用于图像认证的半脆弱性数字水印研究

重点分析了半脆弱数字水印按空间域和变换域分类的各种算法，比较了它们的优缺点，总结了半脆弱水印有可能受到的攻击行为。最后提出了半脆弱数字水印技术研究需要解决的问题和发展方向。     

Fragile watermarking scheme for image authentication

A simple fragile watermarking scheme for image authentication is proposed. The insertion of a binary watermark is accomplished by replacing the least significant bit of the host image with a random binary image obtained by random permutations and suitable XOR operations among bitplanes. The proposed scheme is secure, fast, and capable of detecting and localising modification, and is immune of the well-known quantisation attack.     

一种基于量化的图像水印方案

本文提出一种图像自适应的扩频水印方案，利用量化-替换过程将水印嵌到图像DCT域的中高频系数上，为保证水印的鲁棒性和透明性，根据视觉模型自适应地确定水印的量化区间，在水印检测时不需要参考原图。对水印图像的攻击实验表明：算法对噪声干扰及常见的图像处理技术具有较好的鲁棒性。     

图像内容真实性认证的半易损水印方案研究

提出一种基于半易损的图像内容真实性认证方案。该算法既对不改变图像内容的一般性操作具有较好的顽健性，又能对改变图像内容的篡改操作具有较强的敏感性。在水印图像进行一般性处理后能够借助水印差别国和滑动窗检测并定位被篡改的区域。本文还从概率论的角度给出了精确的判决门限确定公式。仿真试验采用一般性操作和篡改操作相结合的方法，实验结果表明本文算法在保护数字图像内容真实性方面是可行的。     

基于自嵌入鲁棒双水印的图像认证算法

通过构造鲁棒零水印并鲁棒的嵌入到图像自身来实现完全盲检测的鲁棒水印方案,从而完成数字图像的全盲版权认证。首先利用整数小波变换构造鲁棒零水印:将原始图像进行三级整数小波变换,并将变换后获取的第三级低频子带(LL3)系数二值化来构造鲁棒零水印。然后利用一种基于奇异值分解(SVD,singular value decomposition)的鲁棒水印方案,将构造的鲁棒零水印嵌入到图像小波变换域低频子带(LL1)的分块奇异值分解的U矩阵的系数中。利用超混沌系统设计了随机置乱和加密方案。在鲁棒嵌入之前,先将二值零水印进行置乱分散并加密,增强了方案的鲁棒性和构造水印安的全性。实验结果证明了所提出的方案的有效性。     

一种半脆弱性数字图像水印算法

本文基于JPEG有损压缩提出了一种半脆弱性数字图像水印算法。实验测试表明,所提出的算法不会明显影响图像的主客观质量,对JPEG有损压缩具有一定的顽健性,可以有效地进行篡改检测,达到检测和标示图像内容真实性的目的。     

Novel watermarking scheme for image authentication using vector quantization approach

In this paper a new invisible watermarking algorithm using vector quantization (VQ) approach for image content authentication application is proposed. At the first stage, a robust verification watermark is embedded using properties of indices of vector quantized image using key based approach. At the second stage, the semi-fragile watermark is embedded by using modified index key based (MIKB) method. Robust watermark andVQenhance the security of the system by providing its double protection. The semi-fragile watermark is utilized for the authentication of the received image. Blind extraction of the watermark is performed independently in two successive steps i.e. in the order of embedding. Further, in oder to classify quantitatively an incidental or intentional attack, which at present is not being followed, a quantitative threshold approach using pixel neighborhood clustering is suggested. The existing methods employ the qualitative approach of identifying incidental or intentional attacks. Imperceptibility of watermarked image is 41 dB on the average, and it is also possible to detect and locate tamper with very high sensitivity. The present scheme is compared with the existing algorithms. The performance of proposed algorithm has been tested on various practical images. It outperforms in distinguishing malicious tampering from content preserving changes and the tampered regions are accurately localized.     

Semi-fragile watermarking for grayscale image authentication and tamper detection based on an adjusted expanded-bit multiscale quantization-based technique

In this work, a semi-fragile watermarking scheme, for grayscale image authentication and tamper detection, is proposed. The proposed watermarking scheme is based on implementing a modified DWT quantization-based algorithm by embedding a random watermark bit sequence into the DWT domain using an expanded-bit multiscale quantization-based technique with adjusted watermarked location. Here, the watermark bit is expanded into three similar bits and embedded in a multiscale fashion into the DWT low-frequency subbands of the 2nd DWT levels (LL2, LL_HL1 and LL_LH1). An adjustment of the quantized coefficients is provided based on modifying their values to fall in more secure locations within the quantization interval. Several designed criteria were used to judge the received image by classifying it into: authenticated, incidentally or maliciously attacked with high accuracy in detecting and classifying attacks. Experimental results have shown the suitability of the proposed approach for tamper detection and accurate authentication.     

