带噪声水平评估的快速灵活盲深度降噪模型

doi:10.19678/j.issn.1000-3428.0058327

摘要/Abstract

摘要： 为提高快速灵活降噪网络（FFDNet）模型的降噪性能，建立一种噪声水平估计（NLE）模型，将其预测的噪声水平估计值输入FFDNet模型中，并以NLE模型作为FFDNet深度降噪模型的前置模块，使FFDNet转换为盲降噪模型。采用浅层卷积神经网络模型将噪声信号从噪声图像中分离出以得到噪声映射图，将噪声映射图的标准差作为噪声水平的初估值。鉴于噪声水平初估值与真值之间具有强相关性的特性，应用BP神经网络模型对噪声水平初估值进行修正。实验结果表明，该NLE模型与FFDNet降噪模型相结合后，降噪效果总体上与使用真实噪声水平值的FFDNet降噪模型接近，在多数噪声水平值下，两者的PSNR值相差小于0.1 dB，NLE模型的估计值可以达到与真实噪声水平值近似的效果，能够充分发挥FFDNet降噪模型的快速和灵活特性。

关键词: 快速灵活降噪, 盲降噪, 卷积神经网络, 初估值, 修正值

Abstract: To improve the denoising performance of the Fast and Flexible Denoising Convolutional Neural Network(FFDNet),this paper proposes a Noise Level Estimation(NLE) model that estimates the level of noise.The estimation result is input into the FFDNet model,and the NLE model is taken as the preceding module of the FFDNet deep denoising model to transform it into a blind denoising model.Then the shallow convolutional neural network model is used to separate noise signals from noisy images to obtain the noise map,the standard deviation of which is taken as the initial estimated value of the noise level.Considering the fact that there exists strong correlation between the initial estimated value and ground-truths of the noise level,a Back-Propagation(BP) neural network model is used to correct the initial estimated value of noise level.Experimental results show that when the proposed NLE model works with the FFDNet model,its denoising performance is close to that of the FFDNet denoising model which uses the ground-truths of noise level.For most of the noise level values,the difference of Peak Signal to Noise Ratio(PSNR) values between the two models is within 0.1 dB,which means the estimation results of the proposed NLE model are similar to the ground-truths of noise level,bringing the fast and flexible characteristics of the FFDNet model into full play.

Key words: fast and flexible denoising, blind denoising, Convolutional Neural Network(CNN), initial estimated value, corrected value

中图分类号:

TP391

于海雯, 易昕炜, 徐少平, 林珍玉. 带噪声水平评估的快速灵活盲深度降噪模型[J]. 计算机工程, 2020, 46(12): 222-230,237.

YU Haiwen, YI Xinwei, XU Shaoping, LIN Zhenyu. Fast and Flexible Blind Deep Denoising Model with Noise Level Estimation[J]. Computer Engineering, 2020, 46(12): 222-230,237.

http://www.ecice06.com/CN/Y2020/V46/I12/222

图/表 13

20201216142230

20201216142233

20201216142236

20201216142242

20201216142248

20201216142252

20201216142320

20201216142325

20201216142328

20201216142332

20201216142336

20201216142339

20201216142342

参考文献

[1] LI Mengmeng,XIA Yang,LI Xinru,et al.Dynamic hybrid recommendation algorithm based on stacked denoising autoencoder[J].Computer Engineering,2019,45(8):184-189.(in Chinese) 李梦梦,夏阳,李心茹,等.基于栈式降噪自动编码器的动态混合推荐算法[J].计算机工程,2019,45(8):184-189.
[2] ZHANG Guoling,WANG Xiaodan,LI Rui,et al.Extreme learning machine based on stacked denoising sparse auto-encoder[J].Computer Engineering,2020,46(9):61-67.(in Chinese) 张国令,王晓丹,李睿,等.基于栈式降噪稀疏自编码器的极限学习机[J].计算机工程,2020,46(9):61-67.
[3] LV Xiaoqi,WU Liang,GU Yu,et al.Low dose CT lung denoising model based on deep convolution neural network[J].Journal of Electronics and Information Technology,2018,40(6):87-93.(in Chinese) 吕晓琪,吴凉,谷宇,等.基于深度卷积神经网络的低剂量CT肺部去噪[J].电子与信息学报,2018,40(6):87-93.
[4] ZHANG Kai,ZUO Wangmeng,GU Shuhang,et al.Learning deep CNN denoiser prior for image restoration[C]//Preceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2017:2808-2817.
[5] ZHANG Kai,ZUO Wangmeng,CHEN Yunjin,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.
[6] LEFKIMMIATIS S.Universal denoising networks:a novel CNN architecture for image denoising[C]//Preceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2018:145-168.
[7] ZHANG Kai,ZUO Wangmeng,ZHANG Lei.FFDNet:toward a fast and flexible solution for CNN based image denoising[J].IEEE Transactions on Image Processing,2018,27(9):4608-4622.
[8] PYATYKH S,HESSER J,ZHENG L.Image noise level estimation by principal component analysis[J].IEEE Transactions on Image Processing,2013,22(2):687-699.
[9] LIU X H,TANAKA M,OKUTOMI M.Single-image noise level estimation for blind denoising[J].IEEE Transactions on Image Processing,2013,22(12):5226-5237.
[10] XU Shaoping,ZENG Xiaoxia,TANG Yiling.Fast noise level estimation algorithm based on two-stage support vector regression[J].Journal of Computer-Aided Design and Computer Graphics,2018,30(3):447-458.(in Chinese) 徐少平,曾小霞,唐祎玲.基于两阶段支持向量回归的快速噪声水平估计算法[J].计算机辅助设计与图形学学报,2018,30(3):447-458.
[11] DANIEL Z,YAIR W.Scale invariance and noise in natural images[C]//Proceedings of 2009 IEEE International Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2009:2209-2216.
[12] DONG Li,ZHOU Jiantao,TANG Yuanyan.Noise level estimation for natural images based on scale-invariant kurtosis and piecewise stationarity[J].IEEE Transactions on Image Processing,2017,26(2):1017-1030.
[13] GUPTA P,BAMPIS C G,JIN Y,et al.Natural scene statistics for noise estimation[C]//Proceedings of 2018 IEEE Southwest Symposium on Image Analysis and Interpretation.Washington D.C.,USA:IEEE Press,2018:85-88.
[14] FANG Zhuang,YI Xuming.A novel natural image noise level estimation based on flat patches and local statistics[EB/OL].[2020-03-10].https://doi.org/10.1007/s11042-018-7137-4.
[15] RAWAT W,WANG Z H.Deep convolutional neural networks for image classification:a comprehensive review[J].Neural Computation,2017,29(9):2352-2449.
[16] IOFFE S,SZEGEDY C.Batch normalization:accelerating deep network training by reducing internal covariate shift[EB/OL].[2020-03-10].https://arxiv.org/abs/1502.03167.
[17] PABLO A,MICHAEL M,CHARLESS F,et al.Contour detection and hierarchical image segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(5):898-916.
[18] ZHANG Lin,ZHANG Lei,MOU Xuanqin,et al.A comprehensive evaluation of full reference image quality assessment algorithms[C]//Proceedings of the 19th IEEE International Conference on Image Processing.Washington D.C.,USA:IEEE Press,2012:1477-1480.
[19] YANG S M,TAI S C.Fast and reliable image-noise estimation using a hybrid approach[J].Journal of Electronic Imaging,2010,19(3):033007.
[20] CHEN G,ZHU F,HENG P A.An efficient statistical method for image noise level estimation[C]//Proceedings of IEEE International Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2015:477-485.

选择文件类型/文献管理软件名称

选择包含的内容