簡易檢索 / 詳目顯示

回結果列表
研究生： 魯姵妤
Lu, Pei-Yu
論文名稱： 一個用於水下影像增強的高效混合特徵融合網路
An Efficient Hybrid Feature Fusion Network for Underwater Image Enhancement
指導教授： 戴顯權
Tai, Shen-Chuan
學位類別： 碩士
Master
系所名稱： 電機資訊學院 - 電機工程學系
Department of Electrical Engineering
論文出版年： 2026
畢業學年度： 114
語文別： 英文
論文頁數： 85
中文關鍵詞： 水下影像增強 、深度學習 、色彩平衡先驗 、注意力機制
外文關鍵詞： Underwater image enhancement, deep learning, color balance prior, attention mechanism
相關次數： 點閱：6  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 水下影像因光的吸收和散射特性，常面臨嚴重的色彩失真、對比度降低和細節損失等問題，限制了海洋探索和自主水下載具等應用。本論文提出一種新穎的深度學習框架來解決這些挑戰。
    我們的方法整合三個關鍵組件：色彩平衡先驗模塊採用雙統計白平衡技術校正色偏；細節恢復器結合多維度轉移注意力機制、輕量級殘差密集區塊和注意力聚合來恢復精細細節；特徵情境化器透過全局上下文聚合和特徵空間注意力整合多尺度特徵。整個框架採用編碼器-解碼器架構，使用結構相似性和學習感知相似性的複合損失函數訓練。
    在多個基準資料集上的廣泛實驗顯示，我們的方法在配對測試集UIEB、EUVP和LSUI上達到最先進的性能，展現優越的色彩恢復和細節保存能力。在非配對測試集RUIE上取得最佳整體影像品質，證明其強大的泛化能力。此外，我們的方法在所有比較方法中達到最低的計算成本，展現性能與效率的極佳平衡。消融實驗驗證了各組件的有效貢獻。本框架為水下影像增強提供完整且高效的解決方案，適合即時應用和資源受限系統。

    Underwater images suffer from severe color distortion, reduced contrast, and detail loss due to light absorption and scattering, limiting marine exploration and autonomous underwater vehicle applications. This Thesis presents a novel deep learning framework to address these challenges.
    The proposed method integrates three key components: a Color Balance Prior module for color correction, a Detail Restorer combining attention mechanisms and residual dense blocksfordetailrecovery, andaFeatureContextualizerformulti-scalefeatureintegration. The framework employs an encoder-decoder architecture trained with a composite loss function.
    Experiments demonstrate state-of-the-art performance on paired datasets UIEB, EUVP, and LSUI, and the unpaired dataset RUIE. The proposed method achieves the lowest computational cost among the compared methods. Consequently, it is highly suitable for real-time and resource-constrained systems.

    中文摘要 i Abstract ii Acknowledgements iii Contents iv ListofTables vii ListofFigures viii 1 Introduction 1 2 RelatedWorks 6 2.1 U-NetArchitecture 6 2.2 AttentionMechanismsandEfficientFeatureExtraction 7 2.2.1 ConvolutionalBlockAttentionModule(CBAM) 7 2.2.2 Multi-DconvHeadTransposedAttention(MDTA) 9 2.2.3 FNet:FFT-BasedTokenMixing 10 2.2.4 GatedMechanisms 11 2.2.5 ResidualDenseBlock(RDB) 11 2.3 TraditionalUnderwaterImageEnhancement 12 2.4 DeepLearning-BasedUnderwaterImageEnhancement 14 2.5 Prior-GuidedHybridEnhancement 16 3 TheProposedMethod 19 3.1 ColorEnhancer 20 3.2 ColorBalancePrior 22 3.3 DetailRestorer 24 3.3.1 QBlock 25 3.3.2 CBAMAggregation 28 3.4 FeatureContextualizer 29 3.4.1 AdjustColorTransformer 30 3.4.2 GatedCrossAttention 32 3.4.3 FNetSelfAttention 32 3.5 ScaleHarmonizer 34 3.6 LossFunction 35 3.6.1 PSNRLoss 35 3.6.2 SSIMLoss 36 3.6.3 LPIPSLoss 36 3.6.4 LossWeightingStrategy 37 4 ExperimentalResults 38 4.1 ExperimentalDataset 38 4.1.1 TrainingDataset 38 4.1.2 TestingDataset 39 4.2 ExperimentalSettings 43 4.2.1 TrainingHyperparameters 43 4.2.2 ImplementationEnvironment 43 4.3 ExperimentalResults 44 4.3.1 EvaluationMetrics 45 4.3.2 QuantitativeResults 48 4.3.3 VisualResultsComparison 53 4.4 AblationExperimentalResults 58 5 ConclusionsandFutureworks 60 5.1 Conclusions 60 5.2 FutureWork 61 References 63

    [1] L.Li,B.Dong,E.Rigall,T.Zhou,J.Dong,andG.Chen,“Marineanimalsegmentation,”IEEE Transactions on Circuits and Systems for Video Technology, vol. 32, no. 4, pp. 2303–2314, 2021.
