《计算机应用》唯一官方网站 ›› 2026, Vol. 46 ›› Issue (1): 144-151.DOI: 10.11772/j.issn.1001-9081.2025020144
收稿日期:2025-02-18
修回日期:2025-05-25
接受日期:2025-05-28
发布日期:2026-01-10
出版日期:2026-01-10
通讯作者:
覃晶滢
作者简介:马英杰(1979—),女,吉林通化人,副教授,博士,主要研究方向:混沌保密通信基金资助:
Yingjie MA, Jingying QIN(
), Geng ZHAO, Jing XIAO
Received:2025-02-18
Revised:2025-05-25
Accepted:2025-05-28
Online:2026-01-10
Published:2026-01-10
Contact:
Jingying QIN
About author:MA Yingjie, born in 1979, Ph. D., associate professor. Her research interests include chaotic secure communication.Supported by:摘要:
针对物联网大量冗余图像数据的传输和存储带来的高资源需求和隐私泄露隐患问题,提出一种面向物联网图像的深度压缩感知(DCS)网络及其混沌加密保护方法。首先,提出一种改进的DCS网络用于图像的高质量压缩重构,该网络通过通道注意力机制改进传统深度重构网络中的残差块,并结合多尺度分支与融合模块进行并行融合设计,改进了传统的基于卷积层残差堆叠的深度重构网络的重构性能;其次,提出一种腔体混沌系统通过球坐标变换和2组奇偶控制的阶梯函数实现球形腔体在X、Y、Z任意单方向或者双方向的腔体拓展,具有较好的混沌特性和随机性,可有效用于图像加密;最后,结合提出的DCS网络和多腔体混沌系统,基于混沌索引置乱和扩散方法对DCS的测量图像进行加解密设计,并进行详细的安全性分析,从而保证图像传输的安全性。实验结果表明,相较于DCS的经典方法CSNet+,该DCS网络重构图像的峰值信噪比(PSNR)平均增长了0.606 dB(0.25~1.42 dB),结构相似性指数(SSIM)平均提高了1.11个百分点(0.69~2.17个百分点)。
中图分类号:
马英杰, 覃晶滢, 赵耿, 肖靖. 面向物联网图像的深度压缩感知网络及其混沌加密保护方法[J]. 计算机应用, 2026, 46(1): 144-151.
Yingjie MA, Jingying QIN, Geng ZHAO, Jing XIAO. Deep compressive sensing network for IoT images and its chaotic encryption protection method[J]. Journal of Computer Applications, 2026, 46(1): 144-151.
| 方法 | 不同采样比r下的PSNR | 平均PSNR | ||||
|---|---|---|---|---|---|---|
| r=0.1 | r=0.2 | r=0.3 | r=0.4 | r=0.5 | ||
| DWT | 23.46 | 27.26 | 29.23 | 30.72 | 32.17 | 28.568 |
| MH | 25.16 | 28.09 | 29.85 | 31.35 | 32.86 | 29.462 |
| GSR | 25.91 | 29.18 | 31.33 | 33.20 | 34.94 | 30.912 |
| CSNet+ | 28.53 | 31.05 | 33.08 | 34.91 | 36.68 | 32.850 |
| 本文方法 | 28.78 | 32.47 | 33.42 | 35.51 | 37.10 | 33.456 |
表1 不同方法在BSDS100数据集上的PSNR比较 ( dB)
Tab. 1 PSNR comparison of different methods on BSDS100 dataset
| 方法 | 不同采样比r下的PSNR | 平均PSNR | ||||
|---|---|---|---|---|---|---|
| r=0.1 | r=0.2 | r=0.3 | r=0.4 | r=0.5 | ||
| DWT | 23.46 | 27.26 | 29.23 | 30.72 | 32.17 | 28.568 |
| MH | 25.16 | 28.09 | 29.85 | 31.35 | 32.86 | 29.462 |
| GSR | 25.91 | 29.18 | 31.33 | 33.20 | 34.94 | 30.912 |
| CSNet+ | 28.53 | 31.05 | 33.08 | 34.91 | 36.68 | 32.850 |
| 本文方法 | 28.78 | 32.47 | 33.42 | 35.51 | 37.10 | 33.456 |
| 方法 | 不同采样比r下的SSIM | 平均SSIM | ||||
|---|---|---|---|---|---|---|
| r=0.1 | r=0.2 | r=0.3 | r=0.4 | r=0.5 | ||
| DWT | 0.634 3 | 0.751 6 | 0.810 8 | 0.852 4 | 0.886 2 | 0.787 0 |
| MH | 0.667 3 | 0.774 6 | 0.830 7 | 0.869 5 | 0.901 2 | 0.808 6 |
| GSR | 0.707 1 | 0.815 6 | 0.872 3 | 0.909 6 | 0.935 9 | 0.848 1 |
| CSNet+ | 0.783 4 | 0.872 1 | 0.917 1 | 0.944 3 | 0.961 8 | 0.895 7 |
| 本文方法 | 0.805 1 | 0.883 4 | 0.924 8 | 0.952 1 | 0.968 7 | 0.906 8 |
表2 不同方法在BSDS100数据集上的SSIM比较
