《计算机应用》唯一官方网站 ›› 2024, Vol. 44 ›› Issue (3): 772-779.DOI: 10.11772/j.issn.1001-9081.2023040477
所属专题: 网络空间安全
董炜娜1,2, 刘佳1,2(), 潘晓中1,2, 陈立峰1,2, 孙文权1,2
收稿日期:
2023-04-26
修回日期:
2023-07-06
接受日期:
2023-07-10
发布日期:
2023-12-04
出版日期:
2024-03-10
通讯作者:
刘佳
作者简介:
董炜娜(1997—),女,山东烟台人,硕士研究生,主要研究方向:信息隐藏基金资助:
Weina DONG1,2, Jia LIU1,2(), Xiaozhong PAN1,2, Lifeng CHEN1,2, Wenquan SUN1,2
Received:
2023-04-26
Revised:
2023-07-06
Accepted:
2023-07-10
Online:
2023-12-04
Published:
2024-03-10
Contact:
Jia LIU
About author:
DONG Weina, born in 1997, M. S. candidate. Her research interests include information hiding.Supported by:
摘要:
针对基于编码-解码网络的大容量隐写模型存在鲁棒性弱、无法抵抗噪声攻击和信道压缩的问题,提出一种基于编码-解码网络的大容量鲁棒图像隐写方案。首先,设计了基于密集连接卷积网络(DenseNet)的编码器、解码器和判别器,编码器将秘密信息和载体图像联合编码成隐写图像,解码器提取秘密信息,判别器用于区分载体图像和隐写图像。在编码器和解码器中间加入噪声层,采用Dropout、JPEG压缩、高斯模糊、高斯噪声和椒盐噪声模拟真实环境下的各类噪声攻击,编码器输出的隐写图像经过不同种类的噪声处理,再由解码器解码;通过训练模型,解码器能够对噪声处理后的隐写图像提取秘密信息,以抵抗噪声攻击。实验结果表明,所提方案在360×360像素的图像上隐写容量达到0.45~0.95 bpp,与次优的鲁棒隐写方案相比,相对嵌入容量提升了2.04倍;解码准确率可达0.72~0.97;与未添加噪声层的隐写方案相比,平均解码准确率提高了44个百分点。所提方案在保证高嵌入量、高编码图片质量的同时具有更强的抗噪声攻击能力。
中图分类号:
董炜娜, 刘佳, 潘晓中, 陈立峰, 孙文权. 基于编码-解码网络的大容量鲁棒图像隐写方案[J]. 计算机应用, 2024, 44(3): 772-779.
Weina DONG, Jia LIU, Xiaozhong PAN, Lifeng CHEN, Wenquan SUN. High-capacity robust image steganography scheme based on encoding-decoding network[J]. Journal of Computer Applications, 2024, 44(3): 772-779.
(λe,λd) | 隐写容量/bpp | PSNR/dB | SSIM | 解码准确率 |
---|---|---|---|---|
(20,1) | 0.95 | 34.27 | 0.95 | 0.97 |
(1,1) | 0.97 | 25.97 | 0.68 | 0.98 |
(1,20) | 0.99 | 21.21 | 0.49 | 0.99 |
表1 不同权重参数下模型的效果
Tab. 1 Effects of model under different weight parameters
(λe,λd) | 隐写容量/bpp | PSNR/dB | SSIM | 解码准确率 |
---|---|---|---|---|
(20,1) | 0.95 | 34.27 | 0.95 | 0.97 |
(1,1) | 0.97 | 25.97 | 0.68 | 0.98 |
(1,20) | 0.99 | 21.21 | 0.49 | 0.99 |
模型 | Dropout | JPEG压缩 | 高斯模糊 | 高斯噪声 | 椒盐噪声 |
---|---|---|---|---|---|
无噪声模型 | 0.43 | 0.16 | 0.22 | 0.28 | 0.11 |
噪声模型 | 0.72 | 0.45 | 0.88 | 0.95 | 0.95 |
表2 模型在不同噪声攻击下的隐写容量对比 (bpp)
Tab. 2 Comparison of steganographic capacity between models under different noise attacks
模型 | Dropout | JPEG压缩 | 高斯模糊 | 高斯噪声 | 椒盐噪声 |
---|---|---|---|---|---|
无噪声模型 | 0.43 | 0.16 | 0.22 | 0.28 | 0.11 |
噪声模型 | 0.72 | 0.45 | 0.88 | 0.95 | 0.95 |
隐写方案 | 图像尺寸 | 绝对嵌入容量/b | 相对嵌入 容量/bpp | 鲁棒性 |
---|---|---|---|---|
文献[ | 64×64 | 1 634 | 0.400 000 | 否 |
文献[ | 32×32 | 410 | 0.400 000 | 否 |
文献[ | 64×64 | 300 | 0.070 000 | 否 |
文献[ | 16×16 | 52 | 0.203 000 | 是 |
文献[ | 128×128 | 8 | 0.000 488 | 是 |
文献[ | 256×256 | 200 | 0.001 020 | 是 |
文献[ | 128×128 | 64 | 0.003 900 | 是 |
文献[ | 64×64 | 1 280 | 0.313 000 | 是 |
本文方案 | 360×360 | 123 120 | 0.950 000 | 是 |
表3 不同隐写方案的容量对比
Tab. 3 Capacity comparison of different scheme
隐写方案 | 图像尺寸 | 绝对嵌入容量/b | 相对嵌入 容量/bpp | 鲁棒性 |
---|---|---|---|---|
文献[ | 64×64 | 1 634 | 0.400 000 | 否 |
文献[ | 32×32 | 410 | 0.400 000 | 否 |
文献[ | 64×64 | 300 | 0.070 000 | 否 |
文献[ | 16×16 | 52 | 0.203 000 | 是 |
