《计算机应用研究》|Application Research of Computers

基于噪声感知残差网络的JPEG隐写分析方法

JPEG steganalysis based on image noise-aware residual network

免费全文下载 (已被下载 次)  
获取PDF全文
作者 李德维,任魏翔,王丽娜,方灿铭,吴畑
机构 武汉大学 国家网络安全学院 空天信息安全与可信计算教育部重点实验室,武汉 430072
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2021)10-047-3148-05
DOI 10.19734/j.issn.1001-3695.2021.01.0081
摘要 为了进一步挖掘自适应JPEG隐写图像中隐写噪声信号特征,提出基于噪声感知残差网络的JPEG隐写分析方法。该方法由噪声感知、噪声分析和判断三部分组成。其中,噪声感知部分提取图像噪声,利用图像去噪网络,更加全面地捕获隐写引入的扰动;噪声分析部分获得噪声信息的统计特征;判断部分确定图像是否携带隐写信息。此外,网络中的残差连接有效融合多尺度特征,并防止训练中出现梯度消失和爆炸。多种条件下的对比实验结果表明,该方法相较于对比算法,能够提升针对JPEG自适应隐写的检测性能并具有更好的泛化能力。
关键词 JPEG隐写分析; 深度学习; 噪声感知; 残差网络; JPEG自适应隐写
基金项目 国家自然科学基金资助项目(U1836112,61876134,U1536204)
本文URL http://www.arocmag.com/article/01-2021-10-047.html
英文标题 JPEG steganalysis based on image noise-aware residual network
作者英文名 Li Dewei, Ren Weixiang, Wang Lina, Fang Canming, Wu Tian
机构英文名 Key Laboratory of Aerospace Information Security & Trusted Computing of Ministry of Education,School of Cyber Science & Engineering,Wuhan University,Wuhan 430072,China
英文摘要 To further explore the characteristics of the steganographic noise signal in the adaptive JPEG steganographic images, this paper proposed a noise-aware residual network for JPEG image steganalysis. The proposed network included three components, noise-aware component, noise analysis component and judgement component. First, noise-aware component extracted the noise of an image, it used denoising network to offer more comprehensive information related to steganography. Then, noise analysis component got discriminative features from image noise related information. Last, judgement network gave the final detection result. Moreover, residual connections integrated high-level features and low-level features and prevented gradient vanishing and exploding. Experiments under various conditions show that the network has an advanced performance, and has better generalization ability.
英文关键词 JPEG steganalysis; deep learning; noise-aware; residual network; JPEG adaptive steganography
参考文献 查看稿件参考文献
 
收稿日期 2021/1/19
修回日期 2021/3/3
页码 3148-3152,3165
中图分类号 TP309.2
文献标志码 A