Optimization and Application of Improved YOLOv9s-UI for Underwater Object Detection
Source abstract
Abstract
The You Only Look Once (YOLO) series of object detection models is widely recognized for its efficiency and real-time performance, particularly under the challenging conditions of underwater environments, characterized by insufficient lighting and visual disturbances. By modifying the YOLOv9s model, this study aims to improve the accuracy and real-time capabilities of underwater object detection, resulting in the introduction of the YOLOv9s-UI detection model. The proposed model incorporates the Dual Dynamic Token Mixer (D-Mixer) module from TransXNet to improve feature extraction capabilities. Additionally, it integrates a feature fusion network design from the LocalMamba network, employing channel and spatial attention mechanisms. These attention modules effectively guide the feature fusion process, significantly enhancing detection accuracy while maintaining the model’s compact size of only 9.3 M. Experimental evaluation on the UCPR2019 underwater object dataset shows that the YOLOv9s-UI model has higher accuracy and recall than the existing YOLOv9s model, as well as excellent real-time performance. This model significantly improves the ability of underwater target detection by introducing advanced feature extraction and attention mechanisms. The model meets portability requirements and provides a more efficient solution for underwater detection.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Water Quality Monitoring Technologies · Image Enhancement Techniques · Advanced Neural Network Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Power Engineering, Naval University of Engineering, Wuhan 430000, China
References (39)
Show source-provided references
1. Redmon, J., and Farhadi, A. (2018). YOlOv3: An incremental improvement. arXiv.
2. Bochkovskiy, A., Wang, C.-Y., and Liao, H.-Y.M. (2020). YOlOv4: Optimal speed and accuracy of object detection. arXiv.
3. Rethinking general underwater object detection: Datasets, challenges, and solutions10.1016/j.neucom.2022.10.039 ↗
4. Underwater object detection: Architectures and algorithms–a comprehensive review10.1007/s11042-022-12502-1 ↗
5. Awalludin, E.A., Arsad, T.N.T., Yussof, W.N.J.H.W., Bachok, Z., and Hitam, M.S. (2022). A comparative study of various edge detection techniques for underwater images. J. Telecommun. Inf. Technol.10.26636/jtit.2022.155921 ↗
6. Integrate MSRCR and mask R-CNN to recognize underwater creatures on small sample datasets10.1109/ACCESS.2020.3025617 ↗
7. Underwater target detection based on Faster R-CNN and adversarial occlusion network10.1016/j.engappai.2021.104190 ↗
8. Saini, A., and Biswas, M. (2019, January 23–25). Object detection in underwater image by detecting edges using adaptive thresholding. Proceedings of the 2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI), Tirunelveli, India.10.1109/ICOEI.2019.8862794 ↗
9. Yuan, X., Guo, L., Luo, C., Zhou, X., and Yu, C. (2022). A survey of target detection and recognition methods in underwater turbid areas. Appl. Sci., 12.10.3390/app12104898 ↗
10. Underwater cable detection in the images using edge classification based on texture information10.1016/j.measurement.2016.05.030 ↗
11. Lin, Y.-H., Chen, S.-Y., and Tsou, C.-H. (2019). Development of an image processing module for autonomous underwater vehicles through integration of visual recognition with stereoscopic image reconstruction. J. Mar. Sci. Eng., 7.10.1115/OMAE2019-95321 ↗
12. Mandal, R., Connolly, R.M., Schlacher, T.A., and Stantic, B. (2018, January 8–13). Assessing fish abundance from underwater video using deep neural networks. Proceedings of the 2018 International Joint Conference on Neural Networks (IJCNN), Rio de Janeiro, Brazil.10.1109/IJCNN.2018.8489482 ↗
13. Sung, M., Yu, S.-C., and Girdhar, Y. (2017, January 19–22). Vision based real-time fish detection using convolutional neural network. Proceedings of the OCEANS 2017-Aberdeen, Aberdeen, UK.10.1109/OCEANSE.2017.8084889 ↗
14. Real-time detection of uneaten feed pellets in underwater images for aquaculture using an improved yolo-v4 network10.1016/j.compag.2021.106135 ↗
15. Composited fishnet: Fish detection and species recognition from low-quality under-water videos10.1109/TIP.2021.3074738 ↗
16. Er, M.J., Chen, J., Zhang, Y., and Gao, W. (2023). Research challenges, recent advances, and popular datasets in deep learning-based underwater marine object detection: A review. Sensors, 23.10.36227/techrxiv.19350389 ↗
17. A systematic review and analysis of deep learning-based underwater object detection10.1016/j.neucom.2023.01.056 ↗
18. Lightweight deep neural network for joint learning of underwater object detection and color conversion10.1109/TNNLS.2021.3072414 ↗
19. Underwater image processing and object detection based on deep CNN method10.1155/2020/6707328 ↗
20. Zhang, J., Peng, X., and Zhang, G. (2022, January 19–21). Using Improved YOLOX for Underwater Object Recognition. Proceedings of the 2022 5th International Conference on Pattern Recognition and Artificial Intelligence (PRAI), Chengdu, China.10.1109/PRAI55851.2022.9904162 ↗
