基于物理启发式深度神经网络的大气湍流退化图像复原
Restoration of Atmospheric Turbulence Degraded Images Based on Physics-Inspired Deep Neural Networks
摘要: 大气湍流会导致大气光传输数据模糊和扭曲，影响成像质量和后续分析处理。为了更好地恢复由大气湍流所影响的失真图像，本研究基于物理启发式深度神经网络提出了一种大气湍流退化图像复原算法。通过结合金字塔结构和注意力机制，提高特征提取的精度，从而改善复原图像质量。通过湍流模拟器模拟退化图像，提取大气湍流退化图像的先验信息，提升复原效果。实验结果表明，本研究在处理处理弱和中湍流环境下的大气光传输数据时复原能有效提高成像的视觉质量，降低成像的模糊和几何畸变程度。
Abstract: Atmospheric turbulence causes blurring and distortion in atmospheric optical transmission data, affecting imaging quality and subsequent analysis and processing. To better restore distorted images influenced by atmospheric turbulence, this study proposes an atmospheric turbulence degradation image restoration algorithm based on a physics-inspired deep neural network. By combining a pyramid structure and attention mechanisms, the accuracy of feature extraction is enhanced, thereby improving the quality of the restored images. Degraded images are simulated using a turbulence simulator, and prior information about atmospheric turbulence degradation is extracted to improve the restoration effect. Experimental results show that this study effectively enhances the visual quality of imaging and reduces blur and geometric distortion in atmospheric optical transmission data under weak and moderate turbulence environments.
文章引用：原泽, 周林华. 基于物理启发式深度神经网络的大气湍流退化图像复原[J]. 应用数学进展, 2025, 14(4): 205-217. https://doi.org/10.12677/aam.2025.144154

参考文献

[1] Yitzhaky, Y., Dror, I. and Kopeika, N.S. (1997) Restoration of Atmospherically Blurred Images According to Weather-Predicted Atmospheric Modulation Transfer Functions. Optical Engineering, 36, 30643-3072. [Google Scholar] [CrossRef] 
[2] Shacham, O., Haik, O. and Yitzhaky, Y. (2007) Blind Restoration of Atmospherically Degraded Images by Automatic Best Step-Edge Detection. Pattern Recognition Letters, 28, 2094-2103. [Google Scholar] [CrossRef] 
[3] Yasarla, R. and Patel, V.M. (2021) Learning to Restore Images Degraded by Atmospheric Turbulence Using Uncertainty. 2021 IEEE International Conference on Image Processing (ICIP), Anchorage, 19-22 September 2021, 1694-1698. [Google Scholar] [CrossRef] 
[4] Li, D., Mersereau, R.M. and Simske, S. (2007) Atmospheric Turbulence-Degraded Image Restoration Using Principal Components Analysis. IEEE Geoscience and Remote Sensing Letters, 4, 340-344. [Google Scholar] [CrossRef] 
[5] Tian, C., Fei, L., Zheng, W., Xu, Y., Zuo, W. and Lin, C. (2020) Deep Learning on Image Denoising: An Overview. Neural Networks, 131, 251-275. [Google Scholar] [CrossRef] [PubMed]
[6] Gampala, V., Sunil Kumar, M., Sushama, C. and Fantin Irudaya Raj, E. (2020) WITHDRAWN: Deep Learning Based Image Processing Approaches for Image Deblurring. Materials Today: Proceedings. (In Press) [Google Scholar] [CrossRef] 
[7] Gao, Z., Shen, C. and Xie, C. (2018) Stacked Convolutional Auto-Encoders for Single Space Target Image Blind Deconvolution. Neurocomputing, 313, 295-305. [Google Scholar] [CrossRef] 
[8] Zhu, X. and Milanfar, P. (2013) Removing Atmospheric Turbulence via Space-Invariant Deconvolution. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35, 157-170. [Google Scholar] [CrossRef] [PubMed]
[9] Dong, C., Loy, C.C., He, K. and Tang, X. (2014) Learning a Deep Convolutional Network for Image Super-Resolution. Computer Vision—ECCV 2014, Zurich, 6-12 September 2014, 184-199. [Google Scholar] [CrossRef] 
[10] He, K., Zhang, X., Ren, S. and Sun, J. (2016) Deep Residual Learning for Image Recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, 27-30 June 2016, 770-778. [Google Scholar] [CrossRef] 
[11] Jin, D., Chen, Y., Lu, Y., Chen, J., Wang, P., Liu, Z., et al. (2021) Neutralizing the Impact of Atmospheric Turbulence on Complex Scene Imaging via Deep Learning. Nature Machine Intelligence, 3, 876-884. [Google Scholar] [CrossRef] 
[12] Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., et al. (2017) Attention Is All You Need. Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, 4-9 December 2017, 6000-6010.
[13] Hu, J., Shen, L. and Sun, G. (2018) Squeeze-and-Excitation Networks. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, 18-23 June 2018, 7132-7141. [Google Scholar] [CrossRef] 
[14] Zhang, K., Zuo, W., Chen, Y., Meng, D. and Zhang, L. (2017) Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising. IEEE Transactions on Image Processing, 26, 3142-3155. [Google Scholar] [CrossRef] [PubMed]
[15] Fu, K., Chang, Z., Zhang, Y., Xu, G., Zhang, K. and Sun, X. (2020) Rotation-Aware and Multi-Scale Convolutional Neural Network for Object Detection in Remote Sensing Images. ISPRS Journal of Photogrammetry and Remote Sensing, 161, 294-308. [Google Scholar] [CrossRef] 
[16] Du, C. and Gao, S. (2017) Image Segmentation-Based Multi-Focus Image Fusion through Multi-Scale Convolutional Neural Network. IEEE Access, 5, 15750-15761. [Google Scholar] [CrossRef] 
[17] Zhang, T., Fang, B., Yuan, Y., Yan Tang, Y., Shang, Z., Li, D., et al. (2009) Multiscale Facial Structure Representation for Face Recognition under Varying Illumination. Pattern Recognition, 42, 251-258. [Google Scholar] [CrossRef] 
[18] Mei, Y., Fan, Y., Zhang, Y., Yu, J., Zhou, Y., Liu, D., et al. (2023) Pyramid Attention Network for Image Restoration. International Journal of Computer Vision, 131, 3207-3225. [Google Scholar] [CrossRef] 
[19] Yu, Y., Zhang, Y., Cheng, Z., Song, Z. and Tang, C. (2024) Multi-Scale Spatial Pyramid Attention Mechanism for Image Recognition: An Effective Approach. Engineering Applications of Artificial Intelligence, 133, Article 108261. [Google Scholar] [CrossRef] 

为你推荐