Attention Mechanism |
| title_short | An Underwater Side-Scan Sonar Transfer Recognition Method Based on Crossed Point-to-Point Second-Order Self-Attention Mechanism |
| title_sort | underwater side scan sonar transfer recognition method based on crossed point to point second order self attention mechanism |
| topic | attention mechanism side-scan sonar image classification crossed point-to-point multi-modal transfer learning self-supervision |
| url | https://www.mdpi.com/2072-4292/15/18/4517 |
| work_keys_str_mv | AT jianwang anunderwatersidescansonartransferrecognitionmethodbasedoncrossedpointtopointsecondorderselfattentionmechanism AT haisenli anunderwatersidescansonartransferrecognitionmethodbasedoncrossedpointtopointsecondorderselfattentionmechanism AT chaodong anunderwatersidescansonartransferrecognitionmethodbasedoncrossedpointtopointsecondorderselfattentionmechanism AT jingwang anunderwatersidescansonartransferrecognitionmethodbasedoncrossedpointtopointsecondorderselfattentionmechanism AT bingzheng anunderwatersidescansonartransferrecognitionmethodbasedoncrossedpointtopointsecondorderselfattentionmechanism AT tianyaoxing anunderwatersidescansonartransferrecognitionmethodbasedoncrossedpointtopointsecondorderselfattentionmechanism AT jianwang underwatersidescansonartransferrecognitionmethodbasedoncrossedpointtopointsecondorderselfattentionmechanism AT haisenli underwatersidescansonartransferrecognitionmethodbasedoncrossedpointtopointsecondorderselfattentionmechanism AT chaodong underwatersidescansonartransferrecognitionmethodbasedoncrossedpointtopointsecondorderselfattentionmechanism AT jingwang underwatersidescansonartransferrecognitionmethodbasedoncrossedpointtopointsecondorderselfattentionmechanism AT bingzheng underwatersidescansonartransferrecognitionmethodbasedoncrossedpointtopointsecondorderselfattentionmechanism AT tianyaoxing underwatersidescansonartransferrecognitionmethodbasedoncrossedpointtopointsecondorderselfattentionmechanism |