《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (2): 536-544.DOI: 10.11772/j.issn.1001-9081.2022010015
所属专题: 多媒体计算与计算机仿真
收稿日期:
2022-01-07
修回日期:
2022-04-30
接受日期:
2022-05-05
发布日期:
2022-05-24
出版日期:
2023-02-10
通讯作者:
刘文印
作者简介:
陈刚(1977—),男,江西高安人,博士研究生,CCF会员,主要研究方向:人工智能、计算机视觉基金资助:
Gang CHEN1, Yongwei LIAO1, Zhenguo YANG1, Wenying LIU1,2()
Received:
2022-01-07
Revised:
2022-04-30
Accepted:
2022-05-05
Online:
2022-05-24
Published:
2023-02-10
Contact:
Wenying LIU
About author:
CHEN Gang, born in 1977, Ph. D. candidate. His research interests include artificial intelligence, computer vision.Supported by:
摘要:
针对多尺度生成式对抗网络图像修复算法(MGANII)在修复图像过程中训练不稳定、修复图像的结构一致性差以及细节和纹理不足等问题,提出了一种基于多特征融合的多尺度生成对抗网络的图像修复算法。首先,针对结构一致性差以及细节和纹理不足的问题,在传统的生成器中引入多特征融合模块(MFFM),并且引入了一个基于感知的特征重构损失函数来提高扩张卷积网络的特征提取能力,从而改善修复图像的细节性和纹理特征;然后,在局部判别器中引入了一个基于感知的特征匹配损失函数来提升判别器的鉴别能力,从而增强了修复图像的结构一致性;最后,在对抗损失函数中引入风险惩罚项来满足利普希茨连续条件,使得网络在训练过程中能快速稳定地收敛。在CelebA数据集上,所提的多特征融合的图像修复算法与MANGII相比能快速收敛,同时所提算法所修复图像的峰值信噪比(PSNR)、结构相似性(SSIM)比基线算法所修复图像分别提高了0.45%~8.67%和0.88%~8.06%,而Frechet Inception距离得分(FID)比基线算法所修复图像降低了36.01%~46.97%。实验结果表明,所提算法的修复性能优于基线算法。
中图分类号:
陈刚, 廖永为, 杨振国, 刘文印. 基于多特征融合的多尺度生成对抗网络图像修复算法[J]. 计算机应用, 2023, 43(2): 536-544.
Gang CHEN, Yongwei LIAO, Zhenguo YANG, Wenying LIU. Image inpainting algorithm of multi-scale generative adversarial network based on multi-feature fusion[J]. Journal of Computer Applications, 2023, 43(2): 536-544.
迭代轮次/104 | Lr | 迭代轮次/104 | Lr |
---|---|---|---|
0 | 0.167 | 90 | 0.009 |
30 | 0.082 | 120 | 0.006 |
60 | 0.082 |
表1 重构损失Lr的部分数值
Tab. 1 Partial values of reconstruction loss Lr
迭代轮次/104 | Lr | 迭代轮次/104 | Lr |
---|---|---|---|
0 | 0.167 | 90 | 0.009 |
30 | 0.082 | 120 | 0.006 |
60 | 0.082 |
算法 | PSNR/dB | SSIM | FID |
---|---|---|---|
CE[ | 24.980 | 0.8622 | 6.568 |
GMCNN[ | 26.123 | 0.9017 | 6.865 |
PENNet[ | 26.011 | 0.8923 | 6.857 |
PICNet[ | 26.425 | 0.9106 | 6.815 |
MGANII[ | 27.025 | 0.9236 | 7.926 |
本文算法 | 27.146 | 0.9317 | 4.203 |
表2 不同算法的修复效果比较
Tab. 2 Comparison of inpainting effects of different algorithms
算法 | PSNR/dB | SSIM | FID |
---|---|---|---|
CE[ | 24.980 | 0.8622 | 6.568 |
GMCNN[ | 26.123 | 0.9017 | 6.865 |
PENNet[ | 26.011 | 0.8923 | 6.857 |
PICNet[ | 26.425 | 0.9106 | 6.815 |
MGANII[ | 27.025 | 0.9236 | 7.926 |
本文算法 | 27.146 | 0.9317 | 4.203 |
1 | LIU G L, REDA F A, SHIH K J, et al. Image inpainting for irregular holes using partial convolutions[C]// Proceedings of the 2018 European Conference on Computer Vision, LNCS 11215. Cham: Springer, 2018: 89-105. |
2 | CAI W W, WEI Z G. PiiGAN: generative adversarial networks for pluralistic image inpainting[J]. IEEE Access, 2020, 8: 48451-48463. 10.1109/ACCESS.2020.2979348 |
3 | XIONG W, YU J H, LIN Z, et al. Foreground-aware image inpainting[C]// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2019: 5833-5841. 10.1109/cvpr.2019.00599 |
4 | WIN K N, LI K L, CHEN J G, et al. Fingerprint classification and identification algorithms for criminal investigation: a survey[J]. Future Generation Computer Systems, 2020, 110: 758-771. 10.1016/j.future.2019.10.019 |
5 | HOU M L YANG S, HU Y G, et al. Novel method for virtual restoration of cultural relics with complex geometric structure based on multiscale spatial geometry[J]. ISPRS International Journal of Geo-Information, 2018, 7(9): No.353. 10.3390/ijgi7090353 |
6 | WU Q D, LI Y B, LIN Y. Medical image restoration method via multiple nonlocal prior constraints[J]. Journal of Intelligent and Fuzzy Systems, 2020, 38(1): 5-19. 10.3233/jifs-179375 |
