基于OGS-HL的遥感图像混合噪声去除算法
Hybrid Noise Removal Algorithm of Remote Sensing Image Based on OGS-HL
摘要: 遥感图像在成像过程中容易受到混合噪声污染，包括高斯噪声、条纹噪声和脉冲噪声等。这些混合噪声降低了遥感图像的质量，限制了其后续应用。为了解决这一问题，首先，通过对遥感图像的梯度值进行统计分析，发现遥感图像的空间梯度值是符合重尾分布，因此，设计了关于遥感图像空间梯度值的OGS-HL正则项来去除混合噪声模型，该正则项不仅可以减少全变分带来的阶梯效应，而且还可以对图像的梯度值进行合理的稀疏表示；其次，针对条纹噪声，考虑其具有低秩性且使用核范数来约束，而稀疏噪声则具有全局稀疏分布，并且采用L1范数来约束；最后，采用交替方向乘子法和Majorization-Minimization算法来求解所提出的模型。通过与现有的算法进行比较，结果表明我们提出的算法在去除高水平混合噪声方面具有良好的效果。
Abstract: Remote sensing images are easily contaminated by mixed noise in the imaging process, including Gaussian noise, fringe noise and impulse noise. These mixed noises degrade the quality of remote sensing images and limit their subsequent applications. In order to solve this problem, firstly, through the statistical analysis of the gradient value of the remote sensing image, it is found that the spatial gradient value of the remote sensing image conforms to the heavy-tailed distribution. Therefore, the OGS-HL regularization term about the spatial gradient value of the remote sensing image is designed to remove the mixed noise model. Moreover, the gradient value of the image can be reasonably sparse represented. Secondly, the stripe noise was considered to have low rank and was constrained by the nuclear norm, while the sparse noise had global sparse distribution and was constrained by the L1 norm. Finally, the alternating direction method of multipliers and the Majorization-Minimization algorithm are used to solve the proposed model. By comparing with the existing algorithms, the results show that our proposed algorithm has a good effect in removing high-level mixed noise.
文章引用：张阿松. 基于OGS-HL的遥感图像混合噪声去除算法[J]. 图像与信号处理, 2024, 13(2): 163-178. https://doi.org/10.12677/jisp.2024.132015

参考文献

[1] Rudin, L.I., Osher, S. and Fatemi, E. (1992) Nonlinear Total Variation Based Noise Removal Algorithms. Physica D: Nonlinear Phenomena, 60, 259-268. [Google Scholar] [CrossRef] 
[2] Wang, Y., Yang, J., Yin, W., et al. (2008) A New Alternating Minimization Algorithm for Total Variation Image Reconstruction. SIAM Journal on Imaging Sciences, 1, 248-272. [Google Scholar] [CrossRef] 
[3] He, W., Zhang, H., Zhang, L., et al. (2015) Total-Variation-Regularized Low-Rank Matrix Factorization for Hyperspectral Image Restoration. IEEE Transactions on Geoscience and Remote Sensing, 54, 178-188. [Google Scholar] [CrossRef] 
[4] Yin, M., Adam, T., Paramesran, R., et al. (2022) An ℓ0-Overlapping Group Sparse Total Variation for Impulse Noise Image Restoration. Signal Processing: Image Communication, 102, Article 116620. [Google Scholar] [CrossRef] 
[5] 杨俊慈, 马明溪, 张俊, 吴朝明, 邓承志. 分数阶全变分泊松去噪的快速线性化增广拉格朗日方法[J]. 南昌工程学院学报, 2022(41): 97-101.
[6] Aggarwal, H.K. and Majumdar, A. (2016) Hyperspectral Image Denoising Using Spatio-Spectral Total Variation. IEEE Geoscience and Remote Sensing Letters, 13, 442-446. [Google Scholar] [CrossRef] 
[7] Chang, Y., Yan, L., Fang, H., et al. (2015) Anisotropic Spectral-Spatial Total Variation Model for Multispectral Remote Sensing Image Destriping. IEEE Transactions on Image Processing, 24, 1852-1866. [Google Scholar] [CrossRef] 
[8] Takeyama, S., Ono, S. and Kumazawa, I. (2020) A Constrained Convex Optimization Approach to Hyperspectral Image Restoration with Hybrid Spatio-Spectral Regularization. Remote Sensing, 12, Article 3541. [Google Scholar] [CrossRef] 
[9] 陈育群, 陈颖频, 林凡, 王灵芝. 基于高阶交叠组稀疏正则项的图像恢复方法[J]. 科学技术与工程, 2020(20): 13747-13756.
[10] Lu, X., Wang, Y. and Yuan, Y. (2013) Graph-Regularized Low-Rank Representation for Destriping of Hyperspectral images. IEEE Transactions on Geoscience and Remote Sensing, 51, 4009-4018. [Google Scholar] [CrossRef] 
[11] Chang, Y., Yan, L., Wu, T., et al. (2016) Remote Sensing Image Stripe Noise Removal: From Image Decomposition Perspective. IEEE Transactions on Geoscience and Remote Sensing, 54, 7018-7031. [Google Scholar] [CrossRef] 
[12] 石建飞, 唐玉波, 孙裕超. 结合非局部先验性与加权核范数最小化的声纳图像去噪[J]. 电声技术, 2020(44): 17-21.
[13] Zhang, H., Cai, J., He, W., et al. (2021) Double Low-Rank Matrix Decomposition for Hyperspectral Image Denoising and Destriping. IEEE Transactions on Geoscience and Remote Sensing, 60, Article No. 5502619. [Google Scholar] [CrossRef] 
[14] Jon, K., Liu, J., Wang, X., et al. (2021) Weighted Hyper-Laplacian Prior with Overlapping Group Sparsity for Image Restoration under Cauchy Noise. Journal of Scientific Computing, 87, Article No. 64. [Google Scholar] [CrossRef] 
[15] 余义斌, 吴承鑫, 彭念, 袁仕芳. 基于超拉普拉斯先验与核谱特性噪声图像盲去模糊[J]. 工程数学学报, 2018(35): 648-654.
[16] Cheng, M.H., Huang, T.Z., Zhao, X.L., et al. (2019) A Variational Model with Hybrid Hyper-Laplacian Priors for Retinex. Applied Mathematical Modelling, 66, 305-321. [Google Scholar] [CrossRef] 
[17] Zhang, H., He, W., Zhang, L., et al. (2013) Hyperspectral Image Restoration Using Low-Rank Matrix Recovery. IEEE Transactions on Geoscience and Remote Sensing, 52, 4729-4743. [Google Scholar] [CrossRef] 
[18] Chen, Y., He, W., Yokoya, N., et al. (2019) Hyperspectral Image Restoration Using Weighted Group Sparsity-Regularized Low-Rank Tensor Decomposition. IEEE Transactions on Cybernetics, 50, 3556-3570. [Google Scholar] [CrossRef] 
[19] Zhuang, L. and Ng, M.K. (2020) Hyperspectral Mixed Noise Removal by L1-Norm-Based Subspace Representation. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 13, 1143-1157. [Google Scholar] [CrossRef] 
[20] Chen, Y., Huang, T.Z., He, W., et al. (2021) Hyperspectral Image Denoising Using Factor Group Sparsity-Regularized Nonconvex Low-Rank Approximation. IEEE Transactions on Geoscience and Remote Sensing, 60, 1-16. [Google Scholar] [CrossRef] 
[21] Xu, S., Zhang, J. and Zhang, C. (2022) Hyperspectral Image Denoising by Low-Rank Models with Hyper-Laplacian Total Variation Prior. Signal Processing, 201, Article 108733. [Google Scholar] [CrossRef]