Multi-stage dynamic wavelet gated aggregation network for underwater image denoising

Wei Li; Haiyan Xie; Jiaxi Li; Huiyu Zhou; Wei Li; Haiyan Xie; Jiaxi Li; Huiyu Zhou

doi:10.3934/era.2026270

Electronic Research Archive

2026, Volume 34, Issue 9: 6121-6152. doi: 10.3934/era.2026270

Previous Article Next Article

Research article Special Issues

Multi-stage dynamic wavelet gated aggregation network for underwater image denoising

1.
School of Science, Dalian Maritime University, Dalian 116026, China
2.
School of Computing and Mathematical Sciences, University of Leicester, Leicester, LE1 7RH, United Kingdom

Received: 20 March 2026 Revised: 17 June 2026 Accepted: 29 June 2026 Published: 15 July 2026

Underwater image quality is inevitably degraded by the selective absorption and scattering of light, which introduce severe blurring and abrupt high-frequency noise. These interferences obscure latent structural information and make robust restoration a significant challenge. To address this, we present MWGANet, a multi-stage dynamic wavelet-gated aggregation network designed to decouple complex underwater noise distributions caused by environmental disturbances. The core of our architecture was a parallel strategy with three pathways tailored for feature extraction across domains. Specifically, the Gated Wavelet Transform Branch (GWTB) isolated noise spikes in the frequency domain to effectively safeguard critical structural textures from corruption. Our adaptive reuse branch (ARB) incorporated a dynamic aggregation block at multiple scales to facilitate feature flow across layers while adaptively filtering redundant information through learnable gating mechanisms. Furthermore, the dilated context branch (DCB) expanded the model's receptive field via graduated dilation rates to enhance the perception of diverse object priors. Extensive evaluations confirmed that MWGANet achieves a superior balance between denoising fidelity and computational efficiency. In particular, the model maintained a high-fidelity output on the EUVP dataset, achieving a PSNR of 31.06 dB even under severe synthetic additive white Gaussian noise (AWGN) ($ \sigma = 50 $). Such robustness and efficiency indicated that MWGANet has the potential to serve as a practical real-time solution for resource-constrained autonomous underwater vehicle (AUV) platforms.
- underwater image denoising,
- wavelet transform,
- attention mechanism,
- multi-scale aggregation,
- feature reuse
Citation: Wei Li, Haiyan Xie, Jiaxi Li, Huiyu Zhou. Multi-stage dynamic wavelet gated aggregation network for underwater image denoising[J]. Electronic Research Archive, 2026, 34(9): 6121-6152. doi: 10.3934/era.2026270

Related Papers:

Abstract

Underwater image quality is inevitably degraded by the selective absorption and scattering of light, which introduce severe blurring and abrupt high-frequency noise. These interferences obscure latent structural information and make robust restoration a significant challenge. To address this, we present MWGANet, a multi-stage dynamic wavelet-gated aggregation network designed to decouple complex underwater noise distributions caused by environmental disturbances. The core of our architecture was a parallel strategy with three pathways tailored for feature extraction across domains. Specifically, the Gated Wavelet Transform Branch (GWTB) isolated noise spikes in the frequency domain to effectively safeguard critical structural textures from corruption. Our adaptive reuse branch (ARB) incorporated a dynamic aggregation block at multiple scales to facilitate feature flow across layers while adaptively filtering redundant information through learnable gating mechanisms. Furthermore, the dilated context branch (DCB) expanded the model's receptive field via graduated dilation rates to enhance the perception of diverse object priors. Extensive evaluations confirmed that MWGANet achieves a superior balance between denoising fidelity and computational efficiency. In particular, the model maintained a high-fidelity output on the EUVP dataset, achieving a PSNR of 31.06 dB even under severe synthetic additive white Gaussian noise (AWGN) ($ \sigma = 50 $). Such robustness and efficiency indicated that MWGANet has the potential to serve as a practical real-time solution for resource-constrained autonomous underwater vehicle (AUV) platforms.