Tab. 2 SSIM comparison of different methods on BSDS100 datasets
| 方法 | 不同采样比r下的SSIM | 平均SSIM | ||||
|---|---|---|---|---|---|---|
| r=0.1 | r=0.2 | r=0.3 | r=0.4 | r=0.5 | ||
| DWT | 0.634 3 | 0.751 6 | 0.810 8 | 0.852 4 | 0.886 2 | 0.787 0 |
| MH | 0.667 3 | 0.774 6 | 0.830 7 | 0.869 5 | 0.901 2 | 0.808 6 |
| GSR | 0.707 1 | 0.815 6 | 0.872 3 | 0.909 6 | 0.935 9 | 0.848 1 |
| CSNet+ | 0.783 4 | 0.872 1 | 0.917 1 | 0.944 3 | 0.961 8 | 0.895 7 |
| 本文方法 | 0.805 1 | 0.883 4 | 0.924 8 | 0.952 1 | 0.968 7 | 0.906 8 |
| 统计检验 | P-value | 结果 |
|---|---|---|
| Frequency | 0.678 686 | Pass |
| BlockFrequency | 0.816 537 | Pass |
| CumulativeSums1 | 0.657 933 | Pass |
| CumulativeSums2 | 0.867 692 | Pass |
| Runs | 0.129 620 | Pass |
| LongestRun | 0.383 827 | Pass |
| Rank | 0.678 686 | Pass |
| FFT | 0.304 126 | Pass |
| NonOverlappingTemplate | 0.719 747 | Pass |
| OverlappingTemplate | 0.171 876 | Pass |
| Universal | 0.202 268 | Pass |
| ApproximateEntropy | 0.304 126 | Pass |
| RandomExcursions | 0.657 933 | Pass |
| RandomExcursionsVariant | 0.319 084 | Pass |
| Serial 1 | 0.699 313 | Pass |
| Serial 2 | 0.350 485 | Pass |
| LinearComplexity | 0.699 313 | Pass |
表3 NIST测试结果
Tab. 3 NIST test results
| 统计检验 | P-value | 结果 |
|---|---|---|
| Frequency | 0.678 686 | Pass |
| BlockFrequency | 0.816 537 | Pass |
| CumulativeSums1 | 0.657 933 | Pass |
| CumulativeSums2 | 0.867 692 | Pass |
| Runs | 0.129 620 | Pass |
| LongestRun | 0.383 827 | Pass |
| Rank | 0.678 686 | Pass |
| FFT | 0.304 126 | Pass |
| NonOverlappingTemplate | 0.719 747 | Pass |
| OverlappingTemplate | 0.171 876 | Pass |
| Universal | 0.202 268 | Pass |
| ApproximateEntropy | 0.304 126 | Pass |
| RandomExcursions | 0.657 933 | Pass |
| RandomExcursionsVariant | 0.319 084 | Pass |
| Serial 1 | 0.699 313 | Pass |
| Serial 2 | 0.350 485 | Pass |
| LinearComplexity | 0.699 313 | Pass |
| 采样比 | SSIM | PSNR/dB |
|---|---|---|
| 0.1 | 0.825 7 | 29.47 |
| 0.2 | 0.904 9 | 32.54 |
| 0.3 | 0.940 1 | 34.71 |
| 0.4 | 0.961 4 | 36.43 |
| 0.5 | 0.970 8 | 38.21 |
表4 本文方法在SET14数据集上的解密重构图像质量评价
Tab. 4 Evaluation of decrypted and reconstructed image quality by proposed method on SET14 dataset
| 采样比 | SSIM | PSNR/dB |
|---|---|---|
| 0.1 | 0.825 7 | 29.47 |
| 0.2 | 0.904 9 | 32.54 |
| 0.3 | 0.940 1 | 34.71 |
| 0.4 | 0.961 4 | 36.43 |
| 0.5 | 0.970 8 | 38.21 |
| 方法 | PSNR/dB | 加密图像信息熵 | 加密图像相邻像素相关性 | ||
|---|---|---|---|---|---|
| 水平方向 | 垂直方向 | 对角线方向 | |||
| 文献[ | 30.64 | 7.994 8 | -0.017 5 | 0.005 9 | -0.018 5 |
| 文献[ | 34.37 | 7.998 6 | -0.001 7 | 0.007 2 | -0.000 9 |
| 本文方法 | 41.57 | 7.993 5 | 0.000 4 | -0.007 9 | 0.005 1 |
表5 不同方法的加解密效果对比
Tab. 5 Comparison of encryption and decryption effects of different methods
| 方法 | PSNR/dB | 加密图像信息熵 | 加密图像相邻像素相关性 | ||
