隐写术
散列函数
隐写分析技术
稳健性(进化)
计算机科学
信息隐藏
安全散列算法
沙-2
完美哈希函数
隐写工具
直方图
端到端原则
密码学
人工智能
信息丢失
算法
图像(数学)
理论计算机科学
密码哈希函数
计算机安全
基因
生物化学
化学
作者
Laijin Meng,Xinghao Jiang,Zhenzhen Zhang,Zhaohong Li,Tanfeng Sun
出处
期刊:IEEE Transactions on Circuits and Systems for Video Technology
[Institute of Electrical and Electronics Engineers]
日期:2023-07-01
卷期号:33 (7): 3542-3558
被引量:2
标识
DOI:10.1109/tcsvt.2022.3232790
摘要
Recently, coverless steganography algorithms have attracted increased research attention due to their ability to completely resist steganalysis algorithms. However, the existing algorithms do not attain the same robust balance against geometric and non-geometric attacks. In addition, most of the existing methods need to transmit some auxiliary information along with the stego-images, which increases the cost of the hidden information. In this paper, a robust coverless image steganography algorithm based on a hash generation model is proposed. Different from the existing methods, the hash sequences are generated by an end-to-end CNN model, where the input is the original images, and the output is the corresponding hash sequences. Therefore, no auxiliary information needs to be transmitted when hiding the secret information. Moreover, the attention mechanism and adversarial training are introduced to improve the robustness of the model. The loss function is redesigned to accommodate these operations. Finally, an index structure is built to enhance the mapping efficiency. The experimental results show that the proposed method possesses better robustness and security compared with the state-of-the-art coverless image steganography algorithms.
科研通智能强力驱动
Strongly Powered by AbleSci AI