文献[ | 128×128 | 8 | 0.000 488 | 是 |
文献[ | 256×256 | 200 | 0.001 020 | 是 |
文献[ | 128×128 | 64 | 0.003 900 | 是 |
文献[ | 64×64 | 1 280 | 0.313 000 | 是 |
本文方案 | 360×360 | 123 120 | 0.950 000 | 是 |
模型 | 攻击类别 | PSNR/dB | SSIM |
---|---|---|---|
无噪声模型 | Dropout | 35.68 | 0.92 |
JPEG压缩 | 35.68 | 0.92 | |
高斯模糊 | 35.68 | 0.92 | |
高斯噪声 | 30.47 | 0.72 | |
椒盐噪声 | 35.68 | 0.92 | |
噪声模型 | Dropout | 34.04 | 0.84 |
JPEG压缩 | 33.92 | 0.80 | |
高斯模糊 | 34.76 | 0.88 | |
高斯噪声 | 27.23 | 0.57 | |
椒盐噪声 | 33.83 | 0.89 |
表4 模型在不同噪声攻击下生成图像的PSNR和SSIM对比
Tab. 4 Comparison of PSNR and SSIM of images generated by models under different noise attacks
模型 | 攻击类别 | PSNR/dB | SSIM |
---|---|---|---|
无噪声模型 | Dropout | 35.68 | 0.92 |
JPEG压缩 | 35.68 | 0.92 | |
高斯模糊 | 35.68 | 0.92 | |
高斯噪声 | 30.47 | 0.72 | |
椒盐噪声 | 35.68 | 0.92 | |
噪声模型 | Dropout | 34.04 | 0.84 |
JPEG压缩 | 33.92 | 0.80 | |
高斯模糊 | 34.76 | 0.88 | |
高斯噪声 | 27.23 | 0.57 | |
椒盐噪声 | 33.83 | 0.89 |
模型 | Dropout | JPEG压缩 | 高斯模糊 | 高斯噪声 | 椒盐噪声 |
---|---|---|---|---|---|
无噪声模型 | 0.71 | 0.58 | 0.61 | 0.64 | 0.55 |
噪声模型 | 0.85 | 0.72 | 0.94 | 0.97 | 0.97 |
表5 模型在不同噪声攻击下的解码准确率对比
Tab. 5 Comparison of decoding accuracy of models under different noise attacks
模型 | Dropout | JPEG压缩 | 高斯模糊 | 高斯噪声 | 椒盐噪声 |
---|---|---|---|---|---|
无噪声模型 | 0.71 | 0.58 | 0.61 | 0.64 | 0.55 |
噪声模型 | 0.85 | 0.72 | 0.94 | 0.97 | 0.97 |
隐写方案 | 查看原图 | 不查看原图 | ||
---|---|---|---|---|
隐写容量/bpp | 解码准确率 | 隐写容量/bpp | 解码准确率 | |
文献[ | 0.24 | 0.62 | 0.02 | 0.51 |
本文方案 | 0.66 | 0.83 | 0.08 | 0.54 |
表6 不同方案在查看原图和不查看原图两种方式上的隐写容量和解码准确率对比
Tab. 6 Comparison of steganographic capacity and decoding accuracy among different schemes with/without viewing original image
隐写方案 | 查看原图 | 不查看原图 | ||
---|---|---|---|---|
隐写容量/bpp | 解码准确率 | 隐写容量/bpp | 解码准确率 | |
文献[ | 0.24 | 0.62 | 0.02 | 0.51 |
本文方案 | 0.66 | 0.83 | 0.08 | 0.54 |
1 | 张祎,罗向阳,王金伟,等.数字图像鲁棒隐写综述[J]. 中国图象图形学报,2022,27(1): 3-26. 10.11834/jig.210449 |
ZHANG Y, LUO X Y, WANG J W, et al. Research progress on digital image robust steganography[J]. Journal of Image and Graphics, 2022, 27(1): 3-26. 10.11834/jig.210449 | |
2 | XIAN Y-J, WANG X-Y, ZHANG Y-Q, et al. Fractal sorting vector-based least significant bit chaotic permutation for image encryption [J]. Chinese Physics B, 2021, 30(6): 060508. 10.1088/1674-1056/abda35 |
3 | TAMIMI A A, ABDALLA A M, ALALLAF O. Hiding an image inside another image using variable-rate steganography [J]. International Journal of Advanced Computer Science and Applications, 2013, 4(10): 18-21. 10.14569/ijacsa.2013.041004 |
4 | HOLUB V, FRIDRICH J. Designing steganographic distortion using directional filters [C]// Proceedings of the 2012 IEEE International Workshop on Information Forensics and Security. Piscataway: IEEE, 2012: 234-239. 10.1109/wifs.2012.6412655 |
5 | HOLUB V, FRIDRICH J, DENEMARK T. Universal distortion function for steganography in an arbitrary domain [J]. EURASIP Journal on Information Security, 2014, 2014(1): Article No.1. 10.1186/1687-417x-2014-1 |
