| 322 | 4 | 62 |
| Downloads | Citas | Reads |
The most image denoising models for current convolution neural network have been trained via fixed noise level image sets. This image denoising model can lead to the problem of poor denoising for the lack of flexibility in noise level. To solve this problem, the image denoising method based on double convolution neural network is proposed. The proposed method takes the noise level image as the network input, down-sample the input image, uses the expansion convolution to increase the feeling wild, and optimizes the network by using batch renormalization(BN)and leakyrectified linear unit(Leaky ReLu). It can not only flexibly handle different levels of noise but also increase image denoising speed through GPU parallel processing. Experimental results demonstrate the proposed method is better than several state-of-the-art methods in terms of visual comparison. The average peak signal to noise ratio(PSNR) and structural similarity image measurement(SSIM) values for the proposed method are 0.4-1.8 dB and0.000 3-0.047 7 higher than other state-of-the-art methods. The computing time of the proposed method is 0.02-0.06 seconds faster than other state-of-the-art methods.
[1] ANDREWS H C, HUNT B R. Digital image restoration[M].Englewood Cliffs:Prentice-Hall, 1977.
[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] GU S H, ZHANG L, ZUO W M, et al. Weighted nuclear norm minimization with application to image denoising[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition, June 23-28, 2014,Columbus, OH, USA. New York:IEEE Xplore, 2014:2862-2869.
[4] BURGER H C, SCHULER C J, HARMELING S. Image denoising:can plain neural networks compete with BM3D?[C]//Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition, June 16-21, 2012,Providence, RI, USA. New York:IEEE Xplore, 2012:2392-2399.
[5] JAIN V, SEUNG H S. Natural image denoising with convolutional networks[C]//Proceedings of the 22nd Annual Conference on Neural Information Processing Systems,NIPS 2008, December 8-11, 2008, Vancouver, BC,Canada. Vancouver:NIPS, 2008:769-776.
[6] GAI S, BAO Z Y. New image denoising algorithm via improved deep convolutional neural network with perceptive loss[J]. Expert Systems with Applications, 2019, 138:112815.
[7] CHEN L C, PAPANDREOU G, KOKKINOS I, et al.DeepLab:semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(4):834-848.
[8] CHEN Y J, YU W, POCK T. On learning optimized reaction diffusion processes for effective image restoration[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition, June 7-12, 2015, Boston,MA, USA. New York:IEEE Xplore, 2015:5261-5269.
[9] ZHANG K, ZUO W M, ZHANG L. FFDNet:toward a fast and flexible solution for CNN-based image denoising[J].IEEE Transactions on Image Processing, 2018, 27(9):4608-4622.
[10] JIA X X, LIU S Y, FENG X C, et al. FOCNet:a fractional optimal control network for image denoising[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR), June 15-20, 2019,Long Beach, CA, USA. New York:IEEE Xplore, 2019:6047-6056.
[11] TAI Y, YANG J, LIU X M, et al. MemNet:a persistent memory network for iamge restoration[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision(ICCV), October 22-29, 2017, Venice, Italy. New York:IEEE Xplore, 2017:4549-4557.
[12]田娟，李英祥，李彤岩.激活函数在卷积神经网络中的对比研究[J].计算机系统应用，2018, 27(7):43-49.TIAN J, LI Y X, LI T Y. Contrastive study of activation function in convolutional neural network[J]. Computer Systems and Applications, 2018, 27(7):43-49.(in Chinese)
[13] HE K M, ZHANG X Y, REN S Q, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(9):1904-1916.
[14] IOFFE S. Batch renormalization:towards reducing minibatch dependence in batch-normalized models[EB/OL].(2017-03-30)[2021-01-10]. https：//arxiv. org/pdf/1702.03275. pdf.
[15] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6):84-90.
[16] LI X X, XIAO J, ZHOU Y Y, et al. Detail retaining convolutional neural network for image denoising[J]. Journal of Visual Communication and Image Representation, 2020,71:102774.
[17] YANG B, CAO J M, NI R R, et al. Facial expression recognition using weighted mixture deep neural network based on double-channel facial images[J]. IEEE Access,2018, 6:4630-4640.
[18] NEWELL A, YANG K Y, DENG J. Stacked hourglass networks for human pose estimation[C]//Proceedings of the14th European Conference on Computer Vision(ECCV2016), October 11-14, 2016, Amsterdam, The Netherlands. Cham:Springer, 2016:483-499.
[19] DENG J, DONG W, SOCHER R, et al. ImageNet:a large-scale hierarchical image database[C]//Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition, June 20-25, 2009, Miami, FL, USA.New York:IEEE Xplore, 2009:248-255.
[20] WANG T Y, SUN M X, HU K N. Dilated deep residual network for image denoising[C]//Proceedings of the 2017IEEE 29th International Conference on Tools with Artificial Intelligence(ICTAI), November 6-8, 2017, Boston, MA,USA. New York:IEEE Xplore, 2017:1272-1279.
[21] ROTH S, BLACK M J. Fields of experts:a framework for learning image priors[C]//Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR′05), June 20-25, 2005, San Diego, CA, USA. New York:IEEE Xplore, 2005:860-867.
Basic Information:
DOI：10.12194/j.ntu.20210118002
China Classification Code:TP391.41;TP183
Citation Information:
[1]SHEN Weiheng,WANG Luwei,ZHU Yonggui.A double layer convolution neural network method for image denoising[J].Journal of Nantong University (Natural Science Edition),2022,21(01):52-57.DOI:10.12194/j.ntu.20210118002.
Fund Information:
中国传媒大学中央高校基本科研业务费专项资金资助项目（CUC2019A002,CUC2019B021）
2021-04-08
2021
2021-09-01
2021
1
2022-03-20
2022-03-20