基于快速非局部均值和超分辨率重建的图像降噪算法

doi:10.3969/j.issn.1000-1093.2021.08.016

摘要/Abstract

摘要： 针对实际图像噪声强度变化范围宽，目前已有的图像降噪算法多数只能用于处理范围有限的噪声情况，结合快速非局部均值和基于深度残差卷积网的超分辨率重建，提出一种适用于不同噪声强度的图像降噪算法。利用改进的非局部均值算法和Boxfilter滤波器对图像进行初步降噪，再通过深度残差卷积网络对初步降噪的图像实现端到端的低分辨率图像超分辨重建。仿真实验结果表明:当噪声强度分别为15、25、40、50、60时，相比于其他经典降噪算法，新算法能获得更高的峰值信噪比和结构相似性，且随着噪声强度的升高，优势越来越明显；新算法适用于已知噪声水平的降噪，也适用于盲噪声降噪，且盲降噪性能优于其他经典降噪算法；此外，该算法还可以更好地恢复图像细节，产生较好的视觉效果。

关键词: 图像降噪, 非局部均值, Boxfilter滤波器, 深度残差卷积网络, 超分辨率

Abstract: Most of the existing image denoising algorithms can only deal with the noise intensity varying in a limited range. For the actual image noise intensity varying in a wide range, an image denoising algorithm is proposed based on the fast non-local means algorithm and the deep residual convolutional network-based super-resolution reconstruction algorithm. The improved non-local means algorithm and Box filter are used to denoise the image initially, and then the initial denoised image is reconstructed with end-to-end super-resolution of low-resolution images by the deep residual convolutional network. The simulated results show that the proposed algorithm can be used to obtain higher peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) compared with other classical denoising algorithms when the noise intensities are 15, 25, 40, 50 and 60. And with the increase in noise intensity, its advantage is more and more obvious. The proposed algorithm is suitable for denoising of known noise and blind noise. And the proposed algorithm is superior to the classical noise reduction algorithms in blind noise reduction. In addition, the proposed algorithm can restore the image details better and generate a better visual effect.

Key words: imagedeniosing, non-localmean, Boxfilter, deepresidualconvolutionalnetwork, super-resolution

中图分类号:

TP751.1

李静, 刘哲, 黄文准. 基于快速非局部均值和超分辨率重建的图像降噪算法[J]. 兵工学报, 2021, 42(8): 1716-1727.

LI Jing, LIU Zhe, HUANG Wenzhun. An Image Deniosing Algorithm Based on Fast Non-local Mean and Super-resolution Reconstruction[J]. Acta Armamentarii, 2021, 42(8): 1716-1727.

