融合多头注意力机制的图像降噪网络模型

doi:10.11896/jsjkx.230100091

计算机科学 ›› 2023, Vol. 50 ›› Issue (11A): 230100091-8.doi: 10.11896/jsjkx.230100091

• 图像处理&多媒体技术 • 上一篇下一篇

融合多头注意力机制的图像降噪网络模型

李玥玥¹, 刘万平¹, 黄东²

1 重庆理工大学计算机科学与工程学院重庆 400054
2 贵州大学现代制造技术教育部重点实验室贵阳 550025

发布日期:2023-11-09
通讯作者: 刘万平(wpliu@cqut.edu.cn)
作者简介:(lyy98599@2020.cqut.edu.cn)
基金资助:
重庆市自然科学基金(cstc2021jcyj-msxmX0594);重庆理工大学研究生创新项目资助(gzlcx20223212)

Image Denoising Network Model Combined with Multi-head Attention Mechanism

LI Yueyue¹, LIU Wanping¹, HUANG Dong²

1 College of Computer Science and Engineering,Chongqing University of Technology,Chongqing 400054,China
2 Key Laboratory of Advanced Manufacturing Technology of the Ministry of Education,Guizhou University,Guiyang 550025,China

Published:2023-11-09
About author:LI Yueyue,born in 1998,postgraduate.Her main research interests include deep learning and image denoising.
LIU Wanping,born in 1986,Ph.D,associate professor,master supervisor,is a member China Computer Federation.His main research interests include network and information security.
Supported by:
Natural Science Foundation of Chongqing,China(cstc2021jcyj-msxmX0594) and Graduate Student Innovation Program of Chongqing University of Technology(gzlcx20223212).

摘要/Abstract

摘要： 由于GPU计算的快速发展,深度学习近年来在图像降噪方面得到了应用。大多数深度学习方法都需要无噪声图像作为训练标签,但通常它们很难获得,甚至不可能获得。于是,有学者开始研究使用噪声图像进行降噪网络训练,但其恢复的图像却面临丢失细节信息的问题。受Noise2Noise(N2N)的思想启发,文中使用成对的噪声图像训练神经网络,学习同一范围的同类型噪声之间的分布关系,实现了一种新的降噪网络模型。新开发的模型(MA-UNet)基于经典UNet架构,融合了多头注意力机制(Multi-head Attention)和简易的残差网络,可以更好地挖掘图像的关键信息,掌握特征的全局信息,从而恢复更清晰的图像。与传统算法(CBM3D)和其他方法(如DnCNN和B2U)相比,MA-UNet的性能参数优良。从视觉图像观察来看,所提模型恢复了更清晰的图像细节。与N2N设计的模型相比,在不同噪声幅值下,所提模型在4个经典数据集上的峰值信噪比和结构相似性指数的均值均有显著提高。

关键词: 深度学习, 注意力机制, 细节信息, 图像降噪, 全局特征

Abstract: Due to the rapid development of GPU computing,deep learning has been applied in image denoising recently.Most of the deep learning methods require noise-free images as training labels,but they are usually difficult or even impossible to obtain.Therefore,some scholars begin to study the use of noisy images for noise reduction network training,but the restored image is faced with the problem of losing details.Inspired by the idea of Noise2Noise(N2N),this paper uses pairs of noised images to train the neural network,to learn the distribution relationship between the same type of noise in the same range,and realize a new novel image denoising network model.The newly-developed model(MA-UNet) is based on the classic UNet architecture and combines the multi-head attention mechanism and simple residual network.It can capture the key information of the image,master the glo-bal information of the feature,so as to recover clearer images.Compared with the traditional algorithm CBM3D and other me-thods,such as DnCNN and B2U,MA-UNet has excellent performance in terms of parameters.Through the comparison of visual images,our model restores much clearer image details.Compared with the model designed by N2N,under different noise magnitude,the mean value of the peak signal-to-noise ratio and the structural similarity index of the proposed model on four classical data sets improve significantly.

Key words: Deep learning, Attention mechanism, Detail information, Image denoising, Global feature

中图分类号:

TP391

李玥玥, 刘万平, 黄东. 融合多头注意力机制的图像降噪网络模型[J]. 计算机科学, 2023, 50(11A): 230100091-8. https://doi.org/10.11896/jsjkx.230100091

LI Yueyue, LIU Wanping, HUANG Dong. Image Denoising Network Model Combined with Multi-head Attention Mechanism[J]. Computer Science, 2023, 50(11A): 230100091-8. https://doi.org/10.11896/jsjkx.230100091

