單張影像去霧是影像修復領域中重要且具挑戰性的任務，旨在從受大氣霧霾影響、導致視覺品質下降的影像中恢復清晰自然的畫面。霧霾會降低對比度、模糊邊緣並造成色彩偏移，進而影響後續電腦視覺系統的準確性與穩定性，尤其在自動駕駛與智慧監控等應用場景中更為關鍵。現有學習式方法雖具備強大特徵學習能力，但多依賴大型架構，導致模型參數量龐大且部署困難，同時，傳統卷積網路因感受野限制，難以有效建模全局資訊，且未充分強化細節特徵，影響去霧效果。為此，本論文提出一種輕量且高效的單張影像去霧架構，融合多個關鍵模組，包括差分卷積以加強邊緣與紋理感知，多尺度調製模組捕捉不同層次的霧霾特徵，並行大型卷積核實現多尺度局部與全局資訊整合，以及增強型並行注意力模組以強化空間與語意特徵表達。這些模組協同作用，有效提升細節保留與整體一致性，實現更自然且清晰的去霧效果。此外，透過自適應融合模組促進編碼器與解碼器間的資訊交流，進一步優化特徵整合與細節還原。在多個公開去霧資料集上的實驗結果表明，本方法在模型參數量與推理速度方面具明顯優勢，且去霧品質優異，整體性能可媲美甚至超越現有先進方法，展現良好的實用性與部署潛力。

Single image dehazing is a crucial and challenging task in the field of image restoration. Its goal is to recover clear and natural images from those degraded by atmospheric haze, which reduces contrast, blurs edges, and causes color distortion. These degradations adversely affect the accuracy and stability of subsequent computer vision systems, especially in applications such as autonomous driving and intelligent surveillance. Although existing learning-based methods demonstrate strong feature learning capabilities, they often rely on large-scale architectures that result in high model complexity and difficult deployment. Additionally, conventional convolutional neural networks are limited by their receptive fields, making it challenging to effectively model global information and sufficiently enhance fine details, thereby impacting dehazing performance. To address these issues, this Thesis proposes a lightweight and efficient single image dehazing framework that integrates several key modules. These include a differential convolution module to enhance edge and texture awareness, a multi-scale modulation module to capture haze features at different levels, a multi-scale parallel large convolution kernel to integrate local and global information across scales, and an enhanced parallel attention module to strengthen spatial and semantic feature representation. Together, these modules improve detail preservation and overall consistency, resulting in more natural and clearer dehazing outcomes. Furthermore, an adaptive fusion module is introduced to facilitate information exchange between the encoder and decoder, further optimizing feature integration and detail restoration. Extensive experiments on multiple public dehazing datasets demonstrate that the proposed method achieves significant advantages in model size and inference speed while delivering superior dehazing quality. Its overall performance is comparable to or surpasses current state-of-the-art approaches, showcasing strong practicality and deployment potential.

[1] C. Sakaridis, D. Dai, S. Hecker, and L. Van Gool, “Model adaptation with synthetic and real data for semantic dense foggy scene understanding,” in Proceedings of the European Conference on Computer Vision, pp. 687–704, 2018.
[2] K. He, J. Sun, and X. Tang, “Single image haze removal using dark channel prior,” in 2009 IEEE Conference on Computer Vision and Pattern Recognition, pp. 1956–1963, 2009.
[3] 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.
[4] B. Cai, X. Xu, K. Jia, C. Qing, and D. Tao, “Dehazenet: An end-to-end system for single image haze removal,” IEEE Transactions on Image Processing, vol. 25, no. 11, pp. 5187–5198, 2016.
[5] A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, M. Dehghani, M. Minderer, G. Heigold, S. Gelly et al., “An image is worth 16x16 words: Transformers for image recognition at scale,” arXiv preprint arXiv:2010.11929, 2020.
[6] Z. Wang, X. Cun, J. Bao, W. Zhou, J. Liu, and H. Li, “Uformer: A general u-shaped transformer for image restoration,” in 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 17 662–17 672, 2022.
[7] Y. Cui, W. Ren, X. Cao, and A. Knoll, “Revitalizing convolutional network for image restoration,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 46, no. 12, pp. 9423–9438, 2024.
[8] L. Lu, Q. Xiong, B. Xu, and D. Chu, “Mixdehazenet: Mix structure block for image dehazing network,” in 2024 International Joint Conference on Neural Networks, pp. 1–10, 2024.
[9] Z. Chen, Z. He, and Z.-M. Lu, “Dea-net: Single image dehazing based on detail-enhanced convolution and content-guided attention,” IEEE Transactions on Image Processing, vol. 33, pp. 1002–1015, 2024.
[10] S. Narasimhan and S. Nayar, “Contrast restoration of weather degraded images,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, no. 6, pp. 713–724, 2003.