参考文献

［1］ BUADES A, COLL B, MOREL J M. A non-local algorithm for image denoising［C］∥Proceedings of Computer Society Conference on Computer Vision and Pattern Recognition. San Diego, CA, US: IEEE, 2005: 60-65.
［2］ DABOV K, FOI A, KATKOVNIK V, et al. Image denoising by sparse 3-D transform-domain collaborative filtering［J］. IEEE Transactions on Image Processing, 2007, 16(8): 2080-2095.
［3］ JENKINS G, TUPPER P. A dynamic neural gradient model of two-item and intermediate transposition［J］. Neural Computation, 2018, 30(7): 1961-1982.
［4］ ZHANG J L, WU J M, HU B B. Dictionary learning via a mixed noise model for sparse representation classification of rolling bearings［J］. IEEE Access, 2020, 8: 213416-213425.
［5］ SU W K, NI J Q, HU X L. Image steganography with symmetric embedding using Gaussian Markov random field model［J］. IEEE Transactions on Circuits and Systems for Video Technology, 2021, 31(3): 1001-1015.
［6］ DABOV K, FOI A, KATKOVNIK V, et al. Color image denoising via sparse 3D collaborative filtering with grouping constraint in luminance-chrominance space ［C］∥Proceedings of IEEE International Conference on Image Processing.San Antonio, TX, US: IEEE, 2007: 301-313.
［7］ XU J, ZHANG L, ZUO W M, et al. Patch group based nonlocal self-similarity prior learning for image denoising［C］∥Proceedings of IEEE International Conference on Computer Vision. Santiago, Chile: IEEE, 2015: 244-252.
［8］周飞燕, 金林鹏, 董军. 卷积神经网络研究综述［J］. 计算机学报, 2017, 40(6): 1229-1251.
ZHOU F Y, JIN L P, DONG J. Review of convolutional neural networks［J］. Chinese Journal of Computers, 2017, 40(6): 1229-1251.(in Chinese)
［9］唐贤伦, 杜一铭, 刘雨微,等. 基于条件深度卷积生成对抗网络的图像识别方法［J］. 自动化学报, 2018, 44(5): 855-864.
TANG X L, DU Y M, LIU Y W, et al. Image recognition with conditional deep convolutional generative adversarial networks ［J］. Acta Automatica Sinica, 2018,44(5): 855-864.(in Chinese)
［10］李文英, 曹斌, 曹春水, 等. 一种基于深度学习的青铜器铭文识别方法［J］. 自动化学报, 2018, 44(11):2023-2030.
LI W Y, CAO B, CAO C S, et al. A deep learning based method for bronze inscreption recognition［J］. Acta Automatica Sinica, 2018, 44(11):2023-2030.(in Chinese)
［11］ MAO X J, SHEN C, YANG Y B. Image restoration using convolutional auto-encoders with symmetric skip connetions［J］. IEEE Transactions on Image Processing, 2016, 20(8): 1024-1037.
［12］ ZHANG K, ZUO W M, GU S H, et al. Learning deep CNN denoiser prior for image restoration［C］∥Proceedings of the 4th Asian Conference on Pattern Recognition. Nanjing, China: Nanjing University, 2017:2808-2817.
［13］ ZHANG K, ZUO W M, CHEN Y J, et al. Beyond a Gaussian denoiser:residual learning of deep CNN for image denoising［J］. IEEE Transactions on Image Processing, 2017, 26(7): 3142-3155.
［14］盖杉, 鲍中运. 基于深度学习的高噪声图像去噪算法［J］. 自动化学报, 2020, 46(12):2672-2680.
GAI S, BAO Z Y. High noise image denoising algorithm based on deep learning［J］. Acta Automatica Sinica, 2020, 46(12):2672-2680.(in Chinese)
［15］ DONG C, LOY C C, HE K M, et al. Image super resolution using deep convolutional networks［J］. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2016,38(2): 295-307.
［16］ DONG C, LOY C C, TANG X O. Accelerating the super resolution convolutional neural network［J］. Lecture Notes in Computer Science, 2016, 9906: 391-407.
［17］ LUO Y M, ZHOU L G, WANG S, et al. Video satellite imagery super resolution via convolutional neural networks［J］. IEEE Geoscience & Remote Sensing Letters, 2017, 14(12):2398-2402.
［18］ DUCOURNAU A, FABLET R. Deep Learning for ocean remote sensing: an application of convolutional neural networks for super-resolution on satellite derived SST data［C］∥Proceedings of IEEE 9th IAPR Workshop on Pattern Recognition in Remote Sensing. Cancún, Mexico: IEEE, 2016.
［19］ SHI W Z, CABALLERO J, HUSZAR F,et al. Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network［C］∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, US: IEEE, 2016: 1874-1883.
［20］ KIM J, LEE J K, LEE K M. Accurate image super-resolution using very deep convolutional networks［C］∥ Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, US: IEEE, 2016:1646-1654.
［21］ LAI W S, HUANG J B, AHUJA N, et al. Deep Laplacian pyramid networks for fast and accurate super-resolution［C］∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, HI, US: IEEE, 2017:5835-5843.
［22］ TIMOFTE R, AGUTSSON E, GOOL L V, et al. NTIRE 2017 challenge on single imagesuper-resolution: methods and results［C］∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition Workshops. Honolulu, HI, US:IEEE, 2017:1122-1131.
［23］ ROTH S, BLACK M J. Fields of experts［J］. International Journal of Computer Vision, 2009,82(2): 205-229.
［24］ GU S H, ZHANG L, ZUO W M, et al. Weighted nuclear norm minimization with application to image denoising［C］∥Procee- dings of IEEE Conference on Computer Vision and Pattern Recognition. Columbus, OH, US: IEEE Computer Society, 2014:2862-2869.
［25］ ZHANG K, ZUO W M, CHEN Y J, et al. Beyond a Gaussian denoiser:residual learning of deep CNN for image denoising［J］. IEEE Transactions on Image Processing, 2017, 26(7): 3142-3155.
［26］ ABUBAKAR A, ZHAO X J, MAEN T, et al. A hybrid denoising algorithm of BM3D and KSVD for Gaussian noise in DoFP polarization images［J］. IEEE Access, 2020, 8: 57451-57459.
［27］ CHEN Y J, POCK T. Trainable nonlinear reaction diffusion: a flexible framework for fast and effective image restoration［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2016, 39(6):1256-1272.

[1]	申清明，王国博，赵建中，许管利. 基于混合噪声模型的射线图像降噪方法[J]. 兵工学报, 2014, 35(12): 2087-2091.
[2]	杨民，孟凡勇，梁丽红，魏东波. 动态射线数字图像序列降噪算法及快速实现[J]. 兵工学报, 2011, 32(11): 1331-1336.