    [2] M. Bernardi et al., “Aurora, a multi-sensor dataset for robotic ocean exploration,”International Journal of Robotics Research, vol. 41, no. 5, pp. 461–469, 2022.
    [3] S. S. Afzal et al., “Battery-free wireless imaging of underwater environments,”Nature Communications, vol. 13, no. 1, p. 5546, 2022.
    [4] D. K. Rout, M. Kapoor, B. N. Subudhi, V. Thangaraj, V. Jakhetiya, and A. Bansal,“Underwater visual surveillance: A comprehensive survey,”Ocean Engineering, vol. 309, p. 118367, 2024.
    [5] M. J. Islam, A. Q. Li, Y. A. Girdhar, and I. Rekleitis, “Computer vision applications in underwater robotics and oceanography,”in Proceedings of the Computer Vision: Challenges, Trends, and Opportunities, pp. 173–204, 2024.
    [6] S. Raveendran, M. D. Patil, and G. K. Birajdar, “Underwater image enhancement: A comprehensivereview,recenttrends,challengesandapplications,”ArtificialIntelligence Review, vol. 54, no. 7, pp. 5413–5467, 2021.
    [7] J.S.Jaffe, “Computermodelingandthedesignofoptimalunderwaterimagingsystems,”IEEE Journal of Oceanic Engineering, vol. 40, no. 2, pp. 314–324, 2015. 63
    [8] Y.-T. Peng and P. C. Cosman, “Underwater image restoration based on image blurriness and light absorption,”IEEE Transactions on Image Processing, vol. 26, no. 4, pp. 1579–1594, 2017.
    [9] C.O.Ancuti, C.Ancuti, C. D.Vleeschouwer, and P. Bekaert, “Color balance and fusion for underwater image enhancement,”IEEE Transactions on Image Processing, vol. 27, no. 1, pp. 379–393, 2018.
    [10] S. An, L. Xu, Z. Deng, and H. Zhang, “Hfm: A hybrid fusion method for underwater image enhancement,”Engineering Applications of Artificial Intelligence, vol. 127, p. 107219, 2024.
    [11] W. Zhang, L. Zhou, P. Zhuang, G. Li, X. Pan, W. Zhao, and C. Li, “Underwater image enhancement via weighted wavelet visual perception fusion,”IEEE Transactions on Circuits and Systems for Video Technology, vol. 34, no. 4, pp. 2469–2483, 2024.
    [12] W. Zhang, Q. Liu, H. Lu, J. Wang, and J. Liang, “Underwater image enhancement via wavelet decomposition fusion of advantage contrast,”IEEE Transactions on Circuits and Systems for Video Technology, vol. 35, no. 8, pp. 7807–7820, 2025.
    [13] Z. Wang, D. Zhou, Z. Li, Z. Yuan, and C. Yang, “Underwater image enhancement via adaptive color correction and stationary wavelet detail enhancement,”IEEE Access, vol. 12, pp. 11066–11082, 2024.
    [14] G. Hou, N. Li, P. Zhuang, K. Li, H. Sun, and C. Li, “Non-uniform illumination under water image restoration via illumination channel sparsity prior,”IEEE Transactions on Circuits and Systems for Video Technology, vol. 34, no. 2, pp. 799–814, 2024. 64
    [15] Y. Zhou, Q. Wu, K. Yan, L. Feng, and W. Xiang, “Underwater image restoration using color-line model,”IEEE Transactions on Circuits and Systems for Video Technology, vol. 29, no. 3, pp. 907–911, 2019.
    [16] C. Li, S. Anwar, and F. Porikli, “Underwater scene prior inspired deep underwater image and video enhancement,”Pattern Recognition, vol. 98, p. 107038, 2020.
    [17] J. Wen, J. Cui, Z. Zhao, R. Yan, Z. Gao, L. Dou, and B. M. Chen, “Syreanet: A physically guided underwater image enhancement framework integrating synthetic and real images,”in Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pp. 5177–5183, 2023.
    [18] G. Han, M. Wang, H. Zhu, and C. Lin, “Uiegan: Adversarial learning based photo realistic image enhancement for intelligent underwater environment perception,”IEEE Transactions on Geoscience and Remote Sensing, vol. 61, p. 5611514, 2023.
    [19] H.Yanetal., “Uw-cyclegan: Model-driven cyclegan for underwater image restoration,”IEEE Transactions on Geoscience and Remote Sensing, vol. 61, p. 4207517, 2023.
    [20] L.Peng,C.Zhu,andL.Bian,“U-shapetransformerforunderwaterimageenhancement,”IEEE Transactions on Image Processing, vol. 32, pp. 3066–3079, 2023.
    [21] Z.Gao,J.Yang,F.Jiang,X.Jiao,K.Dashtipour,M.Gogate,andA.Hussain,“Ddformer: Dimension decomposition transformer with semi-supervised learning for underwater image enhancement,”Knowledge-Based Systems, vol. 297, p. 111977, 2024.
    [22] C. Zhao, W. Cai, C. Dong, and Z. Zeng, “Toward sufficient spatial-frequency interac tion for gradient-aware underwater image enhancement,”in Proceedings of the IEEE 65 International Conference on Acoustics, Speech and Signal Processing, pp. 3220–3224, 2024.
    [23] S. Liu, H. Fan, Q. Wang, Z. Han, Y. Guan, and Y. Tang, “Wavelet-pixel domain progressive fusion network for underwater image enhancement,”Knowledge-Based Systems, vol. 299, p. 112049, 2024.
    [24] Y. Liu, Q. Jiang, X. Wang, T. Luo, and J. Zhou, “Underwater image enhancement with cascaded contrastive learning,”IEEE Transactions on Multimedia, vol. 27, pp. 1512–1525, 2025.
    [25] A. Saleh, M. Sheaves, D. Jerry, and M. R. Azghadi, “Adaptive deep learning frame work for robust unsupervised underwater image enhancement,”Expert Systems with Applications, vol. 268, p. 126314, 2025.
    [26] X. Guo, X. Chen, S. Wang, and C.-M. Pun, “Underwater image restoration through a prior guided hybrid sense approach and extensive benchmark analysis,”IEEE Trans actions on Circuits and Systems for Video Technology, vol. 35, no. 5, pp. 4784–4800, 2025.
    [27] M. J. Islam, Y. Xia, and J. Sattar, “Fast underwater image enhancement for improved visual perception,”IEEE Robotics and Automation Letters, vol. 5, no. 2, pp. 3227–3234, 2020.
    [28] X. Xue, Z. Li, L. Ma, Q. Jia, R. Liu, and X. Fan, “Investigating intrinsic degradation factors by multi-branch aggregation for real-world underwater image enhancement,”Pattern Recognition, vol. 133, p. 109041, 2023. 66
    [29] O.Ronneberger, P.Fischer, andT.Brox, “U-net: Convolutionalnetworksforbiomedical image segmentation,”in Proceedings of the Medical Image Computing and Computer Assisted Intervention– MICCAI 2015, pp. 234–241, 2015.
    [30] Z. Zhou, M. M. R. Siddiquee, N. Tajbakhsh, and J. Liang, “Unet++: A nested u-net architecture for medical image segmentation,”in Proceedings of the Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, pp. 3–11, 2018.
    [31] O. Oktay, J. Schlemper, L. L. Folgoc, M. Lee, M. Heinrich, K. Misawa, K. Mori, S. McDonagh, N. Y. Hammerla, B. Kainz, B. Glocker, and D. Rueckert, “Attention u-net: Learning where to look for the pancreas,”arXiv preprint arXiv:1804.03999, 2018.
    [32] ¨ Ozg¨un C¸ic¸ek, A. Abdulkadir, S. S. Lienkamp, T. Brox, and O. Ronneberger, “3d u net: Learning dense volumetric segmentation from sparse annotation,”in Proceedings of the Medical Image Computing and Computer-Assisted Intervention– MICCAI, pp. 424–432, 2016.
    [33] J. Hu, L. Shen, and G. Sun, “Squeeze-and-excitation networks,”in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 7132–7141, 2018.
