13 基于SN-GAN的大面积缺损图像修复算法研究

贺佳馨;吕晓琪;张继凯;李菁;

• 1:内蒙古科技大学信息工程学院

• 2:内蒙古工业大学信息工程学院

• 
  

摘要(Abstract)：

针对传统图像修复算法存在的图像细节修复效果差、视觉连贯性不佳以及训练不稳定等问题，将生成对抗网络和孪生神经网络进行结合，孪生神经网络当成GAN中的判别器，并在生成网络中使用均方误差，孪生网络中使用对比损失.实验结果显示，SN-GAN模型在细节修复及视觉连贯性上得到一定提升，并且也更适用于大面积缺损图像的修复.

关键词(KeyWords)： 图像修复;生成对抗网络;孪生神经网络;缺损图像

基金项目(Foundation): 国家自然科学基金资助项目（61771266）;; 内蒙古自治区自然科学基金资助项目（2019BS06005）;; 内蒙古自治区高等学校科学研究基金资助项目（NJZY20095）;; 内蒙古自治区科技计划基金资助项目（2019GG138）

作者(Authors): 贺佳馨;吕晓琪;张继凯;李菁;

DOI: 10.16559/j.cnki.2095-2295.2022.02.014

参考文献(References)：

• [1] Bertalmio M,Sapiro G,Caselles V,et al. Image inpainting[C]//Proceedings of the 27th annual conference on computer graphics and interactive techniques. New York,US:ACM Press/Addison-Wesley Publishing Co.,2000:417.

• [2] Chan T F,Shen J. Nontexture inpainting by curvaturedriven diffusions[J]. Journal of Visual Communication&Image Representation,2001,12(4):436.

• [3]何凯，张涛，梁然．一种基于全变分模型的图像修复改进算法[J]．光电子·激光，2010(12):1890.

• [4] Criminisi A,Perez P,Toyama K. Region filling and object removal by exemplar-basedimage inpainting[J].IEEE Transactions on Image Processing,2004,13(9):1200.

• [5] Sun J,Yuan L,Jia J,et al. Image completion with structure propagation[J]. ACM Transactions on Graphics. 2005,24(3):861.

• [6] Shen B,Hu W,Zhang Y,et al. Image inpainting via sparse representation[C]//IEEE international conference on acoustics,speech and signal processing(ICASSP). Washington,USA:IEEE Computer Society,2009:697.

• [7] Sahoo S K,Lu W. Image inpainting using sparse approximation with adaptive windowselection[C]//IEEE 7th international symposium on intelligent signal processing(WISP). Piscataway,USA:IEEE,2011:1.

• [8] Tang F,Ying Y,Wang J. A novel texture synthesisbased algorithm for object removal in photographs[C]//Annual asian computing science conference. Heidelberg,Germany:Springer,2004:248.

• [9]张如．大区域图像修补与图像特技制作研究[D]．北京：北京交通大学，2009.

• [10]梁淑芬，郭敏，梁湘群．改进的Criminisi算法的数字图像修复技术[J]．计算机工程与设计，2016,37(5):1314.

• [11] Goodfellow I J,Pouget-Abadie J,Mirza M,et al. Generativeadversarial networks[C]//Proceedings of the international conference on neural information processing systems. Cambridge,USA:MIT Press,2014:2672.

• [12] Pathak D,Krahenbuhl P,Donahue J,et al. Context encoders:featurelarningbyinpainting[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. Piscataway,USA:IEEE,2016:2536.

• [13] Radford A,Metz L,Chintala S. Unsupervised representation learning with deep convolutional generative adversarial networks[J]. ARXiv Preprint ARXiv,2015,1511. 06434.

• [14] Yeh R A,Chen C,Lim T Y,et al. Semantic image inpainting with deep generative models[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. Hawaii,USA:2017:5485.

• [15] Li Y,Liu S,Yang J,et al. Generative face completion[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. Piscataway,USA:IEEE,2017:3911.

• [16] Iizuka S,Simo-Serra E,Ishikawa H. Globally and locally consistent image completion[J]. ACM Transactions on Graphics,2017,36(4):1.

• [17] Chen Y,Zhang H,Liu L,et al. Research on image inpainting algorithm of improved GAN based on two-discriminationsnetworks[J]. Applied Intelligence,2020,51(1):1.

• [18] Hedjazi M A,Genc Y. Efficient texture-aware multiGAN for image inpainting[J]. Knowledge-Based Systems,2021,217:106789.

• [19] Yu J H,Lin Z H,Yang J M,et al. Generative image inpainting with contextual attention[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. Piscataway,USA:IEEE,2018:5505.

• [20] Bromley J,Guyon I,LeCun Y,et al. Signature verification using a“Siamese”time delay neural network[C]//Proceedings of the 6th international conference on neural information processing systems. San Francisco,USA:Morgan Kaufmann Publishers Inc,1993:737.

• [21] Chopra S,Hadsell R,LeCun Y. Learning a similarity metric discriminatively,with application to face verification[C]//Proceedings of 2005 IEEE computer society conference on computer vision and pattern recognition.Piscataway,USA:IEEE,2005:539.

• [22]杨雪，郑婷婷，戴阳．基于孪生卷积神经网络的图像融合[J]．计算机系统应用，2020,29(5):196.

• [23]徐先峰，张丽，郎彬，等．引入感知模型的改进孪生卷积神经网络实现人脸识别算法研究[J]．电子学报，2020,48(4):643.