6 | LI B, WANG M, HUANG J, et al. A new cost function for spatial image steganography [C]// Proceedings of the 2014 IEEE International Conference on Image Processing. Piscataway: IEEE, 2014: 4206-4210. 10.1109/icip.2014.7025854 |
7 | DE VLEESCHOUWER C, J-F DELAIGLE, MACQ B. Circular interpretation of bijective transformations in lossless watermarking for media asset management [J]. IEEE Transactions on Multimedia, 2003, 5(1): 97-105. 10.1109/tmm.2003.809729 |
8 | ZOU D, SHI Y Q, NI Z, et al. A semi-fragile lossless digital watermarking scheme based on integer wavelet transform [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2006, 16(10): 1294-1300. 10.1109/tcsvt.2006.881857 |
9 | NI Z, SHI Y Q, ANSARI N, et al. Robust lossless image data hiding designed for semi-fragile image authentication [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2008, 18(4): 497-509. 10.1109/tcsvt.2008.918761 |
10 | 李凌敏,侯梦然,陈琨,等.深度学习的可解释性研究综述[J]. 计算机应用,2022,42(12): 3639-3650. |
LI L M, HOU M R, CHEN K, et al. Survey on interpretability research of deep learning[J]. Journal of Computer Applications, 2022, 42(12): 3639-3650. | |
11 | 刘佳,柯彦,雷雨,等.生成对抗网络在图像隐写中的应用[J]. 武汉大学学报(理学版),2019,65(2): 139-152. |
LIU J, KE Y, LEI Y, et al. Application of generative adversarial network in image steganography[J]. Journal of Wuhan University (Natural Science Edition), 2019, 65(2): 139-152. | |
12 | 刘明明,张敏情,刘佳,等.基于生成对抗网络的无载体信息隐藏[J].应用科学学报,2018,36(2): 371-382. 10.3969/j.issn.0255-8297.2018.02.015 |
LIU M M, ZHANG M Q, LIU J, et al. Coverless information hiding based on generative adversarial networks [J]. Journal of Applied Sciences, 2018, 36(2): 371-382. 10.3969/j.issn.0255-8297.2018.02.015 | |
13 | KE Y, ZHANG M-Q, LIU J, et al. Generative steganography with Kerckhoffs’ principle [J]. Multimedia Tools and Applications, 2019, 78(10): 13805-13818. 10.1007/s11042-018-6640-y |
14 | 李福海,蒋慕容,杨磊,等.基于生成对抗网络的梯度引导太阳斑点图像去模糊方法[J]. 计算机应用,2021,41(11): 3345-3352. 10.11772/j.issn.1001-9081.2020121898 |
LI F H, JIANG M R, YANG L, et al. Solar speckle image deblurring method with gradient guidance based on generative adversarial network[J]. Journal of Computer Applications, 2021, 41(11): 3345-3352. 10.11772/j.issn.1001-9081.2020121898 | |
15 | 赵杨,李波.基于增强生成器条件生成对抗网络的单幅图像去雾[J]. 计算机应用,2021,41(12): 3686-3691. |
ZHAO Y, LI B. Single image dehazing based on conditional generative adversarial network with enhanced generator[J]. Journal of Computer Applications, 2021, 41(12): 3686-3691. | |
16 | BALUJA S. Hiding images 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. |
17 | WU P, YANG Y, LI X. StegNet: mega image steganography capacity with deep convolutional network [J]. Future Internet, 2018, 10(6): 54. 10.3390/fi10060054 |
