《计算机应用》唯一官方网站 ›› 2024, Vol. 44 ›› Issue (7): 2102-2109.DOI: 10.11772/j.issn.1001-9081.2023070919
收稿日期:
2023-07-11
修回日期:
2023-09-14
接受日期:
2023-09-19
发布日期:
2023-10-26
出版日期:
2024-07-10
通讯作者:
刘丽
作者简介:
侯海金(1997—),男,山西晋中人,硕士研究生,主要研究方向:信息隐藏;基金资助:
Li LIU(), Haijin HOU, Anhong WANG, Tao ZHANG
Received:
2023-07-11
Revised:
2023-09-14
Accepted:
2023-09-19
Online:
2023-10-26
Published:
2024-07-10
Contact:
Li LIU
About author:
HOU Haijin, born in 1997. M. S. candidate. His research interests include information hiding.Supported by:
摘要:
针对现有生成式信息隐藏算法嵌入容量低且提取的秘密图像视觉质量欠佳的问题,提出基于多尺度注意力的生成式信息隐藏算法。首先,设计基于多尺度注意力的双编码-单解码生成器,载体图像与秘密图像的特征在编码端分两个支路独立提取,在解码端通过多尺度注意力模块进行融合,并利用跳跃连接为解码端提供不同尺度的细节特征,从而获得高质量的载密图像。其次,在U-Net结构的提取器中引入自注意力模块,以弱化载体图像特征、增强秘密图像深层特征,并利用跳跃连接弥补秘密图像细节特征,提高秘密信息提取的准确率;同时,多尺度判决器与生成器的对抗训练可以有效提升载密图像的视觉质量。实验结果表明,所提算法在嵌入容量为24 bpp的情况下,生成的载密图像峰值信噪比(PSNR)和结构相似性(SSIM)平均可达到40.93 dB和0.988 3,且提取的秘密图像PSNR和SSIM平均可达到30.47 dB和0.954 3。
中图分类号:
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, 2024, 44(7): 2102-2109.
Li LIU, Haijin HOU, Anhong WANG, Tao ZHANG. Generative data hiding algorithm based on multi-scale attention[J]. Journal of Computer Applications, 2024, 44(7): 2102-2109.
层号 | 输出大小 | 卷积核大小 | 操作 | 层号 | 输出大小 | 卷积核大小 | 操作 |
---|---|---|---|---|---|---|---|
1 | 256×256×3 | — | — | 8 | 16×16×512 | 4×4 | Self Atten_concat_Deconv_BN_LeakyReLU |
2 | 256×256×64 | 5×5 | Conv_BN_LeakyReLU | 9 | 32×32×512 | 4×4 | Self Atten_concat_Deconv_BN_LeakyReLU |
3 | 128×128×128 | 4×4 | Conv_BN_LeakyReLU | 10 | 64×64×256 | 4×4 | Self Atten_concat_Deconv_BN_LeakyReLU |
4 | 64×64×256 | 4×4 | Conv_BN_LeakyReLU | 11 | 128×128×128 | 4×4 | concat_Deconv_BN_LeakyReLU |
5 | 32×32×512 | 4×4 | Conv_BN_LeakyReLU | 12 | 256×256×64 | 4×4 | concat_Deconv_BN_LeakyReLU |
6 | 16×16×512 | 4×4 | Conv_BN_LeakyReLU | 13 | 256×256×3 | 5×5 | Deconv_Sigmoid |
7 | 8×8×512 | 4×4 | Conv_BN_LeakyReLU |
表1 提取器参数列表
Tab. 1 Parameter list of extractor
层号 | 输出大小 | 卷积核大小 | 操作 | 层号 | 输出大小 | 卷积核大小 | 操作 |
---|---|---|---|---|---|---|---|
1 | 256×256×3 | — | — | 8 | 16×16×512 | 4×4 | Self Atten_concat_Deconv_BN_LeakyReLU |
2 | 256×256×64 | 5×5 | Conv_BN_LeakyReLU | 9 | 32×32×512 | 4×4 | Self Atten_concat_Deconv_BN_LeakyReLU |
3 | 128×128×128 | 4×4 | Conv_BN_LeakyReLU | 10 | 64×64×256 | 4×4 | Self Atten_concat_Deconv_BN_LeakyReLU |
4 | 64×64×256 | 4×4 | Conv_BN_LeakyReLU | 11 | 128×128×128 | 4×4 | concat_Deconv_BN_LeakyReLU |
5 | 32×32×512 | 4×4 | Conv_BN_LeakyReLU | 12 | 256×256×64 | 4×4 | concat_Deconv_BN_LeakyReLU |
6 | 16×16×512 | 4×4 | Conv_BN_LeakyReLU | 13 | 256×256×3 | 5×5 | Deconv_Sigmoid |
7 | 8×8×512 | 4×4 | Conv_BN_LeakyReLU |
算法 | 载体图像/载密图像 | |
---|---|---|
PSNR/dB | SSIM | |
Yu算法[ | 33.50 | 0.964 5 |
Ying算法[ | 33.94 | 0.951 7 |
HiDDeN算法 | 37.24 | 0.979 1 |
本文算法 | 40.93 | 0.988 3 |
表2 在COCO数据集下不同算法图像隐藏的效果对比
Tab. 2 Comparison of image hiding effects using different algorithms in COCO dataset
算法 | 载体图像/载密图像 | |
---|---|---|
PSNR/dB | SSIM | |
Yu算法[ | 33.50 | 0.964 5 |
Ying算法[ | 33.94 | 0.951 7 |
HiDDeN算法 | 37.24 | 0.979 1 |
本文算法 | 40.93 | 0.988 3 |
网络结构 | 载体图像/载密图像 | |||
---|---|---|---|---|
PSNR/dB | SSIM | MAE | RMSE | |
Base | 35.75 | 0.985 4 | 0.013 | 0.016 |
Base+MSA | 37.93 | 0.989 0 | 0.010 | 0.013 |
Base+MSA+A | 40.93 | 0.998 3 | 0.007 | 0.009 |
表3 消融实验对载密图像的影响
Tab. 3 Effects of ablation experiments on stego-images
网络结构 | 载体图像/载密图像 | |||
---|---|---|---|---|
PSNR/dB | SSIM | MAE | RMSE | |
Base | 35.75 | 0.985 4 | 0.013 | 0.016 |
Base+MSA | 37.93 | 0.989 0 | 0.010 | 0.013 |
Base+MSA+A | 40.93 | 0.998 3 | 0.007 | 0.009 |
网络结构 | 秘密图像/提取的秘密图像 | |||
---|---|---|---|---|
PSNR/dB | SSIM | MAE | RMSE | |
Base | 18.80 | 0.803 7 | 0.102 | 0.126 |
Base+MSA | 26.60 | 0.944 1 | 0.045 | 0.056 |
Base+MSA+A | 30.47 | 0.954 3 | 0.035 | 0.044 |
表4 消融实验对提取的秘密图像的影响
Tab. 4 Effects of ablation experiments on extracted secret images
网络结构 | 秘密图像/提取的秘密图像 | |||
---|---|---|---|---|
PSNR/dB | SSIM | MAE | RMSE | |
Base | 18.80 | 0.803 7 | 0.102 | 0.126 |
Base+MSA | 26.60 | 0.944 1 | 0.045 | 0.056 |
Base+MSA+A | 30.47 | 0.954 3 | 0.035 | 0.044 |
1 | 武晓帅,徐明,乔通,等.图像空域可逆信息隐藏研究进展[J].中国图象图形学报, 2022, 27(1): 125-149. |
WU X S, XU M, QIAO T, et al. Review of reversible data hiding based on the spatial domain of images [J]. Journal of Image and Graphics, 2022, 27(1): 125-149. | |
2 | 付章杰,李恩露,程旭,等.基于深度学习的图像隐写研究进展[J].计算机研究与发展, 2021, 58(3): 548-568. |
FU Z J, LI E L, CHENG X, et al. Recent advances in image steganography based on deep learning [J]. Journal of Computer Research and Development, 2021, 58(3): 548-568. | |
3 | GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial nets [C]// Proceedings of the 27th International Conference on Neural Information Processing Systems. Cambridge: MIT Press, 2014: 2672-2680. |
4 | XIAO M, LI X, MA B, et al. Efficient reversible data hiding for jpeg images with multiple histograms modification [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2021, 31(7): 2535-2546. |
5 | HE W, CAI Z, WANG Y. High-fidelity reversible image watermarking based on effective prediction error-pairs modification [J]. IEEE Transactions on Multimedia, 2021, 23: 52-63. |
6 | YU C, ZHANG X, WANG D, et al. Reversible data hiding with pairwise PEE and 2D-PEH decomposition [J]. Signal Processing, 2022, 196: 108527. |
7 | KOUHI A, SEDAAGHI M H. Prediction error distribution with dynamic asymmetry for reversible data hiding [J]. Expert Systems with Applications, 2021, 184: 115475. |
8 | TANG W, TAN S, LI B, et al. Automatic steganographic distortion learning using a generative adversarial network [J]. IEEE Signal Processing Letters, 2017, 24(10): 1547-1551. |
9 | VOLKHONSKIY D, BORISENKO B, BURNAEV E. Generative adversarial networks for image steganography [EB/OL]. [2022-06-30]. . |
10 | 王耀杰,钮可,杨晓元.基于生成对抗网络的信息隐藏方案[J].计算机应用, 2018, 38(10): 2923-2928. |
WANG Y J, NIU K, YANG X Y. Information hiding scheme based on generative adversarial network [J]. Journal of Computer Applications, 2018, 38(10): 2923-2928. | |
11 | ZHU J, KAPLAN R, JOHNSON J, et al. HiDDeN: hiding data with deep networks [C]// Proceedings of the 15th European Conference on Computer Vision. Cham: Springer, 2018: 682-697. |
12 | ZHANG H, WANG H, CAO Y, et al. Robust data hiding using inverse gradient attention [EB/OL]. [2023-07-01]. . |
13 | TAN J, LIAO X, LIU J, et al. Channel attention image steganography with generative adversarial networks [J]. IEEE Transactions on Network Science and Engineering, 2022, 9(2): 888-903. |
14 | WEI P, LI S, ZHANG X, et al. Generative steganography network [C]// Proceedings of the 30th ACM International Conference on Multimedia. New York: ACM, 2022: 1621-1629. |
15 | LIU X, MA Z, MA J, et al. Image disentanglement autoencoder for steganography without embedding [C]// Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Piscataway: IEEE, 2022: 2293-2302. |
16 | ZHOU Z, DONG X, MENG R, et al. Generative steganography via auto-generation of semantic object contours [J]. IEEE Transactions on Information Forensics and Security, 2023, 18: 2751-2765. |
17 | BALUJA S. Hiding image in plain sight: deep steganography [C]// Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook: Curran Associates Inc., 2017: 2066-2076. |
