计算机应用 ›› 2014, Vol. 34 ›› Issue (10): 2840-2843.DOI: 10.11772/j.issn.1001-9081.2014.10.2840

• 计算机安全 • 上一篇    下一篇

基于运动矢量直方图不变的数字视频隐写算法

郭朝江,张敏情,钮可   

  1. 武警工程大学 电子技术系,西安 710086
  • 收稿日期:2014-03-10 修回日期:2014-04-20 出版日期:2014-10-01 发布日期:2014-10-30
  • 通讯作者: 郭朝江
  • 作者简介:郭朝江(1989-),男,安徽淮北人,硕士研究生,主要研究方向:信息隐藏;
    张敏情(1967-),女,陕西西安人,教授,博士,主要研究方向:信息安全、密码学;
    钮可(1981-),男,浙江湖州人,硕士,主要研究方向:信息隐藏。
  • 基金资助:

    国家自然科学基金资助项目

Video steganography algorithm based on invariant histogram of motion vector

GUO Chaojiang,ZHANG Minqing,NIU Ke   

  1. Department of Electronic Technology, Engineering University of Armed Police Force, Xian Shaanxi 710086, China
  • Received:2014-03-10 Revised:2014-04-20 Online:2014-10-01 Published:2014-10-30
  • Contact: GUO Chaojiang

摘要:

针对现有基于运动矢量(MV)的数字视频隐写算法对载体直方图统计特性带来较大改变这一问题,提出一种基于运动矢量直方图不变的数字视频隐写算法。利用保持直方图数据映射方法,把秘密信息隐藏在视频运动矢量中;同时,利用数据匹配编码对嵌入之前的秘密信息进行编码处理,得到了与视频运动矢量统计特征基本匹配的数据流,使算法接近于信息论下的完美安全。实验结果表明:算法有效控制了运动矢量直方图的改变,同时码率的增加也被有效地控制在1%以内,隐写分析检出率平均下降了30%~50%。

Abstract:

To solve the problem that some video steganography algorithms based on Motion Vector (MV) change statistical features of histogram, a new video steganography algorithm based on keeping histogram of MV was proposed. In this paper, the secret information were hidden in video MV by using the data mapping of histogram. At the same time, a code-matching method was used to encode the secret information before being embedded, then the data stream would have the same statistical characteristics by MV, which made the scheme absolutely secure in theory. The experimental results show that: the change of histogram features is effectively controlled, the increment of bitrate is controlled in 1%, and the steganalysis detection rate is decreased by an average of 30% to 50%.

中图分类号: