基于密集反馈网络的单幅图像超分辨率重建

doi:10.19678/j.issn.1000-3428.0059576

摘要/Abstract

摘要： 基于深度学习的单幅图像超分辨率网络模型体积庞大，导致参数利用率低且难以部署，对中间层特征利用不充分。提出一种密集反馈注意力网络（DFAN）模型。在同一特征图中通过多尺度残差注意力模块（MRAB）提取不同尺度的深层特征，以增加特征的多样性。同时将每个MRAB的输出均作为同组中其他残差模块的输入，使各层之间的信息流最大化，从而减小训练难度。实验结果表明，相比VDSR、DRRN、MemNet等模型，DFAN模型具有较优的重建效果，其在重建放大倍数为4的Set5数据集上计算复杂度仅为VDSR模型的0.14倍左右，而峰值信噪比提高了0.57 dB。

关键词: 单幅图像超分辨率重建, 深度学习, 密集反馈模型, 注意力机制, 残差模块

Abstract: The existing deep learning-based models for the super-resolution reconstruction of a single image are large-sized, which reduces the utilization of parameters and middle-layer features,and cause difficulty to model deployment.To address the problems,a model is proposed based on Dense Feedback Attention Network(DFAN).This model uses multi-scale residual modules with attention mechanism to extract deep features of different scales in the same feature map,so the diversity of features is increased.The output of each residual module is broadcast to the input of other residual modules in the same group, so that the information flows between layers are maximized,thereby reducing the difficulty of training.Experimental results show that the proposed network exhibits better reconstruction effects than VDSR,DRRN,MemNet and other models.When reconstructing the Set5 dataset at an amplification factor of 4,the Peak Signal to Noise Ratio(PSNR) of DFAN is 0.57 dB higher than that of VDSR,while the computational complexity of DFAN is only about 0.14 times that of VDSR algorithm.

Key words: super-resolution reconstruction of a single image, Deep Learning(DL), dense feedback model, attention mechanism, residual module

中图分类号:

TP391

刘锡泽, 范红, 海涵, 王鑫城, 许武军, 倪林. 基于密集反馈网络的单幅图像超分辨率重建[J]. 计算机工程, 2021, 47(11): 254-261.

LIU Xize, FAN Hong, HAI Han, WANG Xincheng, XU Wujun, NI Lin. Super-Resolution Reconstruction of a Single Image Based on Dense Feedback Network[J]. Computer Engineering, 2021, 47(11): 254-261.

https://www.ecice06.com/CN/Y2021/V47/I11/254

图/表 16

20211113145331

20211113145334

20211113145337

20211113145340

20211113145343

20211113145347

20211113145350

20211113145353

20211113145355

20211113145358

20211113145401

20211113145404

20211113145407

20211113145411

20211113145414

20211113145417

参考文献

[1] SEEMA R,BAILEY K.Multi-frame image super-resolution by interpolation and iterative backward projection[C]//Proceedings of the 2nd International Conference on Signal Processing and Communication.Washingston D C.,USA:IEEE Press,2019:36-40.
[2] CASTRO S E,NAKANO M,SÁNCHEZ G,et al.Improvement of image super-resolution algorithms using iterative back projection[J].IEEE Latin America Transactions,2017,15(11):2214-2219.
[3] WU T Y,ZHANG Y,HUANG Y L,et al.Fast deconvolution method for angular super-resolution imaging based on sub-space embedding[J].Journal of Engineering,2019,19:6291-6294.
[4] LI W,ZHU W,ZHU Q.Single-image super-resolution based on morphological component analysis and coupled dictionary learning in wavelet domain[C]//Proceedings of Intelligent Transportation,Big Data & Smart City.Washington D.C.,USA:IEEE Press,2020:938-941.
[5] DONG C,LOY C C,HE K,et al.Learning a deep convolutional network for image super-resolution[C]//Proceedings of European Conference Computer Vision.Berlin,Germany:Springer,2014:184-199.
[6] KIM J,KWON L J,MU L K.Accurate image super-resolution using very deep convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2016:1646-1654.
[7] HE K,ZHANG X,REN S,et al.Deep residual learning for image recognition[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2016:770-778.
[8] TAI Y,YANG J,LIU X,et al.Image super-resolution via deep recursive residual network[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2017:2790-2798.
[9] DONG C,LOY C C,TANG X,et al.Accelerating the super-resolution convolutional neural network[C]//Proceedings of European Conference on Computer Vision.Berlin,Germany:Springer,2016:391-407.
[10] SHI W,CABALLERO J,HUSZÁR 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.Washington D.C.,USA:IEEE Press,2016:1874-1883.
[11] LIM B,SON S,KIM H,et al.Enhanced deep residual networks for single image super-resolution[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition Workshops.Washington D.C.,USA:IEEE Press,2017:136-144.
[12] ZHANG Y,LI K,LI L,et al.Image super-resolution using very deep residual channel attention networks[C]//Proceedings of European Conference on Computer Vision.Berlin,Germany:Springer,2018:286-301.
[13] TONG T,LI G,LIU X,et al.Image super-resolution using dense skip connections[C]//Proceedings of IEEE International Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2017:4809-4817.
[14] TAI Y,YANG J,LIU X,et al.MemNet:a persistent memory network for image restoration[C]//Proceedings of IEEE International Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2017:4549-4557.
[15] HUI Z,WANG X,GAO X.Fast and accurate single image super-resolution via information distillation network[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2018:723-731.
[16] 季秀怡,李建华.基于双路注意力机制的化学结构图像识别[J].计算机工程,2020,46(9):213-220. JI X Y,LI J H.Chemical structure image recognition based on dual attention mechanism[J].Computer Engineering,2020,46(9):213-220.(in Chinese)
[17] YANG Y,ZHONG Z,SHEN T,et al.Convolutional neural networks with alternately updated clique[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2018:2413-2422.
[18] AGUSTSSON E,TIMOFTE R.Ntire 2017 challenge on single image super-resolution:dataset and study[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition Workshops.Washington D.C.,USA:IEEE Press,2017:1122-1131.
[19] BEVILACQUA M,ROUMY A,GUILLEMOT C,et al.Low-complexity single-image super-resolution based on nonnegative neighbor embedding[C]//Proceedings of British Machine Vision Conference.Malvern,UK:BMVA Press,2012:1-10.
[20] ZEYDE R,ELAD M,PROTTER M.On single image scale-up using sparse-representations[C]//Proceedings of International Conference on Curves & Surfaces.Berlin,Germany:Springer,2010:711-730.
[21] ARBELÁEZ P,MAIRE M,FOWLKES C,et al.Contour detection and hierarchical image segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(5):898-916.
[22] PABLO A,MAIRE M,FOWLKES C,et al.Contour detection and hierarchical image segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(5):898-916.
[23] HUANG J B,SINGH A,AHUJA N.Single image super-resolution from transformed self-exemplars[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2015:5197-5206.
[24] KIM J W,LEE J K,LEE K M,et al.Deeply-recursive convolutional network for image super-resolution[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2016:1637-1645.
[25] 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.Washington D.C.,USA:IEEE Press,2017:5835-5843.
[26] HAN W,CHANG S,LIU D,et al.Image super-resolution via dual-state recurrent networks[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2018:1654-1663.

选择文件类型/文献管理软件名称

选择包含的内容