基于生成对抗网络的密钥生成方法及其在微光图像加密中的应用

doi:10.3969/j.issn.1000-1093.2022.02.011

兵工学报 ›› 2022, Vol. 43 ›› Issue (2): 337-344.doi: 10.3969/j.issn.1000-1093.2022.02.011

基于生成对抗网络的密钥生成方法及其在微光图像加密中的应用

李锦青^1,2，刘泽飞^1,2，满振龙^1,2

(1.长春理工大学计算机科学技术学院, 吉林长春 130022； 2.吉林省网络与信息安全重点实验室，吉林长春 130022)

收稿日期:2021-03-02 修回日期:2021-03-02 接受日期:2021-03-02 上线日期:2022-03-28
通讯作者: 李锦青（1980—）,女,副教授，博士 E-mail:lijinqing@cust.edu.cn
作者简介:刘泽飞（1998—）,女，硕士研究生。E-mail: 1486873001@qq.com
基金资助:
国家重点研发计划项目（2018YFB1800303）；吉林省科技厅自然科学基金项目（20190201188JC）；吉林省高等教育教学改革研究课题（JLLG685520190725093004）

Key Generation Method Based on Generative Adversarial Network and Its Application in Low-light-level Image Encryption

LI Jinqing^1,2, LIU Zefei^1,2, MAN Zhenlong^1,2

(1.School of Computer Science and Technology, Changchun University of Science and Technology, Changchun 130022, Jilin, China;2.Jilin Key Laboratory of Network and Information Security, Changchun 130022, Jilin China)

Received:2021-03-02 Revised:2021-03-02 Accepted:2021-03-02 Online:2022-03-28

摘要/Abstract

摘要： 生成对抗网络（GAN）是一种深度学习模型，是近年来在复杂分布环境下进行无监督学习的最有前景的方法之一。将GAN开创性地引入到随机密钥生成中，利用GAN对超混沌系统产生的随机密钥进行学习和训练。GAN通过学习训练生成的随机数与混沌系统生成的随机数有着很多相似的优点，如随机性和敏感性，但是同时它也具备了混沌系统所生成随机数不具备的特征，如不可复现性。GAN训练生成的随机数在对信噪比低、灰度等级少的微光图像加密中显示出其快速性与更高的安全性。本文将GAN引入到随机密钥生成中，利用量子细胞神经网络系统产生的伪随机数作为GAN的训练集，通过GAN对超混沌系统产生的随机密钥进行学习和训练得到一个随机密钥池，最后针对这种密钥生成方案在微光图像加密中的应用，提出了一种新的微光图像加密算法，该算法给出了一种与明文相关的2D指针，随机选择密钥池中的两个相位掩膜来实现微光图像的安全。结果表明该学习型密钥生成方案所生成的加密密钥可以通过美国国家标准技术研究所的所有随机测试，并且该方案能够有效抵抗差分攻击、已知明文/选择明文攻击和各种统计分析。同时，与其他同类算法的性能比较也进一步表明了该模型的优越性。

关键词: 密钥生成, 生成对抗网络, 深度学习, 微光图像, 加密算法

Abstract: Generative adversarial network (GAN) is a deep learning model, which is one of the most promising methods for unsupervised learning in complex distributed environment in recent years. The random numbers generated by GAN and chaotic systems have many similar advantages, such as randomness and sensitivity. But at the same time,the random numbers generated by GAN have the features that the random numbers generated by chaotic systems do not have, such as non-reproducibility.The random number generated by GAN training shows its rapidity and higher security in the encryption of low-light-level images with low signal-to-noise ratio and few gray levels. GAN is introduced into the random key generation. The pseudo-random number generated by quantum cellular neural network (QCNN) system is used as the training set of GAN, and a completely random key pool is obtained through GAN to learn and train the random key generated by hyperchaotic system. For the application of the proposed key generation method in low-light-level image encryption, a new low-light-level image encryption algorithm is proposed, which uses a 2D-pointer related to the plaintext to randomly select two phase masks in the key pool to achieve the security of low-light-level images. The results show that the encryption key generated by the learning key generation scheme can pass all the random tests of The National Institute of Standards and Technology, and the scheme can resist differential attacks by cryptanalysts, known/chosen plaintext attacks, and various statistical analyses. The performance comparison with other similar algorithms also shows the superiority of the proposed cryptosystem.

Key words: keygeneration, generativeadversarialnetwork, deeplearning, low-light-levelimage, encryptionalgorithm

中图分类号:

李锦青，刘泽飞，满振龙. 基于生成对抗网络的密钥生成方法及其在微光图像加密中的应用[J]. 兵工学报, 2022, 43(2): 337-344.

LI Jinqing, LIU Zefei, MAN Zhenlong. Key Generation Method Based on Generative Adversarial Network and Its Application in Low-light-level Image Encryption[J]. Acta Armamentarii, 2022, 43(2): 337-344.

参考文献

［1］白乐,赖雪峰,韩维强,等.适应积分时间调整的红外图像非均匀性校正方法[J]. 光子学报，2020, 49(1): 169-177.
BAI L，LAI X F, HAN W Q,et al.An infrared image nonuniformity correction method adapted to adjustment of integration time[J].Acta Photonica Sinica,2020,49(1):169-177.(in Chinese)
[2］李佳,李少娟,段小虎,等.基于Retinex理论与概率非局部均值的红外图像增强方法[J].光子学报, 2020，49（4）:182-191.
LI J,LI S J,DUAN X H, et al.Infrared image enhancement based on retinex and probability nonlocal means filtering[J].Acta Photonica Sinica，2020，49（4）:182-191.(in Chinese)
[3］宗艳桃，蒋晓瑜，汪熙，等.车载微光瞄准镜电子稳像技术研究[J]. 兵工学报, 2012, 33(9): 1041-1047.
ZONG Y T, JIANG X Y, WANG X, et al. Research on electronic image stabilization technique for vehicle-borne low-light-level telescope[J]. Acta Armamentarii, 2012, 33(9): 1041-1047.(in Chinese)
[4］曾桂林, 苏秉华, 牛丽红,等. 基于神经网络的微光图像增强与复原[J]. 兵工学报, 2006, 27(4):652-654.
ZENG G L, SU B H, NIU L H, et al. Neural network-based low-light-level image enhancement and reconstruction[J]. Acta Armamentarii, 2006, 27(4): 652-654.(in Chinese)
[5］ MAN Z L, LI J Q, DI X Q, et al. An image segmentation encryption algorithm based on hybrid chaotic system[J]. IEEE Access, 2019,7: 103047-103058.
[6］ DI X Q, LI J Q, QI H, et al. A semi-symmetric image encryption scheme based on the function projective synchronization of two hyperchaotic systems[J]. PloS One, 2017,12(9): e0184586.
[7］ WANG H X, CAI G L, MIAO S, et al. Nonlinear feedback control of a novel hyperchaotic system and its circuit implementation[J]. Chinese Physics B, 2010，19(3):150-157.
[8］ STINSONDR, PATERSON M. Cryptography: theory and practice[M]. Los Angeles， CA，US：CRC Press,2018.
[9］ TSAFACK N, KENGNE J, ABD-EL-ATTY B, et al. Design and implementation of a simple dynamical 4-D chaotic circuit with applications in image encryption[J]. Information Sciences, 2020, 515: 191-217.
[10］ RAKHEJ P, VIG R,SINGH P, et al. Double image encryption using 3D Lorenz chaotic system, 2D non-separable linear canonical transform and QR decomposition[J]. Optical and Quantum Electronics, 2020, 52(2): 103.
[11］ HUANG C, KAIROUZ P, CHEN X, et al. Context-aware generative adversarial privacy[J]. Entropy, 2017, 19(12): 656.
[12］ LIANG X D, LEE L S, DAI W, et al. Dual motion GAN for future-flow embedded video prediction[C]∥Proceedings of the IEEE International Conference on Computer Vision.Venice, Italy: IEEE,2017.
[13］ WANG X J, ZHANG R, SUN Y, et al. KDGAN: Knowledge distillation with generative adversarial networks[C］∥Proceedings of the 32nd Conference on Neural Information Processing Systems. Montréal, Canada: Neural Information Processing Systems Foundation, Inc.,2018: 775-786.
[14］ LIU J, KE Y, LEI Y, et al, Recent advances of image steganography with generative adversarial networks[J]. IEEE Access, 2020,8: 60575-60597.
[15］ DEBernardiEmail M, KHOUZANI M H R, MALACARIA P. Pseudo-random number generation using generative adversarial networks[C]∥Proceedings of Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Cham, Switzerland:Springer, 2018: 191-200.
[16］ FORTUNA L, PORTO D. Quantum-CNN to generate nanoscale chaotic oscillators[J]. International Journal of Bifurcation and Chaos, 2004,14(3): 1085-1089.
[17］ SEYEDZADEH S M, MIRZAKUCHAKI S. A fast color image encryption algorithm based on coupled two-dimensional piecewise chaotic map[J]. Signal Processing, 2012,92(5): 1202-1215.
[18］ LIU Y J, JIANG Z G, XU X P, et al. Optical image encryption algorithm based on hyper-chaos and public-key cryptography[J]. Optics & Laser Technology, 2020, 127: 106171.
[19］贾忠祥，柳银萍.基于自适应与全局置乱的图像加密新算法[J].华东师范大学学报（自然科学版）,2019(6):61-72.
JIA Z X, LIU Y P. Novel image encryption algorithm based on self-adaptive diffusion and combined global scrambling[J]. Journal of East China Normal University (Natural Science), 2019(6): 61-72.(in Chinese)
[20］郭媛,许鑫,敬世伟,等.一种混合混沌虚拟光学图像加密方法[J].光子学报，2019，48（7）：81-89.
GUO Y, XU X, JING S W, et al.Virtual optical image encryption method based on hybrid chaotic system[J].Acta Photonica Sinica，2019，48（7）：81-89. (in Chinese)
[21］ YU S S,ZHOU N R,GONG L H, et al. Optical image encryption algorithm based on phase-truncated short-time fractional Fourier transform and hyper-chaotic system[J]. Optics and Lasers in Engineering, 2020， 124: 105816.
[22］钟彩，潘梅森.一种基于新超混沌Chen系统的置乱替代相结合的加密算法[J].液晶与显示，2020， 35(1): 91-97.
ZHONG C, PAN M S. Encryption algorithm based on scrambling substitution of new hyperchaotic Chen system[J]. Chinese Journal of Liquid Crystals and Displays, 2020， 35(1): 91-97.（in Chinese）
[23］ YIN Q, WANG C H. A new chaotic image encryption scheme using breadth-first search and dynamic diffusion[J]. International Journal of Bifurcation and Chaos, 2018, 28(4):1850047.
[24］ CHAI X L, ZHENG X Y, GAN Z H, et al. An image encryption algorithm based on chaotic system and compressive sensing[J].Signal Processing,2018, 148: 124-144.
[25］ ZHU H G, ZHANG X D, YU H, et al. An image encryption algorithm based on compound homogeneous hyper-chaotic system[J].Nonlinear Dynamics,2017, 89: 61-79.

