• 中国计算机学会会刊
• CCF推荐中文期刊(B类)
• Scopus收录期刊
• CSCD收录期刊
• 中国科技核心期刊

手机二维码
 
【打印本页】      【下载PDF全文】   查看/发表评论  下载PDF阅读器  关闭
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 132次   下载 213 本文二维码信息
码上扫一扫!
面向无载体信息隐藏的映射关系智能搜索方法
王亚宁,吴槟
分享到: 微信 更多
(中国科学院信息工程研究所信息安全国家重点实验室 北京 中国 100093;中国科学院大学 网络空间安全学院 北京 中国 100049)
摘要:
传统的搜索式无载体信息隐藏技术建立在固定的映射规则与庞大的图像库基础上,依赖于复杂的人工特征提取并且需要进行大量搜索来构建合适的图像库。针对这些问题,本文提出了一种面向无载体信息隐藏的基于深度学习映射关系智能搜索方法,该方法从已有图像库出发,基于深度神经网络,自动搜索一套高容量、高覆盖率的映射关系,从而解决传统人工方法存在的传输开销大、图像库建立困难的问题。除此以外,实验表明我们的方法相较于传统无载体方法有更强的鲁棒性。
关键词:  信息隐藏  隐写术  无载体信息隐藏  深度学习
DOI:10.19363/J.cnki.cn10-1380/tn.2020.05.05
投稿时间:2020-01-25修订日期:2020-05-03
基金项目:本课题得到国家自然科学基金(No.U1936119,No.61941116)和国家重点研发计划课题(GrantNo.2019QY(Y)0602)资助。
An intelligent search method of mapping relation for coverless information hiding
WANG Yaning,WU Bin
State Key Laboratory of Information, Security, Institute of Information Engineering, Chinese Academy of Sciences, Beijing 100093, China;School of Cyber Security, University of Chinese Academy of Sciences, Beijing 100049, China
Abstract:
Based on fixed mapping rules and large image database, traditional searching coverless information hiding technology relies on complex artificial feature extraction and takes a lot of searching to construct available image database. To solve these problems, this paper proposed an intelligent search method of mapping relation for coverless information hiding based on deep learning. This framework based on the collected image set and automatically searched for a set of high-capacity, high-coverage mapping rules based on deep neural networks, thereby solving the problems:large transmission overhead and a large image database of traditional coverless methods. In addition, experiments show that our method is more robust than the traditional coverless methods.
Key words:  information hiding  steganography  coverless steganography  deep learning