【摘 要】
：
Aimed at the problem that existing net-flow identification models only take capacity into consideration but lack of unified evaluation standards of analysis of robustness and invisibility, net-flow id
【机 构】
：
Zhengzhou Information Science and Technology Institute, Zhengzhou, Henan province 450001 ;Henan Prov
【出 处】
：
第十二届全国信息隐藏暨多媒体信息安全学术大会
论文部分内容阅读
Aimed at the problem that existing net-flow identification models only take capacity into consideration but lack of unified evaluation standards of analysis of robustness and invisibility, net-flow identification model based on optimization theory is proposed.Firstly, by analyzing existing net-flow identification systems and net-flow transformation process, it defines the net-flow identification model and summarizes net-flow transformation problems.Secondly, the definition and relationship of robustness and invisibility are given.Besides, according to existing attacks, robustness and invisibility of models are classified into three levels, and K-S goodness-to-fit test is used to judge the level of net-flow identification model.Whats more, it proves the existence of capacity.By using optimization theory, it analyzes maximum capacity of model under different conditions of robustness,invisibility and net-flow transformation problem constraints.Finally, simulation experiments verify the correctness of net-flow identification model based on optimization theory.
其他文献
提出一种基于图像矢量量化压缩和自嵌入机制的脆弱水印算法.通过建立码本并利用矢量量化方法对原始图像每个不重叠的分块进行压缩,获得可表示图像主要内容的压缩码.将置乱后的压缩码通过随机矩阵进行扩展,增加了待嵌入的内容恢复比特中的冗余,并使得原始图像中各个分块与每个内容恢复比特间建立相关性,可提高篡改后图像可恢复面积.实验结果表明,与已有算法相比,本文算法可以较小的水印嵌入容量实现针对大面积篡改情况下更好
In this paper, a memorability feature based-video hashing is proposed as an alternative to appearance feature based-and visual attention based-algorithms.Inspired by our previous study which shows spa
Document digitalization is one of the basic technologies in multimedia information search and retrieval research area.It has offered a powerful way to bridge the gap between massive redundant image in
粉碎文件图像重构是指将碎纸机产生的碎纸片进行重构,目前最有挑战性的是DARPA的碎纸机挑战竞赛,此竞赛已成为文件重构的评测标准之一.目前的方法仅能给出前两个题目的解决方法.本文针对前三个题目,提出一种利用曲线匹配和信息对齐的方法重构粉碎文件的高效率方法.该方法将文件碎片轮廓分割为若干条曲线,通过曲线距离大小来描述两曲线的相似度,进而选择碎片对进行曲线匹配,同时重点提取了碎片的文本线、颜色、字符和图
基于Ideal-SIS(ideal small integer solution)的一个变种,改进了Damien Stehle的陷门生成算法.利用改进的陷门生成算法,对Damien Stehle的选择明文攻击(CPA)安全方案进行了改进,然后结合SWlFFT压缩函数,提出了一种在标准模型下达到适应性选择密文攻击下的不可区分性(IND-CCA2)安全的混合加密方案.在这陷门产生过程中节省了nlogq
位置隐私保护是位置服务中的关键安全问题.当前的位置发布隐私保护方法忽略了位置点之间的时空关联特性对敏感轨迹信息泄露的重要影响.本文提出了一种基于敏感轨迹点还原的风险模型,利用真实的样本轨迹数据,构造基于马尔科夫链的轨迹隐私威胁模型,基于时空上下文信息计算敏感轨迹点被还原的条件概率,还原由删除或抑制而造成的不完整轨迹数据.基于真实的轨迹数据完成实验,验证了该算法在轨迹数据还原准确性方面达到了80％以
本文提出了一种基于分层匹配的高效、精确的网络近重复图像去冗方法.该方法采用分层匹配的近重复图像检测框架,结合粗略过滤的全局特征匹配和精细的局部特征匹配,并分别在两个阶段构建哈希索引结构以提升匹配效率;在冗余图像去除阶段,提出一种新的图像质量评价策略,对图像的内容和属性进行综合评价,以更加准确地选择出近重复图像组中的质心图像并删除非质心图像.实验结果表明,该方法在保证匹配精度的同时大大提升了匹配效率
We proposed a commutative watermarking-enceryption method for the purpose of protecting privacy and copyrights of digital image monitoring system.The method fuses watermarking and encryption based on
With the development of the Internet and mobile terminals, more and more pirate digital music resources are easily disseminated online.To prevent the illegal copies propagating among mobile devices an
云计算环境下国内外针对DAS模式密文查询的研究缺少攻击模式下对隐私的深度分析.针对该问题,本文提出了一种DAS模式下基于隐私保护的桶划分算法.首先根据查询效率指标提出了一种基于遗传算法的桶划分方案;在此基础之上,针对查询的过程中隐私泄漏情况提出了信息泄露的隐私指标体系,并将该指标体系与查询效率进行结合,最后基于遗传算法的桶划分算法对隐私与效率的模型进行最优化,从而获得最优的桶划分方案来确保查询过程