18 | DUAN X, JIA K, LI B, et al. Reversible image steganography scheme based on a U-Net structure [J]. IEEE Access, 2019, 7: 9314-9323. 10.1109/access.2019.2891247 |
19 | RONNEBERGER O, FISCHER P, BROX T. U-net: convolutional networks for biomedical image segmentation [C]// Proceedings of the 18th International Conference on Medical Image Computing and Computer-Assisted Intervention, LNCS 9351. Cham: Springer, 2015: 234-241. |
20 | ZHANG R, DONG S, LIU J. Invisible steganography via generative adversarial networks [J]. Multimedia Tools and Applications, 2019, 78(7): 8559-8575. 10.1007/s11042-018-6951-z |
21 | HAYES J, DANEZIS G. Generating steganographic images via adversarial training [C]// Proceedings of the 31st Annual Conference on Neural Information Processing Systems. Red Hook: Curran Associates Inc., 2017: 1951-1960. |
22 | HU D, WANG L, JIANG W, et al. A novel image steganography method via deep convolutional generative adversarial networks [J]. IEEE Access, 2018, 6: 38303-38314. 10.1109/access.2018.2852771 |
23 | ZHANG K A, CUESTA-INFANTE A, XU L, et al. SteganoGAN: high capacity image steganography with GANs [EB/OL]. (2019-01-30) [2023-02-20]. . |
24 | JING J, DENG X, XU M, et al. HiNet: deep image hiding by invertible network [C]// Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision. Piscataway: IEEE, 2021: 4733-4742. 10.1109/iccv48922.2021.00469 |
25 | GUAN Z, JING J, DENG X, et al. DeepMIH: deep invertible network for multiple image hiding[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 45(1): 372-390. 10.1109/tpami.2022.3141725 |
26 | LU S-P, WANG R, ZHONG T, et al. Large-capacity image steganography based on invertible neural networks [C]// Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2021: 10811-10820. 10.1109/cvpr46437.2021.01067 |
27 | ZHU J, KAPLAN R, JOHNSON J, et al. HiDDeN: hiding data with deep networks [C]// Proceedings of the 15th European Conference on Computer Vision, LNCS 11219. Cham: Springer, 2018: 682-697. |
28 | TANCIK M, MILDENHALL B, NG R. StegaStamp: invisible hyperlinks in physical photographs [C]// Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2020: 2114-2123. 10.1109/cvpr42600.2020.00219 |
29 | XU Y, MOU C, HU Y, et al. Robust invertible image steganography [C]// Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2022: 7875-7884. 10.1109/cvpr52688.2022.00772 |
30 | HUANG G, LIU Z, VAN DER MAATEN L, et al. Densely connected convolutional networks [C]// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2017: 2261-2269. 10.1109/cvpr.2017.243 |
31 | HSU C-T, WU J-L. Hidden digital watermarks in images [J]. IEEE Transactions on Image Processing, 1999, 8(1): 58-68. 10.1109/83.736686 |
32 | 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, LNCS 8693. Berlin: Springer, 2014: 740-755. |
33 | LIU Q, XIANG X, QIN J, et al. A robust coverless steganography scheme using camouflage image [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2022, 32(6): 4038-4051. 10.1109/tcsvt.2021.3108772 |
