基于生成对抗网络的多幅离焦图像超分辨率重建算法

doi:10.19678/j.issn.1000-3428.0058848

计算机工程 ›› 2021, Vol. 47 ›› Issue (9): 266-273. doi: 10.19678/j.issn.1000-3428.0058848

基于生成对抗网络的多幅离焦图像超分辨率重建算法

斯捷¹, 肖雄², 李泾³, 马明勋², 毛玉星²

1. 浙江图盛输变电工程有限公司, 浙江温州 325000;
2. 重庆大学电气工程学院, 重庆 400044;
3. 国网浙江电力有限公司温州供电公司, 浙江温州 325000

收稿日期:2020-07-06 修回日期:2020-09-03 发布日期:2020-09-11
作者简介:斯捷(1976-),男,工程师,主研方向为计算机视觉;肖雄,硕士研究生;李泾,工程师;马明勋,硕士研究生;毛玉星,副教授、博士。
基金资助:
国家重点研发计划“物联网智能感知终端平台系统与应用验证”（2018YFB2100100）。

Super-Resolution Reconstruction Algorithm with Multi-Frame Defocused Images Based on Generative Adversarial Network

SI Jie¹, XIAO Xiong², LI Jing³, MA Mingxun², MAO Yuxing²

1. Zhejiang Tusheng Transmission and Transformation Engineering Co., Ltd., Wenzhou, Zhejiang 325000, China;
2. School of Electrical Engineering, Chongqing University, Chongqing 400044, China;
3. Wenzhou Power Supply Company, State Grid Zhejiang Electric Power Co., Ltd., Wenzhou, Zhejiang 325000, China

Received:2020-07-06 Revised:2020-09-03 Published:2020-09-11

摘要/Abstract

摘要： 为提高超分辨率算法重建出的图像质量，提出融合多幅离焦图像的超分辨率重建算法。以离焦图像作为切入点，利用自编码器提取离焦图像中的重要特征，根据空间特征变换层结构，将离焦特征与原始特征相结合，完成图像的超分辨率重建。在Celeb A人脸数据集上进行实验，结果表明，与传统插值算法及SRGAN算法相比，所提算法在大多数情况下能获得更高峰值信噪比及结构相似性数值，能生成质量更高的重建图像。

关键词: 自编码器, 图像特征提取, 生成对抗网络, 超分辨率重建, 深度神经网络

Abstract: In order to improve the quality of the reconstructed images, a super-resolution reconstruction algorithm using multi-frame defocused images is proposed.The algorithm employs an auto-encoder to extract the important features in the defocused images, and the layer structure is transformed based on spatial features to combine the defocused features with the original features, so the super-resolution reconstruction of the image is completed.The experimental results on the Celeb A face data set show that in most cases, the proposed algorithm provides a higher peak signal-to-noise ratio and structural similarity than the traditional interpolation algorithm and the SRGAN algorithm.This super-resolution algorithm based on multi-frame defocused images can generate better reconstructed images.

Key words: auto encoder, image feature extraction, Generative Adversarial Net(GAN), Super Resolution(SR), deep neural network

中图分类号:

TP391

斯捷, 肖雄, 李泾, 马明勋, 毛玉星. 基于生成对抗网络的多幅离焦图像超分辨率重建算法[J]. 计算机工程, 2021, 47(9): 266-273.

SI Jie, XIAO Xiong, LI Jing, MA Mingxun, MAO Yuxing. Super-Resolution Reconstruction Algorithm with Multi-Frame Defocused Images Based on Generative Adversarial Network[J]. Computer Engineering, 2021, 47(9): 266-273.

