项目名称: 数字视频隐写分析关键技术研究
项目编号: No.61303208
项目类型: 青年科学基金项目
立项/批准年度: 2014
项目学科: 自动化技术、计算机技术
项目作者: 张承乾
作者单位: 天津大学
项目金额: 23万元
中文摘要: 作为信息隐藏对抗技术,隐写分析技术已成为当前信息安全领域的一个前沿研究方向。相比于图像隐写分析技术的快速发展,面向数字视频的隐写分析尚处于发展初期,但其显现的应用前景使得深入展开视频隐写分析研究具有非常重要的理论和现实意义。本项目拟结合视频压缩编码系统特性,围绕视频数据结构中的空域像素,DCT系数和运动矢量三类主要数据元素构建针对性的隐写分析检测模型。定性研究隐写操作对载体统计特性的影响,建立定量模型,对三类数据元素分别引入时空预测模型和高阶谱技术,时域预测和参数估计技术,运动一致性评价技术,在实现隐写嵌入痕迹判别的基础上,进一步有效估测嵌入强度和嵌入容量两个重要隐写参数,为真正实现信息隐藏逆向分析进行有力的探索。本项目将从理论和实际系统两个层面展开对视频隐写的分析研究,其预期成果将形成一系列具有自主知识产权的检测方法及工具,将有助于推动多媒体信息隐写分析技术体系的完善和发展。
中文关键词: 数字视频隐写分析;数字视频隐写;多媒体信息安全;;
英文摘要: As the contrary of steganography, steganalysis has become the highlight in the area of information security. Although video steganalysis is still in its early days compared with image steganalysis, the research on video steganalysis has important theoretical and practical significance. Combined with the video coding system, this project aims at designing steganalysis algorithms to detect steganography using pixels, DCT coefficients and motion vectors. The influences of embedding on cover videos can be investigated by qualitative analysis and then quantificational model can be established. After that, utilizing oftempral-spatial prediction model and high-order spectrum, temporal domain prediction and parameter estimation, motion coherence evaluation respectively for the three kinds of data, the embedding strength and embedding capacity are reliably estimated in addition to successfully exposing the existence of secret message. The video steganalysis research is developed in both of theory and practical system, so that some methods and tools which have independent intellectual property rights can be designed. Therefore, this project can help to improve the development of multimedia steganalysis.
英文关键词: digital video steganalysis;digital video steganography;multimedia security;;