A Novel Underwater Image Enhancement Using Optimal Composite Backbone Network
Source abstract
Abstract
Continuous exploration of the ocean has made underwater image processing an important research field, and plenty of CNN (convolutional neural network)-based underwater image enhancement methods have emerged over time. However, the feature-learning ability of existing CNN-based underwater image enhancement is limited. The networks were designed to be complicated or embed other algorithms for better results, which cannot simultaneously meet the requirements of suitable underwater image enhancement effects and real-time performance. Although the composite backbone network (CBNet) was introduced in underwater image enhancement, we proposed OECBNet (optimal underwater image-enhancing composite backbone network) to obtain a better enhancement effect and shorten the running time. Herein, a comprehensive study of different composite architectures in an underwater image enhancement network was carried out by comparing the number of backbones, connection strategies, pruning strategies for composite backbones, and auxiliary losses. Then, a CBNet with optimal performance was obtained. Finally, cross-sectional research of the obtained network with the state-of-the-art underwater enhancement network was performed. The experiments showed that our optimized composite backbone network achieved better-enhanced images than those of existing CNN-based methods.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Image Enhancement Techniques · Advanced Neural Network Applications · Advanced Image Processing Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Mechanical and Energy Engineering, Southern University of Science and Technology, Shenzhen 518055, China
Health Management System Engineering Center, School of Public Health, Hangzhou Normal University, Hangzhou 311121, China
Institute of Oceanographic Instrumentation, Qilu University of Technology (Shandong Academy of Sciences), Qingdao 266075, China
Funding
Guangdong Education Bureau · Zhejiang Provinces Science Foundation · key R&D Program of Zhejiang Province
References (41)
Show source-provided references
1. Novel event analysis for human-machine collaborative underwater exploration10.1016/j.patcog.2019.106967 ↗
2. Discovering unknowns: Context-enhanced anomaly detection for curiosity-driven autonomous underwater exploration10.1016/j.patcog.2022.108860 ↗
3. Underwater Image Enhancement via Minimal Color Loss and Locally Adaptive Contrast Enhancement10.1109/TIP.2022.3177129 ↗
4. Li, W., Hu, W., Dong, T., and Qu, J. (2018, January 8–9). Depth Image Enhancement Algorithm Based on RGB Image Fusion. Proceedings of the 2018 11th International Symposium on Computational Intelligence and Design (ISCID), Hangzhou, China.10.1109/ISCID.2018.10126 ↗
5. Color Balance and Fusion for Underwater Image Enhancement10.1109/TIP.2017.2759252 ↗
6. Generalization of the Dark Channel Prior for Single Image Restoration10.1109/TIP.2018.2813092 ↗
7. Low-light image enhancement via deep Retinex decomposition and bilateral learning10.1016/j.image.2021.116466 ↗
8. Underwater Image Enhancement Using Adaptive Retinal Mechanisms10.1109/TIP.2019.2919947 ↗
9. Underwater scene prior inspired deep underwater image and video enhancement10.1016/j.patcog.2019.107038 ↗
10. Learning to Enhance Low-Light Image via Zero-Reference Deep Curve Estimation10.1109/TITS.2020.3042973 ↗
11. Underwater Image Enhancement Using a Multiscale Dense Generative Adversarial Network10.1109/JOE.2019.2911447 ↗
12. CBNet: A Novel Composite Backbone Network Architecture for Object Detection
13. CBNet: A Composite Backbone Network Architecture for Object Detection10.1109/TIP.2022.3216771 ↗
14. Diving deeper into underwater image enhancement: A survey10.1016/j.image.2020.115978 ↗
15. Ronneberger, O., Fischer, P., and Brox, T. (2015). Medical Image Computing and Computer-Assisted Intervention 2015, Springer International Publishing.
