对空域图像LSB隐写术的提取攻击

隐写分析的研究一直集中于检测隐藏信息的存在性,而关于如何提取隐藏信息(即对隐写术的提取攻击)的研究还非常少.对于使用密钥的隐写术,提取攻击等价于恢复隐写密钥.文中结合隐写分析中的检测技术和密码分析中的相关攻击技术,对空域图像LSB隐写术提出了一种隐写密钥恢复方法.理论分析说明:此攻击方法的计算复杂度主要由所需的样本量决定,并且当嵌入率r接近0或1时攻击将失败.作者通过混合高斯模型给出了一个估计最小样本量的方法.针对隐写软件"Hide and Seek 4.1"的实验表明:此攻击方法可以成功恢复隐写密钥,从而提取隐藏的消息.如果消息长度L未知,当嵌入率5.3%＜r＜94.7%时攻击可以成功;如果L已知,当1.1%＜r＜98.4%时攻击可以成功,并且当11%＜r＜50%时,使用估计的最小样本量可以将攻击速度提高10%～45%.     

空域序列隐写图像的密钥估计算法*

在已知嵌入消息长度的条件下,隐写密钥可以看做消息嵌入的起始位。将图像像素划分为不同类点,通过分析信息嵌入和最低位置反对各类点的影响,得到图像嵌入信息部分和未嵌入信息部分的差异,最终设计了针对空域序列LSB(least significant bits)隐写图像的密钥估计算法。实验结果表明,该算法可对隐写密钥进行快速有效的估计。     

基于JPEG图像的秘密信息长度估计研究

图像隐写分析就是对信息隐藏系统进行攻击的技术,基于JPEG图像的隐藏信息长度估计的方法最近引起了很大关注。该文分析和讨论了JSteg、F5和OutGuess隐写分析算法。     

辫群上的扭结共轭搜索问题和密码体制研究

通过分析辫群的相关性质及群上的判定问题,结合扭结共轭问题、子群成员判断问题及根搜索问题,提出一种辫群上的公钥加密协议和签名协议,对两者的安全性进行分析,证明敌手无法从公钥中恢复密钥,因此协议可以抵抗长度攻击、惟密钥攻击、一般选择消息攻击、定向选择消息攻击和适应性选择消息攻击。     

P2P网络中基于身份的消息链接密钥隔离加密

针对P2P网络中存在密钥泄露和大量数据加密传输困难的问题,采用密钥隔离机制和消息分组链接加密的方法,提出一种具有消息链接性的基于身份密钥隔离加密方案。在随机预言机模型下给出形式化证明,该方案在双线性Deffie-Hellman假设成立的条件下,满足适应性选择消息攻击不可区分性,且该方案具有非限定时间周期以及随机存取密钥更新的性质。性能分析结果表明,该方案的密文长度仅为基本密钥隔离加密方案的一半,适用于P2P网络中传输大量的加密数据。     

一种新型基于Binary-LWE的认证密钥交换协议

为了设计一种基于格困难问题的强安全认证密钥交换协议,分析了DXL12和DXL14方案中缺少认证功能导致容易遭受中间人攻击等缺陷,提出一种基于Binary-LWE的认证密钥交换协议。该协议具有两轮消息交互,不依赖于数字签名提供隐式密钥认证,并采用2012年Micciancio和Peikert在欧密会上提出的陷门函数来提供双方认证功能。在随机语言机模型下将安全性直接建立在Binary-LWE问题的困难性假设上,具有前向安全性、抗中间人攻击、抗冒充攻击等安全属性。由于该方案的安全性是基于格上困难问题,所以可以抵抗量子攻击。     

基于统计特性的隐写密钥恢复方法

隐写分析技术是检验信息隐藏技术安全性的关键技术,作为隐写分析的最终目标,隐秘信息的提取在取证隐蔽通信方面有重要作用。现有明文嵌入条件下的密钥恢复方法会将部分伪密钥误判为真,为避免这种情况,论文提出了一种基于明文统计特性的隐写密钥恢复方法。该方法在估计嵌入率的基础上确定编码参数的范围,构造密钥空间;然后利用校验矩阵和载密序列通过STC(Syndrome-Trellis Codes)译码方程求出译码序列;最后,基于明文信息的统计特性对隐写密钥进行识别。实验结果表明,针对自适应隐写算法,该方法能够盲识别出正确的隐写密钥。与现有的盲识别算法相比,该方法提高了隐写密钥识别的准确率。     