7 | ZHANG M L, DESROSIERS C. High-quality image restoration using low-rank patch regularization and global structure sparsity[J]. IEEE Transactions on Image Processing, 2019, 28(2): 868-879. 10.1109/tip.2018.2874284 |
8 | 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: ACM, 2000:417-424. 10.1145/344779.344972 |
9 | CRIMINISI A, PÉREZ P, TOYAMA K. Region filling and object removal by exemplar-based image inpainting[J]. IEEE Transactions on Image Processing, 2004, 13(9): 1200-1212. 10.1109/tip.2004.833105 |
10 | BOBIN J, STRARCK J L, FADILI J M, et al. Morphological component analysis: an adaptive thresholding strategy[J]. IEEE Transactions on Image Processing, 2007, 16(11): 2675-2681. 10.1109/tip.2007.907073 |
11 | PATHAK D, KRÄHENBÜHL P, DONAHUE J, et al. Context Encoders: feature learning by inpainting[C]// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2016: 2536-2544. 10.1109/cvpr.2016.278 |
12 | IIZUKA S, SIMO-SERRA E, ISHIKAWA H. Globally and locally consistent image completion[J]. ACM Transactions on Graphics, 2017, 36(4): No.107. 10.1145/3072959.3073659 |
13 | WANG C Y, XU C, WANG C H, et al. Perceptual adversarial networks for image-to-image transformation[J]. IEEE Transactions on Image Processing, 2018, 27(8): 4066-4079. 10.1109/tip.2018.2836316 |
14 | SAGONG M C, SHIN Y G, KIM S W, et al. PEPSI: fast image inpainting with parallel decoding network[C]// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2019: 11352-11360. 10.1109/cvpr.2019.01162 |
15 | ZHENG C X, CHAM T J, CAI J F. Pluralistic image completion[C]// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2019: 1438-1447. 10.1109/cvpr.2019.00153 |
16 | 李克文,张文韬,邵明文,等. 多尺度生成式对抗网络图像修复算法[J]. 计算机科学与探索, 2020, 14(1):159-170. |
LI K W, ZHANG W T, SHAO M W, et al. Multi-scale generative adversarial networks image inpainting algorithm[J]. Journal of Frontiers of Computer Science and Technology, 2020, 14(1):159-170. | |
17 | 赵露露,沈玲,洪日昌. 图像修复研究进展综述[J]. 计算机科学, 2021, 48(3):14-26. 10.11896/jsjkx.210100048 |
ZHAO L L, SHEN L, HONG R C. Survey on image inpainting research progress[J]. Computer Science, 2021, 48(3):14-26. 10.11896/jsjkx.210100048 | |
18 | XIANG C Y, CAO Y J, DUAN P S, et al. An improved exemplar-based image inpainting algorithm[C]// Proceedings of the 9th International Conference on Computer Science and Education. Piscataway: IEEE, 2014:770-775. 10.1109/iccse.2014.6926566 |
19 | AHARON M, ELAD M, BRUCKSTEIN A. K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation[J]. IEEE Transactions on Signal Processing, 2006, 54(11): 4311-4322. 10.1109/tsp.2006.881199 |
20 | 冯浪,张玲,张晓龙. 基于扩张卷积的图像修复[J]. 计算机应用, 2020, 40(3):825-831. 10.11772/j.issn.1001-9081.2019081471 |
FENG L, ZHANG L, ZHANG X L. Image inapinting based on dilated convolution[J]. Journal of Computer Applications, 2020, 40(3):825-831. 10.11772/j.issn.1001-9081.2019081471 | |
21 | 刘波宁,翟东海. 基于双鉴别网络的生成对抗网络图像修复方法[J].计算机应用, 2018, 38(12):3557-3562, 3595. 10.11772/j.issn.1001-9081.2018051097 |
LIU B N, ZHAI D H. Image completion method of generative adversarial networks based on two discrimination networks[J]. Journal of Computer Applications, 2018, 38(12):3557-3562, 3595. 10.11772/j.issn.1001-9081.2018051097 | |
22 | LIU X M, ZHAI D M, ZHOU J T, et al. Sparsity based image error concealment via adaptive dual dictionary learning and regularization[J]. IEEE Transactions on Image Processing, 2017, 26(2): 782-796. 10.1109/tip.2016.2623481 |
