基于生成式对抗网络的发票图像超分辨率研究

doi:10.16182/j.issn1004731x.joss.20-0095

系统仿真学报 ›› 2021, Vol. 33 ›› Issue (6): 1307-1314.doi: 10.16182/j.issn1004731x.joss.20-0095

基于生成式对抗网络的发票图像超分辨率研究

李新利, 邹昌铭, 杨国田, 刘禾

华北电力大学控制与计算机工程学院,北京 102206

收稿日期:2020-02-28 修回日期:2020-05-02 出版日期:2021-06-18 发布日期:2021-06-23
作者简介:李新利(1973-),女,博士,副教授,研究方向为模式识别、图像处理、燃烧过程检测技术。E-mail：lixinli@ncepu.edu.cn

Research of Super-resolution Processing of Invoice Image Based on Generative Adversarial Network

Li Xinli, Zou Changming, Yang Guotian, Liu He

School of Control and Computer Engineering, North China Electric Power University, Beijing, 102206, China

Received:2020-02-28 Revised:2020-05-02 Online:2021-06-18 Published:2021-06-23

摘要/Abstract

摘要： 发票自动识别可有效提高财务工作效率。为避免低分辨率的发票图像影响自动识别的准确性,提出了一种用于对发票图像进行超分辨率处理的ESRGAN (Encoder Super-resolution Generative Adversarial Network)网络。ESRGAN网络是基于带条件的生成式对抗网络,设计了辅助编码器,引导网络生成更加真实的超分辨率图像。基于实际发票图像,将ESRGAN网络与常规图像处理、SRCNN (Super-resolution Convolutional Neural Networks)网络和SRGAN (Super-resolution Generative Adversarial Network)网络进行对比实验,并通过峰值信噪比(Peak Signal to Noise Ratio, PSNR)和结构相似性(Structural Similarity, SSIM)评价指标进行模型评价。实验结果表明基于ESRGAN超分辨率处理的图像在视觉效果和评价指标上均具有良好的效果。

关键词: 发票图像, 超分辨率, 生成式对抗网络, ESRGAN, 评价指标

Abstract: Automatic identification of invoices can effectively improve financial efficiency. But low-resolution invoice image reduces the accuracy of automatic identification, an ESRGAN (Encoder Super-resolution Generative Adversarial Network) network for super-resolution processing of invoice images is proposed. The ESRGAN network is based on a conditional generative adversarial network. An auxiliary encoder is designed to guide the network to generate a more realistic super-resolution image. Based on the actual invoice image, the ESRGAN network and the conventional image processing, SRCNN (Super-resolution Convolutional Neural Networks) network and SRGAN (Super-resolution Generative Adversarial Network) network. The model is evaluated through two evaluation indicators of peak signal-to-noise ratio (PSNR) and structural similarity (SSIM). The experimental results show that the images processed based on ESRGAN super-resolution are better on visual effects and evaluation indicators.

Key words: invoice image, super-resolution, generative adversarial networks, ESRGAN, evaluation indicator

中图分类号:

TP391.4

李新利, 邹昌铭, 杨国田, 刘禾. 基于生成式对抗网络的发票图像超分辨率研究[J]. 系统仿真学报, 2021, 33(6): 1307-1314.

Li Xinli, Zou Changming, Yang Guotian, Liu He. Research of Super-resolution Processing of Invoice Image Based on Generative Adversarial Network[J]. Journal of System Simulation, 2021, 33(6): 1307-1314.