34 | LI Z, ZHANG M, LIU J. Robust image steganography framework based on generative adversarial network [J]. Journal of Electronic Imaging, 2021, 30(2): 023006. 10.1117/1.jei.30.2.023006 |
35 | MA R, GUO M, HOU Y, et al. Towards blind watermarking: combining invertible and non-invertible mechanisms [C]// Proceedings of the 30th ACM International Conference on Multimedia. New York: ACM, 2022: 1532-1542. 10.1145/3503161.3547950 |
36 | PENG F, CHEN G, LONG M. A robust coverless steganography based on generative adversarial networks and gradient descent approximation [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2022, 32(9): 5817-5829. 10.1109/tcsvt.2022.3161419 |
37 | VOLKHONSKIY D, NAZAROV I, BURNAEV E. Steganographic generative adversarial networks [EB/OL]. [2023-04-01]. . 10.1117/12.2559429 |
38 | LIU X, CHENG M, ZHANG H, et al. Towards robust neural networks via random self-ensemble [C]// Proceedings of the 15th European Conference on Computer Vision. Berlin: Springer, 2018: 381-397. 10.1007/978-3-030-01234-2_23 |
39 | LUO Y, ZHOU T, LIU F, et al. IRWArt: levering watermarking performance for protecting high-quality artwork images [C]// Proceedings of the ACM Web Conference 2023. New York: ACM, 2023: 2340-2348. 10.1145/3543507.3583489 |
[1] | 黄云川, 江永全, 黄骏涛, 杨燕. 基于元图同构网络的分子毒性预测[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2964-2969. |
[2] | 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. |
[3] | 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. |
[4] | 王熙源, 张战成, 徐少康, 张宝成, 罗晓清, 胡伏原. 面向手术导航3D/2D配准的无监督跨域迁移网络[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2911-2918. |
[5] | 李顺勇, 李师毅, 胥瑞, 赵兴旺. 基于自注意力融合的不完整多视图聚类算法[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2696-2703. |
[6] | 刘禹含, 吉根林, 张红苹. 基于骨架图与混合注意力的视频行人异常检测方法[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2551-2557. |
[7] | 顾焰杰, 张英俊, 刘晓倩, 周围, 孙威. 基于时空多图融合的交通流量预测[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2618-2625. |
[8] | 石乾宏, 杨燕, 江永全, 欧阳小草, 范武波, 陈强, 姜涛, 李媛. 面向空气质量预测的多粒度突变拟合网络[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2643-2650. |
[9] | 吴筝, 程志友, 汪真天, 汪传建, 王胜, 许辉. 基于深度学习的患者麻醉复苏过程中的头部运动幅度分类方法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2258-2263. |
[10] | 李欢欢, 黄添强, 丁雪梅, 罗海峰, 黄丽清. 基于多尺度时空图卷积网络的交通出行需求预测[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2065-2072. |
[11] | 张郅, 李欣, 叶乃夫, 胡凯茜. 基于暗知识保护的模型窃取防御技术DKP[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2080-2086. |
[12] | 赵亦群, 张志禹, 董雪. 基于密集残差物理信息神经网络的各向异性旅行时计算方法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2310-2318. |
[13] | 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. |
[14] | 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. |
[15] | 孙逊, 冯睿锋, 陈彦如. 基于深度与实例分割融合的单目3D目标检测方法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2208-2215. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||