References

[1]	D. Akkaynak, T. Treibitz, A revised underwater image formation model, in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2018), 6723–6732. https://doi.org/10.1109/CVPR.2018.00703
[2]	C. Li, W. Liu, G. Gong, X. Ding, X. Zhong, SU-YOLO: Spiking neural network for efficient underwater object detection, Neurocomputing, 644 (2025), 130310. https://doi.org/10.1016/j.neucom.2025.130310 doi: 10.1016/j.neucom.2025.130310
[3]	J. S. Jaffe, Computer modeling and the design of optimal underwater imaging systems, IEEE J. Ocean. Eng., 15 (1990), 101–111. https://doi.org/10.1109/48.50695 doi: 10.1109/48.50695
[4]	R. Schettini, S. Corchs, Underwater image processing: State of the art of restoration and image enhancement methods, EURASIP J. Adv. Signal Process., 2010 (2010), 746052. https://doi.org/10.1155/2010/746052 doi: 10.1155/2010/746052
[5]	C. Li, C. Guo, W. Ren, R. Cong, J. Hou, S. Kwong, et al., An underwater image enhancement benchmark dataset and beyond, IEEE Trans. Image Process., 29 (2020), 4376–4389. https://doi.org/10.1109/TIP.2019.2955241 doi: 10.1109/TIP.2019.2955241
[6]	S. Anwar, C. Li, Diving deeper into underwater image enhancement: A survey, Signal Process.: Image Commun., 89 (2020), 115978. https://doi.org/10.1016/j.image.2020.115978 doi: 10.1016/j.image.2020.115978
[7]	K. Zhang, W. Zuo, Y. Chen, D. Meng, L. Zhang, Beyond a gaussian denoiser: Residual learning of deep CNN for image denoising, IEEE Trans. Image Process., 26 (2017), 3142–3155. https://doi.org/10.1109/TIP.2017.2662206 doi: 10.1109/TIP.2017.2662206
[8]	Y. Blau, T. Michaeli, The perception-distortion tradeoff, in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2018), 6228–6237. https://doi.org/10.1109/CVPR.2018.00652
[9]	A. Foi, M. Trimeche, V. Katkovnik, K. Egiazarian, Practical Poissonian-Gaussian noise modeling and fitting for single-image raw-data, IEEE Trans. Image Process., 17 (2008), 1737–1754. https://doi.org/10.1109/TIP.2008.2001399 doi: 10.1109/TIP.2008.2001399
[10]	S. Guo, Z. Yan, K. Zhang, W. Zuo, L. Zhang, Toward convolutional blind denoising of real photographs, in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), (2019), 1712–1722. https://doi.org/10.1109/CVPR.2019.00182
[11]	C. Tian, L. Fei, W. Zheng, Y. Xu, W. Zuo, C. W. Lin, Deep learning on image denoising: An overview, Neural Networks, 131 (2020), 251–275. https://doi.org/10.1016/j.neunet.2020.07.025 doi: 10.1016/j.neunet.2020.07.025
[12]	A. Buades, B. Coll, J. M. Morel, A non-local algorithm for image denoising, in 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05), (2005), 60–65. https://doi.org/10.1109/CVPR.2005.38
[13]	K. Dabov, A. Foi, V. Katkovnik, K. Egiazarian, Image denoising by sparse 3-D transform-domain collaborative filtering, IEEE Trans. Image Process., 16 (2007), 2080–2095. https://doi.org/10.1109/TIP.2007.901238 doi: 10.1109/TIP.2007.901238
[14]	S. Gu, L. Zhang, W. Zuo, X. Feng, Weighted nuclear norm minimization with application to image denoising, in 2014 IEEE Conference on Computer Vision and Pattern Recognition, (2014), 2862–2869. https://doi.org/10.1109/CVPR.2014.366
[15]	J. Mairal, F. Bach, J. Ponce, G. Sapiro, A. Zisserman, Non-local sparse models for image restoration, in 2009 IEEE 12th International Conference on Computer Vision, (2009), 2272–2279. https://doi.org/10.1109/ICCV.2009.5459452
[16]	S. W. Zamir, A. Arora, S. Khan, M. Hayat, F. S. Khan, M. H. Yang, Restormer: Efficient transformer for high-resolution image restoration, in 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), (2022), 5728–5739. https://doi.org/10.1109/CVPR52688.2022.00564
[17]	Z. Wang, X. Cun, J. Bao, W. Zhou, J. Liu, H. Li, Uformer: A general U-shaped transformer for image restoration, in 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), (2022), 17683–17693. https://doi.org/10.1109/CVPR52688.2022.01716
