Research on an Underwater Object Detection Network Based on Dual-Branch Feature Extraction
Underwater object detection is challenging in computer vision research due to the complex underwater environment, poor image quality, and varying target scales, making it difficult for existing object detection networks to achieve high accuracy in underwater tasks. To address the issues of limited d...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/16/3413 |
| _version_ | 1797584881392287744 |
|---|---|
| author | Xiao Chen Mujiahui Yuan Chenye Fan Xingwu Chen Yaan Li Haiyan Wang |
| author_facet | Xiao Chen Mujiahui Yuan Chenye Fan Xingwu Chen Yaan Li Haiyan Wang |
| author_sort | Xiao Chen |
| collection | DOAJ |
| description | Underwater object detection is challenging in computer vision research due to the complex underwater environment, poor image quality, and varying target scales, making it difficult for existing object detection networks to achieve high accuracy in underwater tasks. To address the issues of limited data and multi-scale targets in underwater detection, we propose a Dual-Branch Underwater Object Detection Network (DB-UODN) based on dual-branch feature extraction. In the feature extraction stage, we design a dual-branch structure by combining the You Only Look Once (YOLO) v7 backbone with the Enhanced Channel and Dilated Block (ECDB). It allows for the extraction and complementation of multi-scale features, which enable the model to learn both global and local information and enhance its perception of multi-scale features in underwater targets. Furthermore, we employ the DSPACSPC structure to replace the SPPCSPC structure in YOLOv7. The DSPACSPC structure utilizes atrous convolutions with different dilation rates to capture contextual information at various scales, compensating for potential information loss caused by pooling operations. Additionally, we utilize a dense connection structure to facilitate feature reuse and enhance the network’s representation and generalization capabilities. Experimental results demonstrate that the proposed DB-UODN outperforms the most commonly used object detection networks in underwater scenarios. On the URPC2020 dataset, the network achieves an average detection accuracy of 87.36%. |
| first_indexed | 2024-03-10T23:59:00Z |
| format | Article |
| id | doaj.art-2bf2efbcaf7d45be9d87a81dee8d3cf2 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T23:59:00Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-2bf2efbcaf7d45be9d87a81dee8d3cf22023-11-19T00:53:15ZengMDPI AGElectronics2079-92922023-08-011216341310.3390/electronics12163413Research on an Underwater Object Detection Network Based on Dual-Branch Feature ExtractionXiao Chen0Mujiahui Yuan1Chenye Fan2Xingwu Chen3Yaan Li4Haiyan Wang5School of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology, Xi’an 710021, ChinaSchool of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology, Xi’an 710021, ChinaSchool of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology, Xi’an 710021, ChinaSchool of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology, Xi’an 710021, ChinaSchool of Marine Science and Technology, Northwestern Polytechnical University, Xi’an 710072, ChinaSchool of Electronic Information and Artificial Intelligence, Shaanxi University of Science and Technology, Xi’an 710021, ChinaUnderwater object detection is challenging in computer vision research due to the complex underwater environment, poor image quality, and varying target scales, making it difficult for existing object detection networks to achieve high accuracy in underwater tasks. To address the issues of limited data and multi-scale targets in underwater detection, we propose a Dual-Branch Underwater Object Detection Network (DB-UODN) based on dual-branch feature extraction. In the feature extraction stage, we design a dual-branch structure by combining the You Only Look Once (YOLO) v7 backbone with the Enhanced Channel and Dilated Block (ECDB). It allows for the extraction and complementation of multi-scale features, which enable the model to learn both global and local information and enhance its perception of multi-scale features in underwater targets. Furthermore, we employ the DSPACSPC structure to replace the SPPCSPC structure in YOLOv7. The DSPACSPC structure utilizes atrous convolutions with different dilation rates to capture contextual information at various scales, compensating for potential information loss caused by pooling operations. Additionally, we utilize a dense connection structure to facilitate feature reuse and enhance the network’s representation and generalization capabilities. Experimental results demonstrate that the proposed DB-UODN outperforms the most commonly used object detection networks in underwater scenarios. On the URPC2020 dataset, the network achieves an average detection accuracy of 87.36%.https://www.mdpi.com/2079-9292/12/16/3413underwater object detectiondeep learningYOLOv7feature fusion |
| spellingShingle | Xiao Chen Mujiahui Yuan Chenye Fan Xingwu Chen Yaan Li Haiyan Wang Research on an Underwater Object Detection Network Based on Dual-Branch Feature Extraction Electronics underwater object detection deep learning YOLOv7 feature fusion |
| title | Research on an Underwater Object Detection Network Based on Dual-Branch Feature Extraction |
| title_full | Research on an Underwater Object Detection Network Based on Dual-Branch Feature Extraction |
| title_fullStr | Research on an Underwater Object Detection Network Based on Dual-Branch Feature Extraction |
| title_full_unstemmed | Research on an Underwater Object Detection Network Based on Dual-Branch Feature Extraction |
| title_short | Research on an Underwater Object Detection Network Based on Dual-Branch Feature Extraction |
| title_sort | research on an underwater object detection network based on dual branch feature extraction |
| topic | underwater object detection deep learning YOLOv7 feature fusion |
| url | https://www.mdpi.com/2079-9292/12/16/3413 |
| work_keys_str_mv | AT xiaochen researchonanunderwaterobjectdetectionnetworkbasedondualbranchfeatureextraction AT mujiahuiyuan researchonanunderwaterobjectdetectionnetworkbasedondualbranchfeatureextraction AT chenyefan researchonanunderwaterobjectdetectionnetworkbasedondualbranchfeatureextraction AT xingwuchen researchonanunderwaterobjectdetectionnetworkbasedondualbranchfeatureextraction AT yaanli researchonanunderwaterobjectdetectionnetworkbasedondualbranchfeatureextraction AT haiyanwang researchonanunderwaterobjectdetectionnetworkbasedondualbranchfeatureextraction |