参考文献

[1] Rafael C G, Richard E W.数字图像处理[M]. 3版. 北京: 电子工业出版社, 2014: 97-101.
Rafael C G, Richard E W.Digital Image Processing [M]. 3rd ed. Beijing: Publishing House of Electronic Industry, 2014: 97-101.
[2] Chao D, Chen C L, He K M, et al.Learning a Deep Convolutional Network for Image Super-Resolution[C]// European Conference on Computer Vision (ECCV). Switzerland: Springer International Publishing, 2014: 184-199.
[3] 范明明, 池源, 张铭津, 等. 基于深度学习的芯片图像超分辨率重建[J]. 模式识别与人工智能, 2019, 32(4): 353-360.
Fan Mingming, Chi Yuan, Zhang Mingjin, et al.Super-resolution Reconstruction of Chip Images Based on Deep Learning[J]. Pattern Recognition and Artificial Intelligence, 2019, 32(4): 353-360.
[4] Goodfellow I J, Poutget-Abadie J, Mirza M, et al.Generative Adversarial Nets[C]// The 27th International Conference on Neural Information Processing Systems. Montreal. CANADA: NIPS, 2014:2672-2680.
[5] Rezende D J, Mohamed S, Wierstra D.Stochastic Back-propagation and Approximate Inference in Deep Generative Models[C]// International Conference on Machine Learning (ICML). Beijing: ICML, 2014: 1-9.
[6] Hinton G E, Osindero S, Teh Y W.A Fast Learning Algorithm for Deep Belief Nets[J]. Neural Computation (S0899-7667), 2006, 18(7): 1527-1554.
[7] 王坤峰, 苟超, 段艳杰, 等. 生成式对抗网络 GAN 的研究进展与展望[J]. 自动化学报, 2017, 43(3): 321-332.
Wang Kunfeng, Gou Chao, Duan Yanjie, et al.Research Progress and Prospect of Generative Adversarial Networks[J]. Acta Automatica Sinica, 2017, 43(3): 321-332.
[8] 林懿伦, 戴星原, 李力, 等. 人工智能研究的新前线:生成式对抗网络[J]. 自动化学报, 2018, 44(5): 775-792.
Lin Yilun, Dai Xingyuan, Li Li, et al.A New Frontier for Artificial Intelligence Research: Generative Adversarial Networks[J]. Acta Automatica Sinica, 2018, 44(5): 775-792.
[9] Ledig C, Theis L, Huszar F, et al.Photo-realistic Single Image Super-resolution Using a Generative Adversarial Network[C]// IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, HI, USA: IEEE, 2017: 105-114.
[10] Isola P, Zhu J Y, Zhou T, et al.Image-to-Image Translation with Conditional Adversarial Networks[C]// IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, HI, USA: IEEE, 2017: 5967-5976.
[11] Zhu J Y, Park T, Isola P, et al.Unpaired Image-to-Image Translation Using Cycle-consistent Adversarial Networks[C]// IEEE International Conference on Computer Vision (ICCV). Venice, Italy: IEEE, 2017: 2242-2251.
[12] Choi Y, Choi M, Kim M, et al.StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation[C]// IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR). Salt Lake City, UT, USA: IEEE, 2018: 8789-8797.
[13] He Z L, Zuo W M, Kan M N, et al.AttGAN: Facial Attribute Editing by Only Changing What You Want[J]. IEEE Transactions on Image Processing (S1057-7149), 2019, 28(11): 5464-5478.
[14] Qina R, Tan R T, Yang W H, et al.Attentive Generative Adversarial Network for Raindrop Removal From a Single Image[C]// IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Salt Lake City, UT, USA: IEEE, 2018: 2482-2491.
[15] Mirza M, Osindero S.Conditional Generative Adversarial nets[C]// The 18th Annual Conference of the International Speech Communication Association (INTERSPEECH). Stockholm, Sweden: ISCA, 2017: 2814-2818.
[16] Wang H, Wang Y T, Zhou Z, et al.CosFace: Large Margin Cosine Loss for Deep Face Recognition[C]// IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR). Salt Lake City, UT, USA: IEEE, 2018: 5265-5274.
[17] Wang Y J, Li J H, Lu Yi, et al.Image Quality Evaluation Based on Image Weighted Separating Block Peak Signal to Noise Ratio[C]// 2003 International Conference on Neural Networks and Signal Processing (ICNNSP'03). Nanjing: IEEE, 2003: 994-997.
[18] Wang Z, Bovik A C, Sheikh H R, et al.Image Quality Assessment: from Error Visibility to Structural Similarity[J]. IEEE Transactions on Image Processing (S1057-7149), 2004, 13(4): 600-612.

基于生成式对抗网络的发票图像超分辨率研究

Research of Super-resolution Processing of Invoice Image Based on Generative Adversarial Network

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 9

编辑推荐

Metrics

本文评价

[1]	曹建芳, 贾一鸣, 闫敏敏, 田晓东. 稳定增强生成对抗网络在壁画的超分辨率重建[J]. 系统仿真学报, 2022, 34(5): 1076-1089.
[2]	胡蕾, 王足根, 陈田, 张永梅. 一种改进的SRGAN红外图像超分辨率重建算法[J]. 系统仿真学报, 2021, 33(9): 2109-2118.
[3]	杜金莲, 李淑飞, 金雪云. 三维烟雾流场超分辨率数据生成网络模型的研究[J]. 系统仿真学报, 2021, 33(10): 2381-2389.
[4]	曹东风, 王艳. 基于组合权重的制造系统能效贝叶斯评价方法[J]. 系统仿真学报, 2018, 30(11): 4313-4322.
[5]	彭竹清, 王艳, 纪志成. 基于改进熵值法的离散制造系统能效评价方法[J]. 系统仿真学报, 2017, 29(9): 2121-2127.
[6]	谢士琴, 赵天忠, 王威, 史京京. 高分二号卫星影像融合方法比较研究[J]. 系统仿真学报, 2017, 29(11): 2742-2747.
[7]	侯堃, 李志龙, 冯玉超, 陈谊. 一种面向层次数据可视化的圆形空间填充算法[J]. 系统仿真学报, 2016, 28(9): 2035-2041.
[8]	武兆斌, 陈黎, 赵春霞. 基于灰色聚类和层次分析的模拟训练成绩评定[J]. 系统仿真学报, 2016, 28(2): 416-424.
[9]	陈彦, 王艳. 基于改进主成分分析法的离散制造能耗分析[J]. 系统仿真学报, 2016, 28(12): 3087-3094.