抗隐写密钥分析的LSB隐写算法

LSB算法是一种典型的空间域信息隐藏算法,由于嵌入信息量大、算法简单而成为最常见的信息隐藏算法,也同时成为了隐写分析者很感兴趣的对象。针对隐写位置密钥的攻击,本文提出了一种改进的LSB隐写算法,先在图像各像素的LSB位嵌入噪声信息,再在密钥的控制下有选择性地在部分像素点嵌入隐藏信息。在已知载体图像或已知多幅用同一载体进行信息隐藏的载密图像的情况下,可以有效抵御针对隐写位置密钥的攻击,提高了LSB算法的安全性。同时该思想也可应用于基于DCT域等频域的隐藏算法中。     

基于数字化卡登格的生成图像隐写方案

针对图像隐写中载体修改会留有修改痕迹的问题,文章提出一种基于数字化卡登格的图像隐写方案。首先自动生成数字化卡登格,作为信息隐写和提取的密钥;其次以空白图像作为载体,根据数字化卡登格的区域填充秘密信息;在整个隐写过程保持秘密消息不变的情况下,再利用深度生成模型补全受损的图像,生成具有自然语义的含密图像。文章方案不受载体类型的限制,还可以扩展到其他数字媒体。实验结果表明,新的图像隐写方案具有良好的可行性,相比其他隐写方法,文中方案在实际通信中安全性较高。     

F5隐写算法及其隐写分析研究

图像隐写分析是检测、提取和破坏隐写图像中秘密信息的技术,是信息安全领域的研究热点之一.由Westfeld提出的F5隐写算法是一种重要的图像隐写算法.分析了F5隐写算法及其隐写系统,讨论了一种针对F5隐写算法的隐藏检测方法,并给出了具体实现的算法.实验结果表明,F5算法采用矩阵编码,提高了嵌入效率.     

采用SVR模型进行嵌入率估计的隐写分析方法

为解决大多数通用隐写分析算法不能检测秘密信息长度的问题,提出了一种改进的能估计秘密信息长度的通用隐写分析方法。从隐写图中提取描述DCT域系数相关性的132维特征,用支持向量回归机学习图像特征和相应嵌入改变率之间的映射关系并建立模型,根据映射模型估计测试隐写图的嵌入改变率。使用典型的嵌入算法:F5、outguess与MB进行测验,仿真结果显示提出的秘密信息长度估计算法是切实可行的。     

基于特征的彩色位图隐写分析

针对互联网上流行的真彩位图文件的隐写算法工具InplainView, 根据算法嵌入原理和特征分析,成功实现了对隐秘图像的检测;通过研究从隐秘图像中所提取消息的字符频度分布,提出了对嵌入消息加密的密钥恢复解决方案,有效提高了隐写分析性能。最后的实验结果表明,该隐写分析方法简单、高效,便于实用。     

基于多特征支持向量回归的F5定量隐密分析研究

针对现有F5定量隐密分析方法仅考虑直方图收缩导致定量估计误差较大的问题,提出了一种新的基于多特征支持向量回归(SVR)的定量隐密分析算法.算法通过最大似然估计拟合DCT系数分布模型的参数,利用参数的相对变化量来反映嵌入机制对直方图的修改,并使用共生矩阵描述F5算法对图像像素间相关性的破坏.在确定特征与嵌入量之间的函数关系时,算法采用SVR在高维空间中进行拟合,从而估计待测图像的容量.实验表明,该算法可以对F5算法的嵌入量进行有效估计,而且估计效果要优于已有的算法.     

Quantitative steganalysis of digital images: estimating the secret message length

The objective of steganalysis is to detect messages hidden in cover objects, such as digital images. In practice, the steganalyst is frequently interested in more than whether or not a secret message is present. The ultimate goal is to extract and decipher the secret message. However, in the absence of the knowledge of the stego technique and the stego and cipher keys, this task may be extremely time consuming or completely infeasible. Therefore, any additional information, such as the message length or its approximate placement in image features, could prove very valuable to the analyst. In this paper, we present general principles for developing steganalytic methods that can accurately estimate the number of changes to the cover image imposed during embedding. Using those principles, we show how to estimate the secret message length for the most common embedding archetypes, including the F5 and OutGuess algorithms for JPEG, EzStego algorithm with random straddling for palette images, and the classical LSB embedding with random straddling for uncompressed image formats. The paper concludes with an outline of ideas for future research such as estimating the steganographic capacity of embedding algorithms.     