[11] G. Meng, Y. Wang, J. Duan, S. Xiang, and C. Pan, “Efficient image dehazing with boundary constraint and contextual regularization,” in 2013 IEEE International Conference on Computer Vision, pp. 617–624, 2013.
[12] D. Berman, T. Treibitz, and S. Avidan, “Non-local image dehazing,” in 2016 IEEE Conference on Computer Vision and Pattern Recognition, pp. 1674–1682, 2016.
[13] R. T. Tan, “Visibility in bad weather from a single image,” in 2008 IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–8, 2008.
[14] H. Wu, Y. Qu, S. Lin, J. Zhou, R. Qiao, Z. Zhang, Y. Xie, and L. Ma, “Contrastive learning for compact single image dehazing,” in 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 10546–10555, 2021.
[15] X. Qin, Z. Wang, Y. Bai, X. Xie, and H. Jia, “Ffa-net: Feature fusion attention network for single image dehazing,” in Proceedings of the AAAI conference on artificial intelligence, vol. 34, no. 07, pp. 11908–11915, 2020.
[16] P. Luo, G. Xiao, X. Gao, and S. Wu, “Lkd-net: Large kernel convolution network for single image dehazing,” in 2023 IEEE International Conference on Multimedia and Expo, pp. 1601–1606, 2023.
[17] Y. Cui, W. Ren, X. Cao, and A. Knoll, “Image restoration via frequency selection,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 46, no. 2, pp. 1093–1108, 2024.
[18] Y. Cui, Y. Tao, Z. Bing, W. Ren, X. Gao, X. Cao, K. Huang, and A. Knoll, “Selective frequency network for image restoration,” in The Eleventh International Conference on Learning Representations, 2023.
[19] O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional networks for biomedical image segmentation,” in International Conference on Medical image computing and computer-assisted intervention. Springer, pp. 234–241, 2015.
[20] S.-J. Cho, S.-W. Ji, J.-P. Hong, S.-W. Jung, and S.-J. Ko, “Rethinking coarse-to-fine approach in single image deblurring,” in 2021 IEEE/CVF International Conference on Computer Vision, pp. 4621–4630, 2021.
[21] T. Ojala, M. Pietikainen, and T. Maenpaa, “Multiresolution gray-scale and rotation invariant texture classification with local binary patterns,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, no. 7, pp. 971–987, 2002.
[22] F. Juefei-Xu, V. Naresh Boddeti, and M. Savvides, “Local binary convolutional neural networks,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 19–28, 2017.
[23] Z. Yu, C. Zhao, Z. Wang, Y. Qin, Z. Su, X. Li, F. Zhou, and G. Zhao, “Searching central difference convolutional networks for face anti-spoofing,” in 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 5294–5304, 2020.
[24] Z. Yu, Y. Qin, H. Zhao, X. Li, and G. Zhao, “Dual-cross central difference network for face anti-spoofing,” arXiv preprint arXiv:2105.01290, 2021.
[25] J. Yang, L. Bai, Y. Sun, C. Tian, M. Mao, and G. Wang, “Pixel difference convolutional network for rgb-d semantic segmentation,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 34, no. 3, pp. 1481–1492, 2024.
[26] Z. Yu, J. Wan, Y. Qin, X. Li, S. Z. Li, and G. Zhao, “Nas-fas: Static-dynamic central difference network search for face anti-spoofing,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 9, pp. 3005–3023, 2021.
[27] Z. Yu, X. Li, X. Niu, J. Shi, and G. Zhao, “Face anti-spoofing with human material perception,” in European conference on computer vision, pp. 557–575, 2020.
[28] Z. Wang, Z. Yu, C. Zhao, X. Zhu, Y. Qin, Q. Zhou, F. Zhou, and Z. Lei, “Deep spatial gradient and temporal depth learning for face anti-spoofing,” in 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 5041–5050, 2020.
[29] Y. Qin, C. Zhao, X. Zhu, Z. Wang, Z. Yu, T. Fu, F. Zhou, J. Shi, and Z. Lei, “Learning meta model for zero-and few-shot face anti-spoofing,” in Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34, no. 07, pp. 11916–11923, 2020.
[30] B. Li, W. Ren, D. Fu, D. Tao, D. Feng, W. Zeng, and Z. Wang, “Benchmarking single-image dehazing and beyond,” IEEE Transactions on Image Processing, vol. 28, no. 1, pp. 492–505, 2019.
[31] Y. Liu, L. Zhu, S. Pei, H. Fu, J. Qin, Q. Zhang, L. Wan, and W. Feng, “From synthetic to real: Image dehazing collaborating with unlabeled real data,” in Proceedings of the 29th ACM international conference on multimedia, pp. 50–58, 2021.