基于QGAN的大容量构造式信息隐藏算法  

A Large Capacity Constructive Information Hiding Algorithm Based on QGAN

作　　者：陈纪霞 CHEN Jixia(University of Science and Technology of China,School of Computer Science and Technology,Hefei 230026,China;Anhui Finance&Trade Vocational College,School of Information Engineering,Hefei 230601,China)

机构地区：[1]中国科学技术大学计算机科学与技术学院,合肥230026 [2]安徽财贸职业学院信息工程学院,合肥230601

出　　处：《西安文理学院学报(自然科学版)》2025年第1期32-37,共6页Journal of Xi’an University(Natural Science Edition)

基　　金：高校优秀青年骨干人才国内访学项目(gxgnfx2021216);安徽省高校自然科学研究重点项目(KJ2020A1119)

摘　　要：大容量构造式信息存在嵌入容量与隐蔽性之间的权衡问题,当嵌入的信息量增加时,会对载体数据造成更大的修改,容易受到攻击者的检测和破解,为增强信息安全性,实现信息隐蔽传输,研究基于QGAN的大容量构造式信息隐藏算法.采用经典生成对抗网络的判别器与参数化变分量子电路的生成器,组合成量子生成对抗网络(Quantum Generative Adversarial Networks, QGAN),引入量子切尔诺夫界,优化QGAN的量子线路与迭代方式;建立秘密信息比特与输入态量子比特两者的映射关系,将秘密信息的映射结果作为优化后QGAN生成器的输入状态;将生成含密量子图像样本与真实量子图像样本作为判别器的输入,进行迭代优化,将秘密信息隐藏在量子图像中,实现信息隐藏.实验结果表明:该算法可以高效地完成信息隐藏,并且具备大容量构造式信息隐藏的有效性.There is a trade-off between embedding capacity and concealment in high-capacity structured information.As the amount of embedded information increases,it causes greater modifications to the carrier data,making it vulnerable to detect and cracking by attackers.To enhance information security and achieve covert information transmission,a high-capacity structured information hiding algorithm based on QGAN is proposed.By combining the discriminator of classical Generative Adversarial NSetwork(GAN)with the generator of parameterized variational quantum circuits,a Quantum Generative Adversarial Network(QGAN)is formed.The quantum Chernov bound is introduced to optimize the quantum circuit and iteration method of QGAN;a mapping relationship between secret information bits and input state quantum bits is established,and the mapping results of secret information are used as the input states of the optimized QGAN generator.The generated encrypted quantum image samples and real quantum image samples are input to the discriminator for iterative optimization,and secret information is hided in the quantum images,thereby achieving information hiding.The experimental results show that this algorithm can efficiently complete information hiding and is effective for high-capacity structured information hiding.

关 键 词：QGAN 大容量 构造式信息 信息隐藏算法 切尔诺夫界 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]