    [34] S. Woo, J. Park, J.-Y. Lee, and I. S. Kweon, “Cbam: Convolutional block attention module,”in Proceedings of the European Conference on Computer Vision (ECCV), pp. 3–19, 2018. 67
    [35] S. W. Zamir, A. Arora, S. Khan, M. Hayat, F. S. Khan, and M.-H. Yang, “Restormer: Efficient transformer for high-resolution image restoration,”in Proceedings of the IEEE/CVFConferenceonComputerVisionandPatternRecognition(CVPR),pp.5718 5729, 2022.
    [36] J. Lee-Thorp, J. Ainslie, I. Eckstein, and S. Onta˜ n´ on, “Fnet: Mixing tokens with fourier transforms,”in Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp. 4296–4313, 2022.
    [37] R. K. Srivastava, K. Greff, and J. Schmidhuber, “Highway networks,”arXiv preprint arXiv:1505.00387, 2015.
    [38] Y. Zhang, Y. Tian, Y. Kong, B. Zhong, and Y. Fu, “Residual dense network for image super-resolution,”in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 2472–2481, 2018.
    [39] K. He, J. Sun, and X. Tang, “Single image haze removal using dark channel prior,”in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1956–1963, 2009.
    [40] P. L. J. Drews, E. R. Nascimento, S. S. C. Botelho, and M. F. M. Campos, “Underwater depth estimation and image restoration based on single images,”IEEE Computer Graphics and Applications, vol. 36, no. 2, pp. 24–35, 2016.
    [41] A. Galdran, D. Pardo, A. Pic´ on, and A. Alvarez-Gila, “Automatic red-channel underwater image restoration,”Journal of Visual Communication and Image Representation, vol. 26, pp. 132–145, 2015.68
    [42] Q. Zhu, J. Mai, and L. Shao, “A fast single image haze removal algorithm using color attenuation prior,”IEEE Transactions on Image Processing, vol. 24, no. 11, pp. 3522 3533, 2015.
    [43] Y.-T. Peng, K. Cao, and P. C. Cosman, “Generalization of the dark channel prior for single image restoration,”IEEE Transactions on Image Processing, vol. 27, no. 6, pp. 2856–2868, 2018.
    [44] X. Fu, P. Zhuang, Y. Huang, Y. Liao, X.-P. Zhang, and X. Ding, “A retinex-based enhancing approach for single underwater image,”in Proceedings of the IEEE International Conference on Image Processing (ICIP), pp. 4572–4576, 2014.
    [45] A. S. A. Ghani and N. A. M. Isa, “Underwater image quality enhancement through integrated color model with rayleigh distribution,”Applied Soft Computing, vol. 27, pp. 219–230, 2015.
    [46] L. S. Saoud, M. Elmezain, A. Sultan, M. Heshmat, L. Seneviratne, and I. Hussain,“Seeing through the haze: A comprehensive review of underwater image enhancement techniques,”IEEE Access, vol. 12, pp.145206–145233, 2024.
    [47] S. Anwar and C. Li, “Diving deeper into underwater image enhancement: A survey,”Signal Processing: Image Communication, vol. 89, p. 115978, 2020.
    [48] C. Li, C. Guo, W. Ren, R. Cong, J. Hou, S. Kwong, and D. Tao, “An underwater image enhancement benchmarkdataset andbeyond,”IEEETransactions on ImageProcessing, vol. 29, pp. 4376–4389, 2020.69
    [49] C. Fabbri, M. J. Islam, and J. Sattar, “Enhancing underwater imagery using generative adversarial networks,”in ProceedingsoftheIEEEInternationalConferenceonRobotics and Automation (ICRA), pp.7159–7165, 2018.
    [50] R.Cong,W.Yang,W.Zhang,C.Li,C.-L.Guo,Q.Huang,andS.Kwong,“Pugan: Physical model-guided underwater image enhancement using gan with dual-discriminators,”IEEE Transactions on Image Processing, vol. 32, pp. 4472–4485, 2023.
    [51] M.Bharathi, A. Amsaveni, S. A.Dharani, S. Akilandeswari, and M. Sriram, “Underwater image enhancement using generative adversarial network,”in Proceedings of the 3rd International Conference on Advancements in Electrical, Electronics, Communication, Computing and Automation (ICAECA), pp. 1–5, 2025.
    [52] Z.Huang,J.Li,Z.Hua,andL.Fan,“Underwaterimageenhancementviaadaptivegroup attention-based multiscale cascade transformer,”IEEE Transactions on Instrumentation and Measurement, vol. 71, pp. 1–18, 2022.