参考文献

[1]ZHOU Z H,ZHANG W T.Based on local structural similarity image denoising algorithm[C]//International Conference on Computer and Information Applications.2013:3313-3317.
[2]MOHAN J,KRISHNAVENI V,GUO Y.A survey on the magnetic resonance image denoising methods[J].Biomedical Signal Processing and Control,2014,9:56-69.
[3]YUE L,SHEN H,LI J,et al.Image super-resolution:The techniques,applications,and future[J].Signal Processing,2016,128:389-408.
[4]BUADES A,COLL B,MOREL J M.A non-local algorithm for image denoising[C]//2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR’05).IEEE,2005,2:60-65.
[5]DABOV K,FOI A,KATKOVNIK V,et al.Image denoising by sparse 3-D transform-domain collaborative filtering[J].IEEE Transactionson Image Processing,2007,16(8):2080-2095.
[6]JAIN V,SEUNG H S.Natural image denoising with convolutional networks[J].Advances in Neural Information Processing Systems,2009,2:769-776.
[7]BURGER H C,SCHULER C J,HARMELING S.Image denoi-sing:Can plain neural networks compete with BM3D?[C]//2012 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).IEEE,2012:2392-2399.
[8]ZHANG K,ZUO W,CHEN Y,et al.Beyond a gaussian denoi-ser:Residual learning of deep CNN for image denoising[J].IEEE Transactions on Image Processing,2017,26(7):3142-3155.
[9]VASWANI A,SHAZEER N,PARMAR N,et al.Attention isall you need[C]//Advances in Neural Information Processing Systems.2017:5998-6008.
[10]LI R,ZHENG S,DUAN C,et al.Classification of hyperspectral image based on double-branch dual-attention mechanism network[J].Remote Sensing,2020,12(3):582.
[11]BAI J,CHEN R,LIU M.Feature-attention module for context-aware image-to-image translation[J].The Visual Computer,2020,36(10):2145-2159.
[12]TIAN C,XU Y,LI Z,et al.Attention-guided CNN for image denoising[J].Neural Networks,2020,124:117-129.
[13]LEHTINEN J,MUNKBERG J,HASSELGREN J,et al.Noise2Noise:learning image restoration without clean data[C]//International Conference on Machine Learning.PMLR,2018:2965-2974.
[14]RONNEBERGER O,FISCHER P,BROX T.U-net:Convolu-tional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Compu-ter-Assisted Intervention.Cham:Springer,2015:234-241.
[15]HU J,SHEN L,SUN G.Squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:7132-7141.
[16]WOO S,PARK J,LEE J Y,et al.CBAM:Convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision(ECCV).2018:3-19.
[17]DOSOVITSKIY A,BEYER L,KOLESNIKOV A,et al.Animage is worth 16x16 words:Transformers for image recognition at scale[C]//International Conference on Learning Representations.2021.
[18]FEI J,DAI Y,WANG H,et al.Learning to mask:Towards ge-neralized face forgery detection[J].arXiv:2212.14309,2022.
[19]HE K,ZHANG X,REN S,et al.Deep residual learning forimage recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:770-778.
[20]GLOROT X,BORDES A,BENGIO Y.Deep sparse rectifierneural networks[C]//Proceedings of The Fourteenth International Conference on Artificial Intelligence and Statistics.JMLR Workshop and Conference Proceedings,2011:315-323.
[21]IOFFE S,SZEGEDY C.Batch normalization:Accelerating deep network training by reducing internal covariate shift[C]//International Conference on Machine Learning.PMLR,2015:448-456.
[22]ULYANOV D,VEDALDI A,LEMPITSKY V.Deep imageprior[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:9446-9454.
[23]XU S P,LI C X,LIN G X,et al.Fast image noise level estimation algorithm based on principal component analysis and deep neural network[J].Acta Electronica Sinica,2019,47(2):274-281.
[24]XU L F,YU J M.Research on the influence of different optimizers on LR performance under Gaussian noise [J].Computer Technology and Development,2020,30(3):7-12.
[25]XU S P,LIU T Y,LIN Z Y,et al.Learning a two-stage blind convolutional denoising model for the removal of random-valued impulse noise[J].Chinese Journal ofComputers,2020,43(9):1673-1690.
[26]MARTIN D,FOWLKES C,TAL D,et al.A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics[C]//IEEE International Conference on ComputerVision.IEEE,2001:416-423.
[27]ZEYDE R,ELAD M,PROTTER M.On Single Image Scale-Up Using Sparse-Representations[C]//International Conference on Curves and Surfaces.Berlin:Springer,2010:711-730.
[28]BEVILACQUA M,ROUMY A,GUILLEMOT C,et al.Low-Complexity Single Image Super-Resolution Based on Nonnegative Neighbor Embedding[C]//British Machine Vision Confe-rence.BMVA Press,2012,135:1-10.
[29]HUANG J B,SINGH A,AHUJA N.Single image super-resolution from transformed self-exemplars[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2015:5197-5206.
[30]TIMOFTE R,DE SMET V,VAN GOOL L.A+:Adjusted anchored neighborhood regression for fast super-resolution[C]//Asian Conference on Computer Vision.Cham:Springer,2015:111-126.
[31]LEDIG C,THEIS L,HUSZÁR F,et al.Photo-realistic singleimage super-resolution using a generative adversarial network[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:4681-4690.
[32]WANG Z,LIU J,LI G,et al.Blind2Unblind:Self-Supervised Image Denoising with Visible Blind Spots[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2022:2027-2036.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

融合多头注意力机制的图像降噪网络模型

Image Denoising Network Model Combined with Multi-head Attention Mechanism

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

Metrics

本文评价

推荐阅读 0