|---|---|---|---|---|---|
| 水平方向 | 垂直方向 | 对角线方向 | |||
| 文献[ | 30.64 | 7.994 8 | -0.017 5 | 0.005 9 | -0.018 5 |
| 文献[ | 34.37 | 7.998 6 | -0.001 7 | 0.007 2 | -0.000 9 |
| 本文方法 | 41.57 | 7.993 5 | 0.000 4 | -0.007 9 | 0.005 1 |
| [1] | LV T, LIN Z, HUANG P, et al. Optimization of the energy-efficient relay-based massive IoT network [J]. IEEE Internet of Things Journal, 2018, 5(4): 3043-3058. |
| [2] | LI X, XU L D. A review of Internet of Things — resource allocation [J]. IEEE Internet of Things Journal, 2021, 8(1): 8657-8666. |
| [3] | LI S, XU L D, WANG X. Compressed sensing signal and data acquisition in wireless sensor networks and Internet of Things [J]. IEEE Transactions on Industrial Informatics, 2013, 9(4): 2177-2186. |
| [4] | LI Z, HUANG H, MISRA S. Compressed sensing via dictionary learning and approximate message passing for multimedia Internet of Things [J]. IEEE Internet of Things Journal, 2017, 4(2): 502-512. |
| [5] | YAMAÇ M, AHISHALI M, PASSALIS N, et al. Multi-level reversible data anonymization via compressive sensing and data hiding [J]. IEEE Transactions on Information Forensics and Security, 2021, 16: 1014-1028. |
| [6] | SUO Z, XIA C, JIANG D, et al. Multitiered reversible data privacy protection scheme for IoT based on compression sensing and digital watermarking [J]. IEEE Internet of Things Journal, 2024, 11(7): 11524-11539. |
| [7] | SHEN M, GAN H, NING C, et al. TransCS: a Transformer-based hybrid architecture for image compressed sensing [J]. IEEE Transactions on Image Processing, 2022, 31: 6991-7005. |
| [8] | ZHANG J, GHANEM B. ISTA-Net: interpretable optimization-inspired deep network for image compressive sensing [C]// Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2018: 1828-1837. |
| [9] | SHI W, JIANG F, LIU S, et al. Image compressed sensing using convolutional neural network [J]. IEEE Transactions on Image Processing, 2020, 29: 375-388. |
| [10] | YOU D, XIE J, ZHANG J. ISTA-Net++: flexible deep unfolding network for compressive sensing [C]// Proceedings of the 2021 IEEE International Conference on Multimedia and Expo. Piscataway: IEEE, 2021: 1-6. |
| [11] | ZHANG M, ZHANG H, YUAN D, et al. Learning-based sparse data reconstruction for compressed data aggregation in IoT networks [J]. IEEE Internet of Things Journal, 2021, 8(14): 11732-11742. |
| [12] | ZHANG M, ZHANG H, ZHANG C, et al. Communication-efficient quantized deep compressed sensing for edge-cloud collaborative industrial IoT networks [J]. IEEE Transactions on Industrial Informatics, 2023, 19(5): 6613-6623. |
| [13] | CHANG Q, TAO D, WANG J, et al. Deep compressed sensing based data imputation for urban environmental monitoring [J]. ACM Transactions on Sensor Networks, 2024, 20(1): No.17. |
