基于深度学习的半监督图像去雾网络
Semi-Supervised Image Dehazing Network Based on Deep Learning
DOI: 10.12677/csa.2024.144089, PDF,    国家科技经费支持
作者: 徐 航, 周 杰, 赵 丽*, 胡 杰：温州大学浙江省安全应急智能信息技术重点实验室，浙江 温州
关键词: 计算机视觉；图像去雾；深度学习；半监督学习；图像增强；Computer Vision； Image Dehazing； Deep Learning； Semi Supervised Learning； Image Enhancement
摘要: 针对真实雾图数据集稀缺，难以满足深度网络数据需求的问题，提出了一种基于深度学习的图像增强模型，利用随机掩码来模拟真实雾霾中的不均匀现象，以提高增强雾图的真实性。基于这个模型，搭建了一个半监督图像去雾网络，通过对增强图像自监督进一步提高去雾模型在真实世界的泛化性。实验结果表明，所提出的模型在合成雾图数据集和真实世界雾图数据集上都优于传统的图像去雾算法。
Abstract: A deep learning-based image enhancement model is proposed to tackle the challenge of limited real haze image datasets and the difficulty in meeting the data demands of deep networks. The model utilizes random masking to simulate the non-uniformity phenomenon in real haze, thereby enhancing the authenticity of the haze image. Based on this model, a semi-supervised image dehazing network was constructed to further enhance the generalization of the dehazing model in real-world scenarios by leveraging image self-supervision. Experimental results demonstrate that the proposed model outperforms traditional image dehazing algorithms on both synthetic haze image datasets and real-world haze image datasets.
文章引用：徐航, 周杰, 赵丽, 胡杰. 基于深度学习的半监督图像去雾网络[J]. 计算机科学与应用, 2024, 14(4): 193-200. https://doi.org/10.12677/csa.2024.144089

参考文献

[1] Li, Y., Zhao, H., Qi, X., Wang, L., Li, Z., Sun, J. and Jia, J. (2021) Fully Convolutional Networks for Panoptic Segmentation. Proceedings of 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville, 20-25 June 2021, 214-223. [Google Scholar] [CrossRef] 
[2] Li, W., Liu, X. and Yuan, Y. (2022) Sigma: Semantic-Complete Graph Matching for Domain Adaptive Object Detection. Proceedings of 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, 18-24 June 2022, 5281-5290. [Google Scholar] [CrossRef] 
[3] He, K., Sun, J. and Tang, X. (2010) Single Image Haze Removal Using Dark Channel Prior. IEEE Transactions on Pattern Analysis and Machine Intelligence, 33, 2341-2353. [Google Scholar] [CrossRef] 
[4] Berman, D., Treibitz, T. and Avidan, S. (2016) Non-Local Image Dehazing. Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, 27-30 June 2016, 1674-1682. [Google Scholar] [CrossRef] 
[5] Zhu, Q., Mai, J. and Shao, L. (2015) A Fast Single Image Haze Removal Algorithm Using Color Attenuation Prior. IEEE Transactions on Image Processing, 24, 3522-3533. [Google Scholar] [CrossRef] 
[6] Ling, P., Chen, H., Tan, X., et al. (2023) Single Image Dehazing Using Saturation Line Prior. IEEE Transactions on Image Processing, 32, 3238-3253. [Google Scholar] [CrossRef] 
[7] Cai, B., Xu, X., Jia, K., et al. (2016) Dehazenet: An End-to-End System for Single Image Haze Removal. IEEE Transactions on Image Processing, 25, 5187-5198. [Google Scholar] [CrossRef] 
[8] Li, B., Gou, Y., Gu, S., Liu, J.Z., Zhou, J.T. and Peng, X. (2021) You Only Look Yourself: Unsupervised and Untrained Single Image Dehazing Neural Network. International Journal of Computer Vision, 129, 1754-1767. [Google Scholar] [CrossRef] 
[9] Zhang, H. and Patel, V.M. (2018) Densely Connected Pyramid Dehazing Network. Proceedings of 2018 IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, 18-23 June 2018, 3194-3203. [Google Scholar] [CrossRef] 
[10] Chen, Z., He, Z. and Lu, Z.M. (2024) DEA-Net: Single Image Dehazing Based on Detail-Enhanced Convolution and Content-Guided Attention. IEEE Transactions on Image Processing, 33, 1002-1015. [Google Scholar] [CrossRef] 
[11] An, S., Huang, X., Wang, L., et al. (2022) Semi-Supervised Image Dehazing Network. The Visual Computer, 38, 2041-2055. [Google Scholar] [CrossRef] 
[12] Sun, Z., Zhang, Y., Bao, F., et al. (2022) SADnet: Semi-Supervised Single Image Dehazing Method Based on an Attention Mechanism. ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM), 18, 1-23. [Google Scholar] [CrossRef] 
[13] Li, L., Dong, Y., Ren, W., et al. (2019) Semi-Supervised Image Dehazing. IEEE Transactions on Image Processing, 29, 2766-2779. [Google Scholar] [CrossRef] 
[14] Shyam, P., Yoon, K.J. and Kim, K.S. (2021) Towards Domain Invariant Single Image Dehazing. Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 35, Online, 2-9 February 2021, 9657-9665. [Google Scholar] [CrossRef] 
[15] Isola, P., Zhu, J.-Y., Zhou, T. and Efros, A.A. (2017) Image-to-Image Translation with Conditional Adversarial Networks. Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, 21-26 July 2017, 1125-1134. [Google Scholar] [CrossRef] 
[16] Zhao, S., Zhang, L., Shen, Y. and Zhou, Y. (2021) Refinednet: A Weakly Supervised Refinement Framework for Single Image Dehazing. IEEE Transactions on Image Processing, 30, 3391-3404. [Google Scholar] [CrossRef] 
[17] Yang, Y., Wang, C., Liu, R., Zhang, L., Guo, X. and Tao, D. (2022) Self-Augmented Unpaired Image Dehazing via Density and Depth Decomposition. Proceedings of 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, 18-24 June 2022, 2037-2046. [Google Scholar] [CrossRef] 
[18] Qu, Y., Chen, Y., Huang, J. and Xie, Y. (2019) Enhanced Pix2pix Dehazing Network. Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, 15-20 June 2019, 8160-8168. [Google Scholar] [CrossRef] 
[19] Dong, Y., Li, Y., Dong, Q., Zhang, H. and Chen, S. (2022) Semi-Supervised Domain Alignment Learning for Single Image Dehazing. IEEE Transactions on Cybernetics, 53, 7238-7250. [Google Scholar] [CrossRef]