16. Cascaded transformer U-net for image restoration10.1016/j.sigpro.2022.108902 ↗
17. MLFcGAN: Multilevel Feature Fusion-Based Conditional GAN for Underwater Image Color Correction10.1109/LGRS.2019.2950056 ↗
18. Investigating intrinsic degradation factors by multi-branch aggregation for real-world underwater image enhancement10.1016/j.patcog.2022.109041 ↗
19. An Underwater Image Enhancement Benchmark Dataset and Beyond10.1109/TIP.2019.2955241 ↗
20. FW-GAN: Underwater image enhancement using generative adversarial network with multi-scale fusion10.1016/j.image.2022.116855 ↗
21. UIEC^2-Net: CNN-based underwater image enhancement using two color space10.1016/j.image.2021.116250 ↗
22. Chen, Y., Ke, W., Kou, L., Li, Q., Lu, D., Bai, Y., and Wang, Z. (Int. J. Bio Inspired Comput., 2023). Multiple Channel Adjustment based on Composite Backbone Network for Underwater Image Enhancement, Int. J. Bio Inspired Comput., to be published.10.1504/IJBIC.2023.135476 ↗
23. Chen, Y., Li, H., Yuan, Q., Wang, Z., Hu, C., and Ke, W. (2023, January 24–26). Underwater Image Enhancement based on Improved Water-Net. Proceedings of the 2022 IEEE International Conference on Cyborg and Bionic Systems (CBS), Wuhan, China.10.1109/CBS55922.2023.10115390 ↗
24. UW-GAN: Single-Image Depth Estimation and Image Enhancement for Underwater Images10.1109/TIM.2021.3120130 ↗
25. Conditional generative adversarial network with dual-branch progressive generator for underwater image enhancement10.1016/j.image.2022.116805 ↗
26. Multi-scale adversarial network for underwater image restoration10.1016/j.optlastec.2018.05.048 ↗
27. MFFN: An Underwater Sensing Scene Image Enhancement Method Based on Multiscale Feature Fusion Network
28. Real-World Underwater Enhancement: Challenges, Benchmarks, and Solutions Under Natural Light10.1109/TCSVT.2019.2963772 ↗
29. Comprehensive Underwater Object Tracking Benchmark Dataset and Underwater Image Enhancement With GAN10.1109/JOE.2021.3086907 ↗
30. Deep variance network: An iterative, improved CNN framework for unbalanced training datasets10.1016/j.patcog.2018.03.035 ↗
31. Efficient densely connected convolutional neural networks10.1016/j.patcog.2020.107610 ↗
32. Res2Net: A New Multi-Scale Backbone Architecture10.1109/TPAMI.2019.2938758 ↗
33. Liang, M., and Hu, X. (2015, January 7–12). Recurrent convolutional neural network for object recognition. Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA.
34. Zhang, X., Zhou, X., Lin, M., and Sun, J. (2018, January 18–23). ShufflfleNet: An extremely effificient convolutional neural network for mobile devices. Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.10.1109/CVPR.2018.00716 ↗
35. Li, Z., Peng, C., Yu, G., Zhang, X., Deng, Y., and Sun, J. (2018). European Conference on Computer Vision, Springer.
36. Hyperspectral Image Instance Segmentation Using Spectral–Spatial Feature Pyramid Network10.1109/TGRS.2023.3240481 ↗
37. CNN Cloud Detection Algorithm Based on Channel and Spatial Attention and Probabilistic Upsampling for Remote Sensing Image
38. UAVformer: A Composite Transformer Network for Urban Scene Segmentation of UAV Images10.1016/j.patcog.2022.109019 ↗
39. Composited FishNet: Fish Detection and Species Recognition from Low-Quality Underwater Videos10.1109/TIP.2021.3074738 ↗
40. An Underwater Color Image Quality Evaluation Metric10.1109/TIP.2015.2491020 ↗
41. Human-Visual-System-Inspired Underwater Image Quality Measures10.1109/JOE.2015.2469915 ↗