Abstract：To address the image degradation caused by light attenuation and scattering in underwater imaging, an underwater image enhancement network based on semantic collaborative perceptual attention is proposed. First, a dual-path competitive perceptual attention mechanism is designed. Sliding window attention and pooling attention are integrated into Softmax to synchronously capture both coarse-grained and fine-grained features of images, thereby enabling multi-scale feature perception. Second, a convolutional gated linear unit is introduced to realize the attentionalization of the channel mixer. Based on the above, a cascaded perceptual attention network is constructed by integrating the perceptual attention mechanism and convolutional gated linear unit to capture and fuse local and global information of underwater degraded images. Finally, a feature dominant module is developed to embed and collaboratively integrate semantic features, and the capability of the model to understand and represent underwater scene semantics is enhanced. This elevates the enhancement process from mere pixel-level restoration to semantic-level scene reconstruction. Experiments demonstrate that the proposed network exhibits superior generalization ability and significant application value in downstream visual engineering tasks.
[1] 丛晓峰,桂杰,贺磊,等.基于视觉多头注意力与跨层白化的水下图像增强网络.模式识别与人工智能, 2023, 36(5): 407-418. (CONG X F, GUI J, HE L, et al. Underwater Image Enhancement Network Based on Visual Multi-head Attention and Skip-Layer Whi-tening. Pattern Recognition and Artificial Intelligence, 2023, 36(5): 407-418.) [2] 徐会希,吕凤天,石凯,等.深海长期驻留自主水下机器人系统研究现状与发展趋势.机器人, 2023, 45(6): 720-736. (XU H X, LŪ F T, SHI K, et al. Research Status and Development Trends of Deep-Sea Long-Term Resident Autonomous Underwater Vehicle Systems. ROBOT, 2023, 45(6): 720-736.) [3] SUBRAMANI B, VELUCHAMY M. Pixel Intensity Optimization and Detail-Preserving Contextual Contrast Enhancement for Underwater Images. Optics & Laser Technology, 2025, 180. DOI: 10.1016/j.optlastec.2024.111464. [4] ZHANG W D, WANG Y D, LI C Y. Underwater Image Enhancement by Attenuated Color Channel Correction and Detail Preserved Contrast Enhancement. IEEE Journal of Oceanic Engineering, 2022, 47(3): 718-735. [5] 黄峰,刘鸿伟,沈英,等.基于多尺度空间自适应注意力网络的轻量级图像超分辨率方法.模式识别与人工智能, 2025, 38(1): 36-50. (HUANG F, LIU H W, SHEN Y, et al. Lightweight Image Super-Resolution Reconstruction Method Based on Multi-scale Spatial Ada-ptive Attention Network. Pattern Recognition and Artificial Intelligence, 2025, 38(1): 36-50.) [6] PENG L T, ZHU C L, BIAN L H. U-Shape Transformer for Underwater Image Enhancement. IEEE Transactions on Image Processing, 2023, 32: 3066-3079. [7] GAO Z, YANG J, JIANG F L, et al. DDformer: Dimension Decomposition Transformer with Semi-supervised Learning for underwater Image Enhancement. Knowledge-Based Systems, 2024, 297. DOI: 10.1016/j.knosys.2024.111977. [8] YANG J, ZHU S B, LIANG H, et al. PAFPT: Progressive Aggre-gator with Feature Prompted Transformer for Underwater Image Enhancement. Expert Systems with Applications, 2025, 262. DOI: 10.1016/j.eswa.2024.125539. [9] ZHAO C, CAI W L, DONG C Y, et al. Wavelet-Based Fourier Information Interaction with Frequency Diffusion Adjustment for Underwater Image Restoration // Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2024: 8281-8291. [10] QI Q, LI K Q, ZHENG H Y, et al. SGUIE-Net: Semantic Attention Guided Underwater Image Enhancement with Multi-scale Per-ception. IEEE Transactions on Image Processing, 2022, 31: 6816-6830. [11] REZVANI Z, KATANFOROUSH A, VAN WEZEL R, et al. Glo-bal and Local Visual Processing: Influence of Perceptual Field Va-riables[C/OL].