21. Shallow mud detection algorithm for submarine channels based on improved YOLOv5s10.1016/j.heliyon.2024.e31029 ↗
22. Zhang, C., and Jiao, P. (arXiv, 2023). YOLO series target detection algorithms for underwater environments, arXiv.
23. UGC-YOLO: Underwater environment object detection based on YOLO with a global context block10.1007/s11802-023-5296-z ↗
24. Zhang, M., Xu, S., Song, W., He, Q., and Wei, Q. (2021). Lightweight underwater object detection based on yolo v4 and multi-scale attentional feature fusion. Remote Sens., 13.10.3390/rs13224706 ↗
25. Jia, J., Fu, M., Liu, X., and Zheng, B. (2022). Underwater object detection based on improved efficientDet. Remote Sens., 14.10.3390/rs14184487 ↗
26. Yuan, S., Luo, X., and Xu, R. (2024, January 8–10). Underwater Robot Target Detection Based On Improved YOLOv5 Network. Proceedings of the 2024 12th International Conference on Intelligent Control and Information Processing (ICICIP), Nanjing, China.10.1109/ICICIP60808.2024.10477835 ↗
27. Wang, C.Y., Yeh, I.H., and Liao, H.Y.M. (2024). YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information. arXiv.
28. Lou, M., Zhou, H.Y., Yang, S., and Yu, Y. (2023). TransXNet: Learning both global and local dynamics with a dual dynamic token mixer for visual recognition. arXiv.
29. Xu, R., Yang, S., Wang, Y., Du, B., and Chen, H. (2024). A survey on vision mamba: Models, applications and challenges. arXiv.
30. Underwater object recognition using transformable template matching based on prior knowledge10.1155/2019/2892975 ↗
31. Chen, R., Zhan, S., and Chen, Y. (2022, January 17–20). Underwater target detection algorithm based on YOLO and Swin transformer for sonar images. Proceedings of the OCEANS 2022, Hampton Roads, VA, USA.10.1109/OCEANS47191.2022.9976986 ↗
32. Exploring underwater target detection algorithm based on improved SSD10.1051/jnwpu/20203840747 ↗
33. Jiang, X., Zhuang, X., Chen, J., Zhang, J., and Zhang, Y. (2024). YOLOv8-MU: An Improved YOLOv8 Underwater Detector Based on a Large Kernel Block and a Multi-Branch Reparameterization Module. Sensors, 24.10.20944/preprints202404.0779.v1 ↗
34. Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.-Y., and Berg, A.C. (2016). SSD: Single Shot MultiBox Detector. Lecture Notes in Computer Science, Springer International Publishing.10.1007/978-3-319-46448-0_2 ↗
35. Tian, Z., Shen, C., Chen, H., and He, T. (November, January 27). FCOS: Fully Convolutional One-Stage Object Detection. Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, Republic of Korea.10.1109/ICCV.2019.00972 ↗
36. Learning to Match Anchors for Visual Object Detection10.1109/TPAMI.2021.3050494 ↗
37. Tan, M., and Le, Q.V. (2019, January 10–15). EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. Proceedings of the 36th International Conference on Machine Learning, Long Beach, CA, USA.
38. Edge Computing-Based Person Detection System for Top View Surveillance: Using CenterNet with Transfer Learning10.1016/j.asoc.2021.107489 ↗
39. Marine Organism Detection Based on Double Domains Augmentation and an Improved YOLOv710.1109/ACCESS.2023.3287932 ↗