    [53] P.Mishra, S.K.Vipparthi, andS.Murala, “U-enhance: Underwaterimageenhancement usingwavelettriple self-attention,”in Proceedings of the ComputerVision–ACCV2024 Workshops, pp. 87–104, 2025.
    [54] H. Zhang, H. Xu, X. Yu, X. Zhang, X. Gao, and C. Wu, “Cdf-uie: Leveraging crossdomain fusion for underwater image enhancement,”IEEE Transactions on Geoscience and Remote Sensing, vol. 63, pp. 1–15, 2025.
    [55] Z. Cheng, G. Fan, J. Zhou, M. Gan, and C. L. P. Chen, “Fdce-net: Underwater image enhancement with embedding frequency and dual color encoder,”IEEE Transactionson Circuits and Systems for Video Technology, vol. 35, no. 2, pp. 1728–1744, 2025.70
    [56] J. Zhou, J. Sun, C. Li, Q. Jiang, M. Zhou, K.-M. Lam, W. Zhang, and X. Fu, “Hclrnet: Hybrid contrastive learning regularization with locally randomized perturbation for underwater image enhancement,”International Journal of Computer Vision, vol. 132, no. 10, pp. 4132–4156, 2024.
    [57] S. Huang, K. Wang, H. Liu, J. Chen, and Y. Li, “Contrastive semi-supervised learning for underwater image restoration via reliable bank,”in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 18145–18155,2023.
    [58] Z. Wang, L. Shen, M. Xu, M. Yu, K. Wang, and Y. Lin, “Domain adaptation for underwater image enhancement,”IEEE Transactions on Image Processing, vol. 32, pp.1442–1457, 2023.
    [59] Z. Zhang, Z. Jiang, L. Ma, J. Liu, X. Fan, and R. Liu, “Hupe: Heuristic underwater perceptual enhancement with semantic collaborative learning,”International Journal of Computer Vision, vol. 133, no. 6, pp. 3259–3277, 2025.
    [60] Y. Tang, H. Kawasaki, and T. Iwaguchi, “Underwater image enhancement by transformer-based diffusion model with non-uniform sampling for skip strategy,”in Proceedings of the 31st ACM International Conference on Multimedia, pp. 5419–5427,2023.
    [61] D. Du, E. Li, L. Si, W. Zhai, F. Xu, J. Niu, and F. Sun, “Uiedp: Boosting underwater image enhancement with diffusion prior,”Expert Systems with Applications, vol. 259,p. 125271, 2025.71
    [62] A.Vaswani, N.Shazeer, N.Parmar, J.Uszkoreit, L. Jones, A. N. Gomez, LukaszKaiser, and I. Polosukhin, “Attention is all you need,”in Proceedings of the Advances in Neural Information Processing Systems, vol. 30, pp. 5998–6008, 2017.
    [63] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality assessment:From error visibility to structural similarity,”IEEE Transactions on Image Processing, vol. 13, no. 4, pp. 600–612, 2004.
    [64] R. Zhang, P. Isola, A. A. Efros, E. Shechtman, and O. Wang, “The unreasonable effectiveness of deep features as a perceptual metric,”in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 586–595, 2018.
    [65] R. Liu, X. Fan, M. Zhu, M. Hou, and Z. Luo, “Real-world underwater enhancement: Challenges, benchmarks, and solutions under natural light,”IEEE Transactions on Circuits and Systems for Video Technology, vol. 30, no. 12, pp. 4861–4875, 2020.
    [66] M.YangandA.Sowmya,“Anunderwatercolorimagequality evaluation metric,”IEEE Transactions on Image Processing, vol. 24, no. 12, pp. 6062–6071, 2015.
    [67] K. Panetta, C. Gao, and S. Agaian, “Human-visual-system-inspired underwater image quality measures,”IEEE Journal of Oceanic Engineering, vol. 41, no. 3, pp. 541–551,2016.
    [68] P. Guo, H. Liu, D. Zeng, T. Xiang, L. Li, and K. Gu, “An underwater image quality assessment metric,”IEEE Transactions on Multimedia, vol. 25, pp. 5093–5106, 2023.
    [69] Y.Liu,K.Gu,J.Cao,S.Wang,G.Zhai,J.Dong,andS.Kwong,“Uiqi: Acomprehensive quality evaluation index for underwater images,”IEEE Transactions on Multimedia, vol. 26, pp. 2560–2573, 2024.

    QR CODE