An extended matrix encoding algorithm for steganography of high embedding efficiency

F5 steganography is much different from most of LSB replacement or matching steganographic schemes, because matrix encoding is employed to increase embedding efficiency while reducing the number of necessary changes. By using this scheme, the hidden message inserted into carrier media imperceptibly can be transferred via a more secure subliminal channel. The embedding domain is the quantized DCT coefficients of JPEG image, which makes the scheme be immune to visual attack and statistical attack from the steganalyst. Based on this effective scheme, an extended matrix encoding algorithm is proposed to improve the performance further in this paper. By changing the hash function in matrix encoding and converting the coding mode, the embedding efficiency and embedding rate get increased to a large extent. Eventually, the experimental results demonstrate the extended algorithm is superior to the classic F5 steganography.     

基于双向马尔可夫模型的JPEG图象隐写分析

提出了一种基于双向马尔可夫模型的JPEG图象的通用隐写分析方法,利用量化后分块DCT系数的中低频系数间的相关性,提取DCT块内和块间的特征,采用阈值贝叶斯分类方法进行识别,并且与SVM分类器的识别效果进行了比较。针对4种公认的JPEG嵌入方法——F5、Outguess、MB1和MB2进行隐写分析,在CorelDraw图象库上做实验,取得了很好的性能。     

A novel adaptive steganography based on local complexity and human vision sensitivity

This paper presents a novel adaptive steganographic scheme that is capable of both preventing visual degradation and providing a large embedding capacity. The embedding capacity of each pixel is dynamically determined by the local complexity of the cover image, allowing us to maintain good visual quality as well as embedding a large amount of secret messages. We classify pixels into three levels based on the variance of the local complexity of the cover image. When determining which level of local complexity a pixel should belong to, we take human vision sensitivity into consideration. This ensures that the visual artifacts appeared in the stego image are imperceptible, and the difference between the original and stego image is indistinguishable by the human visual system. The pixel classification assures that the embedding capacity offered by a cover image is bounded by the embedding capacity imposed on three levels that are distinguished by two boundary thresholds values. This allows us to derive a combination ratio of the maximal embedding capacity encountered with at each level. Consequently, our scheme is capable of determining two threshold values according to the desired demand of the embedding capacity requested by the user. Experimental results demonstrated that our adaptive steganographic algorithm produces insignificant visual distortion due to the hidden message. It provides high embedding capacity superior to that offered by a number of existing schemes. Our algorithm can resist the RS steganalysis attack, and it is statistically invisible for the attack of histogram comparison. The proposed scheme is simple, efficient and feasible for adaptive steganographic applications.     

Exploring pixel-value differencing and base decomposition for low distortion data embedding

This paper presents a low distortion data embedding method using pixel-value differencing and base decomposition schemes. The pixel-value differencing scheme offers the advantage of conveying a large amount of payload, while still maintaining the consistency of an image characteristic after data embedding. We introduce the base decomposition scheme, which defines a base pair for each degree in order to construct a two-base notational system. This scheme provides the advantage of significantly reducing pixel variation encountered due to secret data embedding. We analyze the pixel variation and the expected mean square error caused by concealing with secret messages. The mathematical analysis shows that our scheme produces much smaller maximal pixel variations and expected mean square error while producing a higher PSNR. We evaluate the performance of our method using 6 categories of metrics which allow us to compare with seven other state-of-the-art algorithms. Experimental statistics verify that our algorithm outperforms existing counterparts in terms of lower image distortion and higher image quality. Finally, our scheme can survive from the RS steganalysis attack and the steganalytic histogram attack of pixel-value difference. We conclude that our proposed method is capable of embedding large amounts of a message, yet still produces the embedded image with very low distortion. To the best of our knowledge, in comparison with the current seven state-of-the-art data embedding algorithms, our scheme produces the lowest image distortion while embedding the same or slightly larger quantities of messages.     

Color local complexity estimation based steganographic (CLCES) method

In this paper we present the color local complexity estimation based steganographic (CLCES) method that is able of both preventing visual degradation and providing a large embedding capacity. A preprocessing stage is applied in the proposed scheme to improve the steganography security. The embedding capacity of each pixel is determined by the local complexity of the cover image, allowing good visual quality as well as embedding a large amount of secret messages. We classify the pixels using a threshold based on the standard deviation of the local complexity in the cover image to provide a compromise between the embedding capacity and the image visual quality. The experimental results demonstrated that the algorithm CLCES proposed produces insignificant visual distortion due to the hidden message. It provides a high embedding capacity that is superior respect to the offered by the existing schemes. The proposed method is a secure steganographic algorithm; it can resist the image quality measures (IQM) steganalysis attack. The RGB, YCbCr, and HSV color spaces are incorporated in the proposed scheme to ensure that the difference between the cover image and the stego-image which is indistinguishable by the human visual system (HVS). Finally, the proposed scheme is simple, efficient, and feasible for the adaptive steganographic applications.