[18]	S. G. Mallat, A theory for multiresolution signal decomposition: The wavelet representation, IEEE Trans. Pattern Anal. Mach. Intell., 11 (1989), 674–693. https://doi.org/10.1109/34.192463 doi: 10.1109/34.192463
[19]	D. L. Donoho, De-noising by soft-thresholding, IEEE Trans. Inf. Theory, 41 (1995), 613–627. https://doi.org/10.1109/18.382009 doi: 10.1109/18.382009
[20]	S. G. Chang, B. Yu, M. Vetterli, Adaptive wavelet thresholding for image denoising and compression, IEEE Trans. Image Process., 9 (2000), 1532–1546. https://doi.org/10.1109/83.862633 doi: 10.1109/83.862633
[21]	J. Portilla, V. Strela, M. J. Wainwright, E. P. Simoncelli, Image denoising using scale mixtures of Gaussians in the wavelet domain, IEEE Trans. Image Process., 12 (2003), 1338–1351. https://doi.org/10.1109/TIP.2003.818640 doi: 10.1109/TIP.2003.818640
[22]	C. J. Prabhakar, P. U. P. Kumar, An image based technique for enhancement of underwater images, Int. J. Mach. Intell., 3 (2010), 217–224.
[23]	M. Jian, X. Liu, H. Luo, X. Lu, H. Yu, J. Dong, Underwater image processing and analysis: A review, Signal Process. Image Commun., 91 (2021), 116088. https://doi.org/10.1016/j.image.2020.116088 doi: 10.1016/j.image.2020.116088
[24]	P. Liu, H. Zhang, K. Zhang, L. Lin, W. Zuo, Multi-level wavelet-CNN for image restoration, in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), (2018), 773–782. https://doi.org/10.1109/CVPRW.2018.00121
[25]	H. Huang, R. He, Z. Sun, T. Tan, Wavelet-SRNet: A wavelet-based CNN for multi-scale face super resolution, in 2017 IEEE International Conference on Computer Vision (ICCV), (2017), 1689–1697. https://doi.org/10.1109/ICCV.2017.187
[26]	K. Jiang, Z. Wang, P. Yi, C. C. Wang, Z. Wang, Z. Han, et al., Direction-aware attention wavelet network for image deraining, in Proceedings of the 31st ACM International Conference on Multimedia, (2023), 7065–7074. https://doi.org/10.1145/3581783.3611697
[27]	Y. Rao, W. Zhao, Y. Tang, J. Zhou, S. N. Lim, J. Lu, HorNet: Efficient high-order spatial interactions with recursive gated convolutions, Adv. Neural Inf. Process. Syst., 35 (2022), 10353–10366.
[28]	M. H. Guo, C. Z. Lu, Z. N. Liu, M. M. Cheng, S. M. Hu, Visual attention network, Comput. Vis. Media, 9 (2023), 733–752. https://doi.org/10.1007/s41095-023-0364-2
[29]	S. K. Li, X. Wang, H. Zuo, C. Qian, S. Pu, M. Li, MogaNet: Multi-order gated aggregation network, in International Conference on Learning Representations, (2024), 37393–37427.
[30]	X. Li, W. Wang, X. Hu, J. Yang, Selective kernel networks, in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), (2019), 510–519. https://doi.org/10.1109/CVPR.2019.00060
[31]	H. Zhang, C. Wu, Z. Zhang, Y. Zhu, H. Lin, Z. Zhang, et al., ResNeSt: Split-attention networks, IEEE Trans. Pattern Anal. Mach. Intell., 44 (2022), 5765–5776. https://doi.org/10.1109/CVPRW56347.2022.00309 doi: 10.1109/CVPRW56347.2022.00309
[32]	B. Yang, G. Bender, Q. V. Le, J. Ngiam, CondConv: Conditionally parameterized convolutions for efficient inference, preprint, arXiv: 1904.04971.
[33]	C. Li, A. Zhou, A. Yao, Omni-dimensional dynamic convolution, preprint, arXiv: 2209.07947.
[34]	J. Long, E. Shelhamer, T. Darrell, Fully convolutional networks for semantic segmentation, in 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), (2015), 3431–3440. https://doi.org/10.1109/CVPR.2015.7298965
[35]	G. Huang, Z. Liu, L. van der Maaten, K. Q. Weinberger, Densely connected convolutional networks, in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), (2017), 4700–4708. https://doi.org/10.1109/CVPR.2017.243
[36]	J. Hu, L. Shen, G. Sun, Squeeze-and-excitation networks, in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2018), 7132–7141. https://doi.org/10.1109/CVPR.2018.00745
[37]	S. Woo, J. Park, J. Y. Lee, I. S. Kweon, CBAM: Convolutional block attention module, in Computer Vision – ECCV 2018, (2018), 3–19. https://doi.org/10.1007/978-3-030-01234-2_1
[38]	Y. Zhang, Y. Tian, Y. Kong, B. Zhong, Y. Fu, Residual dense network for image super-resolution, in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2018), 2472–2481. https://doi.org/10.1109/CVPR.2018.00262
[39]	Q. Jiang, Y. Chen, G. Wang, T. Ji, A novel deep neural network for noise removal from underwater image, Signal Process. Image Commun., 87 (2020), 115921. https://doi.org/10.1016/j.image.2020.115921 doi: 10.1016/j.image.2020.115921
[40]	H. Sheng, X. Wang, C. Zhang, J. Wang, P. Duan, Y. Wang, AIGC video detection based on the fusion of spatial-frequency-optical flow multimodal features, J. Syst. Eng. Electron., 36 (2026), 1–15. https://doi.org/10.23919/JSEE.2026.000049 doi: 10.23919/JSEE.2026.000049
[41]	H. Zhao, O. Gallo, I. Frosio, J. Kautz, Loss functions for image restoration with neural networks, IEEE Trans. Comput. Imaging, 3 (2016), 47–57. https://doi.org/10.1109/TCI.2016.2644865 doi: 10.1109/TCI.2016.2644865
[42]	J. Yang, H. Xie, N. Xue, A. Zhang, Research on underwater image denoising based on dual-channel residual network, Comput. Eng., 49 (2023), 188–198. https://doi.org/10.19678/j.issn.1000-3428.0064662 doi: 10.19678/j.issn.1000-3428.0064662
[43]	M. J. Islam, Y. Xia, J. Sattar, Simultaneous enhancement and super-resolution of underwater imagery for improved visual perception, in Robotics: Science and Systems (RSS), 2020. https://doi.org/10.15607/RSS.2020.XVI.018
[44]	M. J. Islam, Y. Xia, J. Sattar, Fast underwater image enhancement for improved visual perception, IEEE Robot. Autom. Lett., 5 (2020), 3227–3234. https://doi.org/10.1109/LRA.2020.2974710 doi: 10.1109/LRA.2020.2974710
[45]	R. Timofte, E. Agustsson, L. Van Gool, M. Yang, L. Zhang, B. Lim, et al., NTIRE 2017 challenge on single image super-resolution: Dataset and study, in 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), (2017) 1122–1131. https://doi.org/10.1109/CVPRW.2017.149
[46]	R. Franzen, Kodak lossless true color image suite, 1999.
[47]	D. Martin, C. Fowlkes, D. Tal, J. Malik, A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics, in Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001, (2001), 416–423. https://doi.org/10.1109/ICCV.2001.937655
[48]	T. Q. Huynh-Thu, M. Ghanbari, Scope of validity of PSNR in image/video quality assessment, Electron. Lett., 44 (2008), 800–801. https://doi.org/10.1049/el:20080522 doi: 10.1049/el:20080522
[49]	Z. Wang, A. C. Bovik, H. R. Sheikh, E. P. Simoncelli, Image quality assessment: From error visibility to structural similarity, IEEE Trans. Image Process., 13 (2004), 600–612. https://doi.org/10.1109/TIP.2003.819861 doi: 10.1109/TIP.2003.819861
[50]	R. Zhang, P. Isola, A. A. Efros, E. Shechtman, O. Wang, The unreasonable effectiveness of deep features as a perceptual metric, in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2018), 586–595. https://doi.org/10.1109/CVPR.2018.00068
[51]	C. Tian, Y. Xu, W. Zuo, L. Fei, C. W. Lin, Attention-guided CNN denoising network (ADNet), Neural Networks, 121 (2020), 461–473. https://doi.org/10.1016/j.neunet.2019.12.024 doi: 10.1016/j.neunet.2019.12.024
[52]	C. Tian, Y. Xu, L. Fei, J. Wang, J. Wen, N. Luo, Enhanced CNN for image denoising, CAAI Trans. Intell. Technol., 4 (2019), 17–23. https://doi.org/10.1049/trit.2018.1054 doi: 10.1049/trit.2018.1054
[53]	C. Tian, Y. Xu, W. Zuo, B. Du, C. W. Lin, D. Zhang, Designing and training of a dual CNN for image denoising, Knowl.-Based Syst., 226 (2021), 107145. https://doi.org/10.1016/j.knosys.2021.106949 doi: 10.1016/j.knosys.2021.106949
[54]	I. Batool, M. Imran, A dual residual dense network for image denoising, Eng. Appl. Artif. Intell., 147 (2025), 110275. https://doi.org/10.1016/j.engappai.2025.110275 doi: 10.1016/j.engappai.2025.110275
[55]	C. Tian, M. Zheng, W. Zuo, B. Zhang, Y. Zhang, D. Zhang, Multi-stage image denoising with the wavelet transform, Pattern Recognit., 134 (2023), 109050. https://doi.org/10.1016/j.patcog.2022.109050 doi: 10.1016/j.patcog.2022.109050

Reader Comments

Your name:*

Email:*
© 2026 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)