| [14] | CHEN B, TANG D, HUANG Y, et al. CMCL: cross-modal compressive learning for resource-constrained intelligent IoT systems [J]. IEEE Internet of Things Journal, 2024, 11(15): 25534-25542. |
| [15] | WANG M, WANG X, WANG C, et al. Spatiotemporal chaos in cross coupled map lattice with dynamic coupling coefficient and its application in bit-level color image encryption [J]. Chaos, Solitons and Fractals, 2020, 139: No.110028. |
| [16] | CHEN D, SUN Z, MA X, et al. Circuit implementation and model of a new multiscroll chaotic system [J]. International Journal of Circuit Theory and Applications, 2014, 42(4): 407-424. |
| [17] | BOUALLEGUE K. Generation of multi-scroll chaotic attractors from fractal and multi-fractal processes [C]// Proceedings of the 4th International Workshop on Chaos-Fractals Theories and Applications. Piscataway: IEEE, 2011: 398-402. |
| [18] | HE S, BANERJEE S. Multicavity formations and complexity modulation in a hyperchaotic discrete system [J]. Physica A: Statistical Mechanics and its Applications, 2018, 490: 366-377. |
| [19] | 摆玉龙,杨阳,唐丽红.一个新多涡卷混沌系统的设计及在图像加密中的应用[J].电子与信息学报, 2021, 43(2): 436-444. |
| BAI Y L, YANG Y, TANG L H. Design of a multi-scroll chaotic system and its application to image encryption [J]. Journal of Electronics and Information Technology, 2021, 43(2): 436-444. | |
| [20] | 徐昌彪,黎金龙,许浩南.一种基于Logistic电平脉冲的多涡卷系统及其图像加密应用[J].电子与信息学报, 2022, 44(12): 4328-4336. |
| XU C B, LI J L, XU H N. A multi-scroll system and its application for image encryption based on logistic level pulse [J]. Journal of Electronics and Information Technology, 2022, 44(12): 4328-4336. | |
| [21] | 刘思洋,安新磊,施倩倩,等.一类多涡卷Chua系统及其在图像加密中的应用[J].复杂系统与复杂性科学, 2024, 21(3): 85-92. |
| LIU S Y, AN X L, SHI Q Q, et al. A multi-scroll Chua system and its application in image encryption [J]. Complex Systems and Complexity Science, 2024, 21(3): 85-92. | |
| [22] | 赵耿,马英杰.混沌应用密码学[M].北京:科学出版社, 2021. |
| ZHAO G, MA Y J. Chaos applied cryptography [M]. Beijing: Science Press, 2021. | |
| [23] | ZHANG Y, TANG Y. A plaintext-related image encryption algorithm based on chaos [J]. Multimedia Tools and Applications, 2018, 77(6): 6647-6669. |
| [24] | LIU W, SUN K, HE S. SF-SIMM high-dimensional hyperchaotic map and its performance analysis [J]. Nonlinear Dynamics, 2017, 89(4): 2521-2532. |
| [25] | XIAO Y, SUN K, HE S. Constructing chaotic map with multi-cavity [J]. European Physical Journal Plus, 2020, 135: No.21. |
| [26] | WU C, SUN K, XIAO Y. A hyperchaotic map with multi-elliptic cavities based on modulation and coupling [J]. The European Physical Journal Special Topics, 2021, 230: 2011-2020. |