18 | ZHANG R, DONG S, LIU J. Invisible steganography via generative adversarial networks [J]. Multimedia Tools and Applications, 2019, 78(7): 8559-8575. |
19 | FU Z, WANG F, CHENG X. The secure steganography for hiding images via GAN [J]. EURASIP Journal on Image and Video Processing, 2020, 2020: No. 46. |
20 | CHEN H, SONG L, QIAN Z, et al. Hiding images in deep probabilistic models [C]// Proceedings of the 36th International Conference on Neural Information Processing Systems. New York: ACM, 2022: 36776-36788. |
21 | MENG L, JIANG X, ZHANG Z, et al. A robust coverless image steganography based on an end-to-end hash generation model [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2023, 33(7): 3542-3558. |
22 | LI Q, WANG X, WANG X, et al. An encrypted coverless information hiding method based on generative models [J]. Information Sciences, 2021, 553: 19-30. |
23 | YU C. Attention based data hiding with generative adversarial networks [C]// Proceedings of the 34th AAAI Conference on Artificial Intelligence. Menlo Park: AAAI Press, 2020: 1120-1128. |
24 | ZHU J-Y, PARK T, ISOLA P, et al. Unpaired image-to-image translation using cycle-consistent adversarial networks [C]// Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway: IEEE, 2017: 2242-2251. |
25 | YING Q, ZHOU H, ZENG X, et al. Hiding images into images with real-world robustness [C]// Proceedings of the 2022 IEEE International Conference on Image Processing. Piscataway: IEEE, 2022: 111-115. |
26 | LI B, TAN S, WANG M, et al. Investigation on cost assignment in spatial image steganography [J]. IEEE Transactions on Information Forensics and Security, 2014, 9(8): 1264-1277. |
27 | BROCK A, DONAHUE J, SIMONYAN K. Large scale GAN training for high fidelity natural image synthesis [EB/OL]. [2023-07-01]. . |
28 | ASHWATH B. Kaggle [DB/OL]. [2022-06-30]. . |
29 | LIN T-Y, MAIRE M, BELONGIE S, et al. Microsoft COCO: common objects in context [C]// Proceedings of the 13th European Conference on Computer Vision. Cham: Springer, 2014: 740-755. |
[1] | 秦璟, 秦志光, 李发礼, 彭悦恒. 基于概率稀疏自注意力神经网络的重性抑郁疾患诊断[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2970-2974. |
[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] | 李力铤, 华蓓, 贺若舟, 徐况. 基于解耦注意力机制的多变量时序预测模型[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2732-2738. |
[4] | 赵志强, 马培红, 黑新宏. 基于双重注意力机制的人群计数方法[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2886-2892. |
[5] | 薛凯鹏, 徐涛, 廖春节. 融合自监督和多层交叉注意力的多模态情感分析网络[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2387-2392. |
[6] | 汪雨晴, 朱广丽, 段文杰, 李书羽, 周若彤. 基于交互注意力机制的心理咨询文本情感分类模型[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2393-2399. |
[7] | 高鹏淇, 黄鹤鸣, 樊永红. 融合坐标与多头注意力机制的交互语音情感识别[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2400-2406. |
[8] | 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. |
[9] | 李钟华, 白云起, 王雪津, 黄雷雷, 林初俊, 廖诗宇. 基于图像增强的低照度人脸检测[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2588-2594. |
[10] | 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. |
[11] | 莫尚斌, 王文君, 董凌, 高盛祥, 余正涛. 基于多路信息聚合协同解码的单通道语音增强[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2611-2617. |
[12] | 徐松, 张文博, 王一帆. 基于时空信息的轻量视频显著性目标检测网络[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2192-2199. |
[13] | 李大海, 王忠华, 王振东. 结合空间域和频域信息的双分支低光照图像增强网络[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2175-2182. |
[14] | 魏文亮, 王阳萍, 岳彪, 王安政, 张哲. 基于光照权重分配和注意力的红外与可见光图像融合深度学习模型[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2183-2191. |
[15] | 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. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||