http://www.ecice06.com/CN/Y2021/V47/I9/266

图/表 12

20210917192657

20210917192700

20210917192705

20210917192715

20210917192719

20210917192722

20210917192726

20210917192729

20210917192733

20210917192736

20210917192740

20210917192743

参考文献

[1] DONG C, LOY C C, HE K, et al.Learning a deep convolutional network for image super-resolution[C]//Proceedings of 2014 European Conference on Computer Vision.Berlin, Germany:Springer, 2014:184-199.
[2] DONG C, LOY C C, HE K, et al.Image super-resolution using deep convolutional networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(2):295-307.
[3] KIM J, LEE J K, LEE K M.Accurate image super-resolution using very deep convolutional networks[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2016:109-112.
[4] KIM J, LEE J K, LEE K M.Deeply-recursive convolutional network for image super resolution[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2016:45-76.
[5] ZHANG Y, TIAN Y, KONG Y, et al.Residual dense network for image super-restoration[C]//Proceedings of 2018 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2018:37-45.
[6] LIM B, SON S, KIM H, et al.Enhanced deep residual networks for single image super-resolution[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops.Washington D.C., USA:IEEE Press, 2017:71-98.
[7] SIMONYAN K, ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[EB/OL].[2020-06-01].https://arxiv.org/abs/1409.1556.
[8] JOHNSON J, ALAHI A, FEI-FEI L.Perceptual losses for real-time style transfer and super-resolution[EB/OL].[2020-06-01] https://arxiv.org/abs/1603.08155.
[9] LEDIG C, THEIS L, HUSZAR F, et al.Photo-realistic single image super-resolution using a generative adversarial network[J].IEEE Computer Society, 2016, 11(1):1-14.
[10] BULAT A, TZIMIROPOULOS G.Super-FAN:integrated facial landmark localization and super-resolution of real-world low resolution faces in arbitrary poses with GANs[C]//Proceedings of 2018 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2018:42-53.
[11] PARK S, SON H, CHO S, et al.SRFeat:single image super-resolution with feature discrimination[C]//Proceedings of 2018 European Conference on Computer Vision.Berlin, Germany:Springer, 2018:32-53.
[12] ZHANG K, ZUO W, ZHANG L.Learning a single convolutional super-resolution network for multiple degradations[C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington D.C., USA:IEEE Press, 2018:124-133.
[13] BULAT A, YANG J, TZIMIROPOULOS G.To learn image super-resolution, use a GAN to learn how to do image degradation first[C]//Proceedings of 2018 European Conference on Computer Vision.Berlin, Germany:Springer, 2018:39-47.
[14] WANG X, YU K, WU S, et al.Esrgan:enhanced super-resolution generative adversarial networks[C]//Proceedings of 2018 European Conference on Computer Vision workshops.Berlin, Germany:Springer, 2018:78-97.
[15] WANG X, YU K, DONG C, et al.Recovering realistic texture in image super-resolution by deep spatial feature transform[C]//Proceedings of 2018 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2018:2160-2176.
[16] LARSEN A B L, SONDERBY SOREN KAAE, WINTHER O, et al.Autoencoding beyond pixels using a learned similarity metric[C]//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2015:1191-1208.
[17] SUN J, CAO W, XU Z, et al.Learning a convolutional neural network for non-uniform motion blur removal[C]//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2015:22-43.
[18] LI XU, JIMMY SJ REN, CE LIU, et al.Deep convolutional neural network for image deconvolution[C]//Proceedings of 2014 Conference on Neural Information Processing Systems.[S.l.]:MIT Press, 2014:1121-1132.
[19] LEIBE B, MATAS J, SEBE N, et al.A neural approach to blind motion deblurring[EB/OL].[2020-06-01].https://www.researchgate.net/publication/301839917_A_Neural_Approach_to_Blind_Motion_Deblurring.
[20] BORACCHI G, FOI A.Modeling the performance of image restoration from motion blur[J].IEEE Transactions on Image Processing, 2012, 21(8):3502-3517.

选择文件类型/文献管理软件名称

选择包含的内容

基于生成对抗网络的多幅离焦图像超分辨率重建算法

Super-Resolution Reconstruction Algorithm with Multi-Frame Defocused Images Based on Generative Adversarial Network

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	靳雁霞, 史志儒, 杨晶, 刘亚变, 乔星宇, 张翎. 布料与精细建模物体间的碰撞检测算法研究[J]. 计算机工程, 2023, 49(7): 269-277.
[2]	沈梦强, 于文年, 易黎, 宋南. 基于GAN的全时间尺度语音增强方法[J]. 计算机工程, 2023, 49(6): 115-122,130.
[3]	丁子轩, 俞雷, 张娟, 李想, 王新宇. 基于深度残差自适应注意力网络的图像超分辨率重建[J]. 计算机工程, 2023, 49(5): 231-238.
[4]	张子宣, 宗学军, 何戡, 连莲. 基于CVAE-CatBoost的工业控制网络异常流量检测研究[J]. 计算机工程, 2023, 49(5): 173-180.
[5]	陈锐, 孙羽菲, 郭强, 隋轶丞, 周振辉, 石昌青, 张玉志. OclDNN:一种可应用于TensorFlow的通用DNN库[J]. 计算机工程, 2023, 49(4): 138-148.
[6]	李培育, 张雅丽. 基于改进SRGAN模型的人脸图像超分辨率重建[J]. 计算机工程, 2023, 49(4): 199-205.
[7]	罗嗣卿, 陈慧. 基于生成对抗网络的图像场景转换[J]. 计算机工程, 2023, 49(4): 217-225.
[8]	翟社平, 张宇航, 柏晓夏. 融合实体邻域信息的知识图谱嵌入负采样方法[J]. 计算机工程, 2023, 49(3): 95-104.
[9]	席荣康, 蔡满春, 芦天亮. 基于数据增强与流数据处理的Tor流量分析模型[J]. 计算机工程, 2023, 49(3): 177-184.
[10]	石磊, 张吉涛, 高宇飞, 卫琳, 陶永才. 基于Transformer与BiLSTM的网络流量入侵检测[J]. 计算机工程, 2023, 49(3): 29-36,57.
[11]	袁立宁, 胡皓, 刘钊. 基于多通道图卷积自编码器的图表示学习[J]. 计算机工程, 2023, 49(2): 150-160,174.
[12]	刘强, 张颖, 周卫祥, 蒋先涛, 周薇娜, 周谋国. 自适应类增量学习的物联网入侵检测系统[J]. 计算机工程, 2023, 49(2): 169-174.
[13]	王春东, 孙嘉琪, 杨文军. 基于矫正理解的中文文本对抗样本生成方法[J]. 计算机工程, 2023, 49(2): 37-45.
[14]	刘金硕, 詹岱依, 邓娟, 王丽娜. 基于深度神经网络和联邦学习的网络入侵检测[J]. 计算机工程, 2023, 49(1): 15-21,30.
[15]	董卫宇, 李海涛, 王瑞敏, 任化娟, 孙雪凯. 基于堆叠卷积注意力的网络流量异常检测模型[J]. 计算机工程, 2022, 48(9): 12-19.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于生成对抗网络的多幅离焦图像超分辨率重建算法

Super-Resolution Reconstruction Algorithm with Multi-Frame Defocused Images Based on Generative Adversarial Network

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献

相关文章 15

编辑推荐

Metrics

本文评价