| 306 | 4 | 33 |
| 下载次数 | 被引频次 | 阅读次数 |
针对现有图像隐写方法载体实用性、隐藏信息安全性以及信息嵌入与输出准确性较低等问题,提出了一种基于信息最大化生成对抗网络(Information Maximizing Generative Adversarial Networks, InfoGAN)的图像隐写方法。该方法使用潜码(Latent Code)来表示数据的标签,作为生成模型的一个输入与噪声一起生成含密图像,并使用标签判别器分离出标签,以达到解密的效果。与传统信息隐藏方法不同,该方法抛弃了传统信息隐藏方法中信息嵌入的步骤,而使用生成对抗网络生成含密图像,并且将秘密信息"直接"与噪声一起生成含密图像。实验表明,基于InfoGAN的图像隐写方法相较于其他隐写方法具有更好的安全性,并且加密、解密方式简单,易于操作。
Abstract：Considering the low carrier practicality, hidden information security and accuracy of information embedding and output in existing image steganography methods, an image steganography method based on Information Maximizing Generative Adversarial Networks(InfoGAN) is proposed, in which latent code is used to represent the label of data as an input of the generative model to generate a secret image together with noise, and a label discriminator is used to separate the labels to achieve the decryption effect.Different from the traditional information hiding method, this method abandons the information embedding step in the traditional information hiding method, and uses a generative adversarial network to generate the secret image, and the secret information is used together with the noise to “directly” generate the secret image.Experiments show that the image steganography method based on InfoGAN has better security than other steganography methods, and the encryption and decryption methods are simple and easy to operate.
[1] SHI H,DONG J,WANG W,et al.SSGAN:Secure Steganography Based on Generative Adversarial Networks[C]//Pacific Rim Conference on Multimedia.Cham:Springer,2017:534-544.
[2] SALOMON M,COUTURIER R,GUYEUX C,et al.Steganalysis via a Convolutional Neural Network Using Large Convolution Filters for Embedding Process with Same Stego Key:A Deep Learning Approach for Telemedicine[J].European Research in Telemedicine/La Recherche Européenne en Télémédecine,2017,6(2):79-92.
[3] YE J,NI J,YI Y.Deep Learning Hierarchical Representations for Image Steganalysis[J].IEEE Transactions on Information Forensics and Security,2017,12(11):2545-2557.
[4] ZHANG R,DONG S,LIU J.Invisible Steganography via Generative Adversarial Networks[J].Multimedia Tools and Applications,2019,78(7):8559-8575.
[5] CHEN X,DUAN Y,HOUTHOOFT R,et al.InfoGAN:Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets[J].Advances in Neural Information Processing Systems,2016(6):2180-2188.
[6] 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.
[7] 周琳娜，吕欣一.基于GAN图像生成的信息隐藏技术综述[J].信息安全研究，2019,5(9):771-777.
[8] 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.
[9] CUN Y L ,BOSER B ,DENKER J ,et al.Handwritten Digit Recognition with a Back-propagation Network[J].Advances in Neural Information Processing Systems,1990:396-404.
[10] BANERJEE C.Feature Representations Using the Reflected Rectified Linear Unit(RReLU) Activation[J].Big Data Mining and Analytics,2020,3(2):20-38.
[11] FAN E.Extended Tanh-function Method and Its Applications to Nonlinear Equations[J].Physics Letters A,2000,277(4-5):212-218.
[12] ZHANG X,ZOU Y,SHI W.Dilated Convolution Neural Network with LeakyReLU for Environmental Sound Classification[C]//2017 22nd International Conference on Digital Signal Processing (DSP).London:IEEE,2017:1-5.
[13] ZHANG Z,SABUNCU M R.Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labels[J].arXiv:1805.07836v4[cs.LG],2018.
[14] CACHIN C.An Information-theoretic Model for Steganography[C]//International Workshop on Information Hiding.Heidelberg:Springer,1998:306-318.
[15] STAMP M.信息安全原理与实践[J].2版.张戈，译.北京：清华大学出版社，2013.
[16] TANG W,TAN S,LI B,et al.Automatic Steganographic Distortion Learning Using a Generative Adversarial Network[J].IEEE Signal Processing Letters,2017,24(99):1547-1551.
[17] 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.
[18] VOLKHONSKIY D,NAZAROV I,BURNAEV E.Steganographic Generative Adversarial Networks[J].arXiv:1703.05502v2[cs.MM],2019.
基本信息:
DOI：
中图分类号:TP309.7
引用信息:
[1]刘义铭,过小宇,牛一如.基于信息最大化生成对抗网络的图像隐写方法[J].无线电工程,2021,51(11):1214-1219.
基金信息:
国家自然科学基金企业创新发展联合基金—人工智能领域(U20B200063); 四川省科技计划项目(2021JDR274)~~