Coverless information hiding based on generative adversarial networks

Number of views： 10

分享到:

摘要：

提出了一种基于生成对抗网络的无载体信息隐藏方法.发送方用伪装图像和Improved Wasserstein GAN生成一张与秘密图像相同的图像,发送方发送伪装图像,接收方收到伪装图像后用生成器生成和秘密图像视觉上相同的图像.与现有的信息隐藏方法相比,该方法不需要修改载体,能够有效地抵抗隐写分析工具的检测,能够达到秘密信息安全传输的目的.

This paper proposes a coverless information hiding method based on generative adversarial networks.The sender uses the camouflage image and the improved wasserstein GAN to generate an image that is visually identical to the secret image.The sender sends the camouflage image,and the receiver uses the camouflage image and the generator to generate an image that is visually identical to the secret image.Compared with the existing information hiding method,this method does not need to modify the carrier.It effectively resists the detection of the steganalysis tool and can achieve the purpose of secure transmission of secret information.

作者：

段新涛李宝霞贾凯郭玳豆

Duan Xintao;Li Baoxia;Jia Kai;Guo Daidou(College of Computer and Information Engineering,Henan Normal University,Xinxiang 453007,China)

机构地区：

betway官方app 计算机与信息工程学院

出处：

《betway官方app 学报：自然科学版》 CAS 北大核心 2020年第1期33-37,共5页

基金：

河南省高等学校重点科研项目(19B510005,20B413004,16A520058)。

关键词：

IMPROVED Wasserstein GAN 无载体信息隐藏信息安全

Improved Wasserstein GAN coverless information hiding information safety