[2025-04-15]. https://arxiv.org/pdf/2307.04435. [12] JR DREWS P, DO NASCIMENTO E, MORAES F, et al. Transmission Estimation in Underwater Single Images // Proc of the IEEE International Conference on Computer Vision Workshops. Washington, USA: IEEE, 2013: 825-830. [13] PENG Y T, COSMAN P C. Underwater Image Restoration Based on Image Blurriness and Light Absorption. IEEE Transactions on Image Processing, 2017, 26(4): 1579-1594. [14] SONG W, WANG Y, HUANG D M, et al. A Rapid Scene Depth Estimation Model Based on Underwater Light Attenuation Prior for Underwater Image Restoration // Proc of the 19th Pacific-Rim Conference on Multimedia. Berlin, Germany: Springer, 2018: 678-688. [15] ZHUANG P X, WU J M, PORIKLI F, et al. Underwater Image Enhancement with Hyper-Laplacian Reflectance Priors. IEEE Tran-sactions on Image Processing, 2022, 31: 5442-5455. [16] CONG R M, YANG W Y, ZHANG W, et al. PUGAN: Physical Model-Guided Underwater Image Enhancement Using GAN with Dual-Discriminators. IEEE Transactions on Image Processing, 2023, 32: 4472-4485. [17] KHAN M R, MISHRA P, MEHTA N, et al. Spectroformer: Multi-domain Query Cascaded Transformer Network for Underwater Image Enhancement // Proc of the IEEE/CVF Winter Conference on Applications of Computer Vision. Washington, USA: IEEE, 2024: 1443-1452. [18] HUANG S R, WANG K Y, LIU H, et al. Contrastive Semi-supervised Learning for Underwater Image Restoration via Reliable Bank // Proc of the IEEE/CVF Conference on Computer Vision and Pa-ttern Recognition. Washington, USA: IEEE, 2023: 18145-18155. [19] 张帆,黄丹,丁璐,等.基于对比学习的水下图像增强和检测方法.机器人, 2025, 47(1): 64-75. (ZHANG F, HUANG D, DING L, et al. Contrastive Learning-based Underwater Image Enhancement and Detection. ROBOT, 2025, 47(1): 64-75.) [20] FABBRI C, ISLAM M J, SATTAR J. Enhancing Underwater Ima-gery Using Generative Adversarial Networks // Proc of the IEEE International Conference on Robotics and Automation. Washington, USA: IEEE, 2018: 7159-7165. [21] LI C Y, ANWAR S, PORIKLI F. Underwater Scene Prior Inspired Deep Underwater Image and Video Enhancement. Pattern Recognition, 2020, 98. DOI: 10.1016/j.patcog.2019.107038. [22] LI C Y, GUO C L, REN W Q, et al. An Underwater Image Enhancement Benchmark Dataset and Beyond. IEEE Transactions on Image Processing, 2020, 29: 4376-4389. [23] WANG Y D, GUO J C, GAO H, et al. UIEC^2-Net: CNN-Based Underwater Image Enhancement Using Two Color Space. Signal Processing: Image Communication, 2021, 96. DOI: 10.1016/j.image.2021.116250. [24] REN T D, XU H Y, JIANG G Y, et al. Reinforced Swin-Convs Transformer for Simultaneous Underwater Sensing Scene Image Enhancement and Super-Resolution. IEEE Transactions on Geosci-ence and Remote Sensing, 2022, 60. DOI: 10.1109/TGRS.2022.3205061. [25] WANG D, MA L, LIU R S, et al. Semantic-Aware Texture-Structure Feature Collaboration for Underwater Image Enhancement // Proc of the International Conference on Robotics and Automation. Washington, USA: IEEE, 2022: 4598-4604. [26] SHAZEER N. GLU Variants Improve Transformer[C/OL]. [2025-04-15].https://arxiv.org/pdf/2002.05202. [27] DAUPHIN Y N, FAN A, AULI M, et al. Language Modeling with Gated Convolutional Networks[C/OL].[2025-04-15]. https://arxiv.org/pdf/1612.08083. [28] ISLAM M J, XIA Y Y, SATTAR J. Fast Underwater Image Enhancement for Improved Visual Perception. IEEE Robotics and Automation Letters, 2020, 5(2): 3227-3234. [29] LIU R S, FAN X, ZHU M, et al. Real-World Underwater Enhancement: Challenges, Benchmarks, and Solutions under Natural Light. IEEE Transactions on Circuits and Systems for Video Technology, 2020, 30(12): 4861-4875. [30] LI C Y, ANWAR S, HOU J H, et al. Underwater Image Enhancement via Medium Transmission-Guided Multi-color Space Embe-dding. IEEE Transactions on Image Processing, 2021, 30: 4985-5000. [31] QI H, ZHOU H Y, DONG J Y, et al. Deep Color-Corrected Multiscale Retinex Network for Underwater Image Enhancement. IEEE Transactions on Geoscience and Remote Sensing, 2024, 62. DOI: 10.1109/TGRS.2023.3338611. [32] LIU S B, FAN H J, WANG Q, et al. Wavelet-Pixel Domain Pro-gressive Fusion Network for Underwater Image Enhancement. Know-ledge-Based Systems, 2024, 299. DOI: 10.1016/j.knosys.2024.112049. [33] CHEN X Y, YU J Z, KONG S H, et al. Towards Real-Time Advancement of Underwater Visual Quality with GAN. IEEE Transactions on Industrial Electronics, 2019, 66(12): 9350-9359. [34] VENKATANATH N, PRANEETH D, BH M C, et al. Blind Image Quality Evaluation Using Perception Based Features // Proc of the 21st National Conference on Communications Mumbai. Washington, USA: IEEE, 2015. DOI: 10.1109/NCC.2015.7084843. [35] GE Z, LIU S T, WANG F, ,et al. YOLOX: Exceeding YOLO Series in. YOLOX: Exceeding YOLO Series in 2021[C/OL]. [2025-04-15]. https://arxiv.org/pdf/2107.08430.
2025, Vol. 38 
