A differential evolution based algorithm for breaking the visual steganalytic system

Image steganography is the process of sending messages secretly by hiding the message in image content. Steganalytic techniques are used to detect whether an image contains a hidden message by analyzing various image features between stego-images (the images containing hidden messages) and cover-images (the images containing no hidden messages). In the past, genetic algorithm (GA) was applied to design a robust steganographic system that breaks the steganalytic systems. However, GA consumes too much time to converge to the optimal solution. In this paper, we use a different evolutionary approach, named differential evolution (DE), to increase the performance of the steganographic system. The key element that DE is distinguished from other population based approaches is differential mutation, which aims to find the global optimum of a multidimensional, multimodal function. Experimental results show that the application of the DE based steganography not only improves the peak signal to noise ratio (PSNR) of the stego-image, but also promotes the normalized correlation (NC) of the extracted secret message for the same number of iterations. It is observed that the percentage increase in PSNR values ranges from 5% to 13% and that of NC values ranges from 0.8% to 3%.     

Listega: list-based steganography methodology

The use of textual list of items, e.g., products, subjects, books, etc., is widely popular and linguistically legible. This motivates the development of List-Based Steganography Methodology (Listega). Listega takes advantage of such use of textual list to camouflage data by exploiting itemized data to conceal messages. Simply, it encodes a message then assigns it to legitimate items in order to generate a text-cover in a form of list. The generated list of items, the text-cover, can be embedded among other legitimate noncoded items for more protection based on a predetermined protocol among communicating parties such as read every other item, every fifth item, or any other way than the use of particular sequence. Listega neither hides data in a noise (errors) nor produces noise. Instead, it camouflages data by manipulating noiseless list of legitimate items. Listega establishes a covert channel among communicating parties by employing justifiably reasons based on the common practice of using textual list of items in order to achieve unsuspicious transmission of generated covers. The presented implementation, validation, and steganalysis of Listega demonstrate: the robustness capabilities of achieving the steganographic goal, the adequate room for concealing data, and the superior bitrate of roughly 1.32 up to 3.87% than contemporary linguistic steganography approaches.     

A high capacity reversible data hiding scheme with edge prediction and difference expansion

To enhance the embedding capacity of a reversible data hiding system, in this paper, a novel multiple-base lossless scheme based on JPEG-LS pixel value prediction and reversible difference expansion will be presented. The proposed scheme employs a pixel value prediction mechanism to decrease the distortion caused by the hiding of the secret data. In general, the prediction error value tends to be much smaller in smooth areas than in edge areas, and more secret data embedded in smooth areas still meets better stego-image quality. The multiple-base notational system, on the other hand, is applied to increase the payload of the image. With the system, the payload of each pixel, determined by the complexity of its neighboring pixels, can be very different. In addition, the cover image processed by the proposed scheme can be fully recovered without any distortion. Experimental results, as shown in this paper, have demonstrated that the proposed method is capable of hiding more secret data while keeping the stego-image quality degradation imperceptible.     

Search-order coding method with indicator-elimination property

Vector quantization (VQ) is a widely used technique for many applications especially for lossy image compression. Since VQ significantly reduces the size of a digital image, it can save the costs of storage space and image delivery. Search-order coding (SOC) was proposed for improving the performance of VQ in terms of compression rate. However, SOC requires extra data (i.e. indicators) to indicate source of codewords so the compression rate may be affected. To overcome such a drawback, in this paper, a search-order coding with the indicator-elimination property was proposed by using a technique of reversible data hiding. The proposed method is the first one using such a concept of data hiding to achieve a better compression rate of SOC. From experimental results, the performance of the SOC method can be successfully improved by the proposed indicator eliminated search-order coding method in terms of compression rate. In addition, compared with other relevant schemes, the proposed method is also more flexible than some existing schemes.     

结合失真函数的稀疏域图像隐写算法

稀疏域图像隐写利用稀疏表示在载体图像中嵌入秘密信息。提出设计稀疏域图像隐写的失真函数,通过选择使失真函数最小的修改方式进行秘密嵌入,从而减小秘密信息嵌入对载体的影响。实验结果表明,该算法可以在保持图像视觉效果的同时,更有效地抵抗现有的隐写分析算法。     

特征保持隐写的博弈分析

为了分析基于统计特征保持的隐写算法的安全性,将隐写对抗分为以特征子集作为策略和以隐写、隐写分析算法作为策略两种情况建立了两种隐写博弈模型。模型将隐写分析方的检测率作为支付函数,用隐写对抗双方特征子集间的差异反映算法抗统计分析的能力。通过对模型进行均衡分析,给出了各种情况下基于统计特征保持的方式提高隐写系统安全性的最优策略,并得到了均衡局势下的期望支付。     

基于DCT域相关性的非对称JPEG隐写

目前用于JPEG隐写的失真代价函数对同一DCT系数的加一与减一操作分配相同的代价值。但由于JPEG图像中DCT域的相关性,加一与减一修改对图像内容的影响不同,因此其对应的代价值也理应不同。文中基于DCT域相关性,提出了一种适用于JPEG隐写的通用代价值优化方法,主要考虑JPEG图像中相邻DCT块相同位置上系数的相关性。利用八邻域块中相同位置的DCT系数求平均的方法给出当前DCT系数的预测值。对现有的JPEG失真代价函数,按照向预测值靠拢的原则区分加一和减一的代价值。经过调整的代价值能引导隐写修改后的DCT系数最大程度地向预测值靠拢,增强DCT域相关性,从而提高隐写的安全性。该方法可以与任何现有的JPEG隐写失真代价函数结合使用。实验表明,所提方法几乎不增加原始算法的时间复杂度,同时能有效提高现有JPEG隐写方法的安全性,可使当前隐写分析工具的测试误差平均提升2.4%。     

An efficient image interpolation increasing payload in reversible data hiding

Image interpolation is a very important branch in image processing. It is widely used in imaging world, for example, image interpolation is often used in 3-D medical image to compensate for information insufficiency during image reconstruction by simulating additional images between two-dimensional images. Reversible data hiding has become significant branch in information hiding field. Reversibility allows the original media to be completely restored without any degradation after the embedded messages have been extracted. This study proposes a high-capacity image hiding scheme by exploiting an interpolating method called Interpolation by Neighboring Pixels (INP) on Maximum Difference Values to improve the performance of data hiding scheme proposed by Jung and Yoo. The proposed scheme offers the benefits of high embedding capacity with low computational complexity and good image quality. The experimental results showed that the proposed scheme has good performance for payload up to 2.28 bpp. Moreover, the INP yields higher PSNRs than other interpolating methods such as NMI, NNI and BI.     

基于K-L差异的隐密术安全性理论研究

隐密术安全性理论研究一直是被广泛关注的研究重点和难点。该文通过对K-L差异的回顾及隐密术安全性的定性分析,指出Cachin给出的安全性定义缺乏一般性。同时注意到除概率分布的差异外,隐密信息的样本量和状态集合的势对隐密信息的安全性也有重要影响。结合对隐密信息安全性的定性分析,给出了一个基于K-L差异及假设检验的隐密术安全性定义,并推导出修改率对于隐密术安全性的影响。最后结合现有的隐密算法和隐密分析算法设计了一组对比试验,以验证该文理论分析的有效性。     

一种新的信息隐藏检测方法的探讨

首先建立了隐蔽通信系统模型,再对该模型进行了安全分析,最后利用信号盲分离理论中的独立组件分析等方法对隐藏对象与掩蔽载体进行了分离,并给出了隐藏信息的估计算法。