23 | XIE J Y, XU L L, CHEN E H. Image denoising and inpainting with deep neural networks[C]// Proceedings of the 25th International Conference on Neural Information Processing Systems - Volume 1. Red Hook, NY: Curran Associates Inc., 2012: 341-349. |
24 | WANG Y, TAO X, QI X J, et al. Image inpainting via generative multi-column convolutional neural networks[C]// Proceedings of the 32nd International Conference and Workshop on Neural Information Processing Systems. Red Hook, NY: Curran Associates Inc., 2018: 329-338. 10.1109/icsp.2018.8652324 |
25 | ZENG Y H, FU J L, CHAO H Y, et al. Learning pyramid-context encoder network for high-quality image inpainting[C]// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2019:1486-1494. 10.1109/cvpr.2019.00158 |
26 | ZHENG H, LI J, GAO X B, et al. Progressive perception-oriented network for single image super-resolution[J]. Information Sciences, 2021, 546: 769-786. 10.1016/j.ins.2020.08.114 |
27 | SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL]. (2015-04-10) [2022-03-26].. |
28 | JOHNSON J, ALAHI A, LI F F. Perceptual losses for real-time style transfer and super-resolution[C]// Proceedings of the 2016 European Conference on Computer Vision, LNCS 9906. Cham: Springer, 2016: 694-711. |
29 | GATYS L, ECKEKR A S, BETHGE M. Texture synthesis using convolutional neural networks[M/OL]// CORTES C, LAWRENCE N, LEE D, et al. Advances in Neural Information Processing Systems 28 (NIPS 2015). [2022-03-26].. 10.1109/cvpr.2016.265 |
30 | GULRAJANI I, AHMED F, ARJOVSKY M, et al. Improved training of Wasserstein GANs[C]// Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook, NY: Curran Associates Inc., 2017: 5769-5779. |
31 | SHEN W, LIU R J. Learning residual images for face attribute manipulation[C]// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2017: 1225-1233. 10.1109/cvpr.2017.135 |
[1] | 潘烨新, 杨哲. 基于多级特征双向融合的小目标检测优化模型[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2871-2877. |
[2] | 戎妍, 刘嘉雯, 李馨蕾. 面向学生课堂情感计算的自适应混合网络[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2919-2930. |
[3] | 陈彤, 杨丰玉, 熊宇, 严荭, 邱福星. 基于多尺度频率通道注意力融合的声纹库构建方法[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2407-2413. |
[4] | 邓凯丽, 魏伟波, 潘振宽. 改进掩码自编码器的工业缺陷检测方法[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2595-2603. |
[5] | 李晨倩, 刘俊. 基于半监督和多尺度级联注意力的超声颈动脉斑块分割方法[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2604-2610. |
[6] | 刘丽, 侯海金, 王安红, 张涛. 基于多尺度注意力的生成式信息隐藏算法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2102-2109. |
[7] | 刘瑞华, 郝子赫, 邹洋杨. 基于多层级精细特征融合的步态识别算法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2250-2257. |
[8] | 唐媛, 陈艳平, 扈应, 黄瑞章, 秦永彬. 基于多尺度混合注意力卷积神经网络的关系抽取模型[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2011-2017. |
[9] | 施赛龙, 方智文. 基于多尺度聚合和共享注意力的注视估计模型[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2047-2054. |
[10] | 熊武, 曹从军, 宋雪芳, 邵云龙, 王旭升. 基于多尺度混合域注意力机制的笔迹鉴别方法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2225-2232. |
[11] | 李伟, 张晓蓉, 陈鹏, 李清, 张长青. 基于正态逆伽马分布的多尺度融合人群计数算法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2243-2249. |
[12] | 刘越, 刘芳, 武奥运, 柴秋月, 王天笑. 基于自注意力机制与图卷积的3D目标检测网络[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1972-1977. |
[13] | 王美, 苏雪松, 刘佳, 殷若南, 黄珊. 时频域多尺度交叉注意力融合的时间序列分类方法[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1842-1847. |
[14] | 程小辉, 黄云天, 张瑞芳. 基于多尺度和加权坐标注意力的轻量化红外道路场景检测模型[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1927-1934. |
[15] | 黄梦源, 常侃, 凌铭阳, 韦新杰, 覃团发. 基于层间引导的低光照图像渐进增强算法[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1911-1919. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||