[1]	李良福，陈卫东，高强，许开銮，刘轩，何曦，钱钧. 基于深度学习的光电系统智能目标识别[J]. 兵工学报, 2022, 43(S1): 162-168.
[2]	普运伟，刘涛涛，郭江，吴海潇. 基于卷积神经网络和模糊函数主脊坐标变换的雷达辐射源信号识别[J]. 兵工学报, 2021, 42(8): 1680-1689.
[3]	朱敏，许爱强，许晴，李睿峰. 基于改进多层核超限学习机的模拟电路故障诊断[J]. 兵工学报, 2021, 42(2): 356-369.
[4]	张凤, 张超, 杨华民, 王发斌. 作战辅助系统中基于深度学习的投影图像颜色补偿方法[J]. 兵工学报, 2021, 42(11): 2418-2423.
[5]	薛灵芝, 曾向阳, 杨爽. 基于生成对抗网络的水声目标识别算法[J]. 兵工学报, 2021, 42(11): 2444-2452.
[6]	余跃，王宏伦. 基于深度学习的高超声速飞行器再入预测校正容错制导[J]. 兵工学报, 2020, 41(4): 656-669.
[7]	闫孟达，杨任农，左家亮，胡东愿，岳龙飞，赵雨. 基于深度学习的空空导弹多类攻击区实时解算[J]. 兵工学报, 2020, 41(12): 2466-2477.
[8]	于军，刘可，郭帅，于广滨，郭振宇. 基于条件深度循环生成对抗网络和动作探索的行星轮轴承剩余寿命预测[J]. 兵工学报, 2020, 41(11): 2170-2178.
[9]	梁杰，任君，李磊，齐航，周红丽. 基于典型几何形状精确回归的机场跑道检测方法[J]. 兵工学报, 2020, 41(10): 2045-2054.
[10]	余永维，彭西，杜柳青，陈天皓. 基于深度学习框架的装配机器人零件实时检测方法[J]. 兵工学报, 2020, 41(10): 2122-2130.
[11]	李从利，张思雨，韦哲，薛松. 基于深度卷积生成对抗网络的航拍图像去厚云方法[J]. 兵工学报, 2019, 40(7): 1434-1442.
[12]	曾桂林，苏秉华，牛丽紅，苏学征. 基于神经网络的微光图像增强与复原[J]. 兵工学报, 2006, 27(4): 652-654.
[13]	陈钱、柏连发、王敏芳. 微光图像序列时域滤波和运动补偿[J]. 兵工学报, 1997, 18(4): 326-329.

基于生成对抗网络的密钥生成方法及其在微光图像加密中的应用

Key Generation Method Based on Generative Adversarial Network and Its Application in Low-light-level Image Encryption

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 13

编辑推荐

Metrics

本文评价