| [27] | MUN S, FOWLER J E. Block compressed sensing of images using directional transforms [C]// Proceedings of the 16th IEEE International Conference on Image Processing. Piscataway: IEEE, 2009: 3021-3024. |
| [28] | CHEN C, TRAMEL E W, FOWLER J E. Compressed-sensing recovery of images and video using multihypothesis predictions [C]// Proceedings of the 45th Asilomar Conference on Signals, Systems and Computers. Piscataway: IEEE, 2011: 1193-1198. |
| [29] | 马英杰,肖靖,赵耿,等.可控网格多涡卷混沌系统族及其硬件电路实现[J].计算机应用, 2023, 43(3): 956-961. |
| MA Y J, XIAO J, ZHAO G, et al. Controllable grid multi-scroll chaotic system family and its hardware circuit implementation [J]. Journal of Computer Applications, 2023, 43(3): 956-961. | |
| [30] | 潘涛,佟晓筠,张淼,等.基于压缩感知和超混沌系统的图像压缩加密方法[J].计算机科学, 2023, 50(6A): No.220200121. |
| PAN T, TONG X J, ZHANG M, et al. Image compression and encryption method based on compressed sensing and hyperchaotic system [J]. Computer Science, 2023, 50(6A): No.220200121. | |
| [31] | 范海菊,岳爽,窦育强,等.基于新型混沌系统和二进制块压缩感知的图像加密算法[J/OL].计算机科学[2025-05-07]. . |
| FAN H J, YUE S, DOU Y Q, et al. Image encryption algorithm based on a novel chaotic system and binary block compressed sensing [J/OL]. Computer Science [2025-05-07]. . |
| [1] | 王丽芳, 任文婧, 郭晓东, 张荣国, 胡立华. 用于低剂量CT图像降噪的多路特征生成对抗网络[J]. 《计算机应用》唯一官方网站, 2026, 46(1): 270-279. |
| [2] | 麦超云, 张洪燚, 秦传波, 曾军英, 王栋. 基于多尺度与空间频率特征的嗜铬细胞瘤图像分割网络[J]. 《计算机应用》唯一官方网站, 2026, 46(1): 280-288. |
| [3] | 李亚男, 郭梦阳, 邓国军, 陈允峰, 任建吉, 原永亮. 基于多模态融合特征的并分支发动机寿命预测方法[J]. 《计算机应用》唯一官方网站, 2026, 46(1): 305-313. |
| [4] | 昝志辉, 王雅静, 李珂, 杨智翔, 杨光宇. 基于SAA-CNN-BiLSTM网络的多特征融合语音情感识别方法[J]. 《计算机应用》唯一官方网站, 2026, 46(1): 69-76. |
| [5] | 郭伟, 王曼婷, 曲海成. 基于多尺度感知的多维空间融合水下图像增强算法[J]. 《计算机应用》唯一官方网站, 2026, 46(1): 224-232. |
| [6] | 王婷婷, 李廷顺, 谭文, 吕博, 陈翼轩. 基于多尺度Patch与卷积交互的电缆温度预测模型[J]. 《计算机应用》唯一官方网站, 2026, 46(1): 314-321. |
| [7] | 吕景刚, 彭绍睿, 高硕, 周金. 复频域注意力和多尺度频域增强驱动的语音增强网络[J]. 《计算机应用》唯一官方网站, 2025, 45(9): 2957-2965. |
| [8] | 邓伊琳, 余发江. 基于LSTM和可分离自注意力机制的伪随机数生成器[J]. 《计算机应用》唯一官方网站, 2025, 45(9): 2893-2901. |
| [9] | 梁一鸣, 范菁, 柴汶泽. 基于双向交叉注意力的多尺度特征融合情感分类[J]. 《计算机应用》唯一官方网站, 2025, 45(9): 2773-2782. |
| [10] | 李维刚, 邵佳乐, 田志强. 基于双注意力机制和多尺度融合的点云分类与分割网络[J]. 《计算机应用》唯一官方网站, 2025, 45(9): 3003-3010. |
| [11] | 王翔, 陈志祥, 毛国君. 融合局部和全局相关性的多变量时间序列预测方法[J]. 《计算机应用》唯一官方网站, 2025, 45(9): 2806-2816. |
| [12] | 敬超, 全育涛, 陈艳. 基于多层感知机-注意力模型的功耗预测算法[J]. 《计算机应用》唯一官方网站, 2025, 45(8): 2646-2655. |
| [13] | 林进浩, 罗川, 李天瑞, 陈红梅. 基于跨尺度注意力网络的胸部疾病分类方法[J]. 《计算机应用》唯一官方网站, 2025, 45(8): 2712-2719. |
| [14] | 周金, 李玉芝, 张徐, 高硕, 张立, 盛家川. 复杂电磁环境下的调制识别网络[J]. 《计算机应用》唯一官方网站, 2025, 45(8): 2672-2682. |
| [15] | 颜承志, 陈颖, 钟凯, 高寒. 基于多尺度网络与轴向注意力的3D目标检测算法[J]. 《计算机应用》唯一官方网站, 2025, 45(8): 2537-2545. |
| 阅读次数 | ||||||
|
全文 |
|
|||||
|
摘要 |
|
|||||