一种用于信息注释和鉴别的二值图像水印算法

针对二值图像应用,提出了一种适用于信息注释和鉴别的分块二值图像水印新算法.算法的计算量较小而嵌入容量较大,通过对可用分块的像素模式划分优先级,提供了水印嵌入容量和图像质量间平衡的手段.水印提取不需原图像,但需提供嵌入位置的信息.     

Hierarchical watermarking for secure image authentication withlocalization

Several fragile watermarking schemes presented in the literature are either vulnerable to vector quantization (VQ) counterfeiting attacks or sacrifice localization accuracy to improve security. Using a hierarchical structure, we propose a method that thwarts the VQ attack while sustaining the superior localization properties of blockwise independent watermarking methods. In particular, we propose dividing the image into blocks in a multilevel hierarchy and calculating block signatures in this hierarchy. While signatures of small blocks on the lowest level of the hierarchy ensure superior accuracy of tamper localization, higher level block signatures provide increasing resistance to VQ attacks. At the top level, a signature calculated using the whole image completely thwarts the counterfeiting attack. Moreover, "sliding window" searches through the hierarchy enable the verification of untampered regions after an image has been cropped. We provide experimental results to demonstrate the effectiveness of our method.     

Multipurpose watermarking for image authentication and protection

We propose a novel multipurpose watermarking scheme, in which robust and fragile watermarks are simultaneously embedded, for copyright protection and content authentication. By quantizing a host image's wavelet coefficients as masking threshold units (MTUs), two complementary watermarks are embedded using cocktail watermarking and they can be blindly extracted without access to the host image. For the purpose of image protection, the new scheme guarantees that, no matter what kind of attack is encountered, at least one watermark can survive well. On the other hand, for the purpose of image authentication, our approach can locate the part of the image that has been tampered with and tolerate some incidental processes that have been executed. Experimental results show that the performance of our multipurpose watermarking scheme is indeed superb in terms of robustness and fragility.     

基于多特征的MPEG视频认证方案

为更加准确地反映MPEG视频的内容,构造了一种将视频时间信息、运动特征信息以及DCT系数块能量关系信息等多种特征融合在一起的多特征数字签名算法,并以不增加MPEG编码复杂度的水印嵌入方式将数字签名嵌入到视频中,并实现了基于多特征的视频认证方案.认证实验结果证明,本方案对于如代码转换、重新压缩等常规处理较之单特征认证方法虚警概率要低,而对于如移除、替换等恶意攻击能做出有效判断,并能对篡改区域进行准确定位.     

A new semi-fragile image watermarking with robust tampering restoration using irregular sampling

This paper presents a semi-fragile watermarking method for the automatic authentication and restoration of the content of digital images. Semi-fragile watermarks are embedded into the original image, which reflect local malicious tampering on the image. When tampered blocks are detected, the restoration problem is formulated as an irregular sampling problem. These blocks are then reconstructed, making use of the information embedded in the same watermarked image, through iterative projections onto convex sets. In contrast to previous methods, the restoration process is robust to common image processing operations such as lossy transcoding and image filtering. Simulation results showed that the scheme keeps the probability of false alarm to a minimum while maintaining the data integrity of the restored images.     

A feature-based digital watermarking scheme for halftone image

It is a challenging work to design a robust halftone image watermarking scheme against desynchronization attacks. In this paper, we propose a feature-based digital watermarking method for halftone images with low computational complexity, good visual quality and reasonable resistance toward desynchronization attacks. Firstly, the feature points are extracted from host halftone image by using multi-scale Harris–Laplace detector, and the local feature regions (LFRs) are constructed according to the feature scale theory. Secondly, discrete Fourier transform (DFT) is performed on the LFRs, and the embedding positions (DFT coefficients) are selected adaptively according to the magnitude spectrum information. Finally, the digital watermark is embedded into the LFRs by quantizing the magnitudes of the selected DFT coefficients. By binding the watermark with the geometrically invariant halftone image features, the watermark detection can be done without synchronization error. Simulation results show that the proposed scheme is invisible and robust against common signals processing such as median filtering, sharpening, noise adding, and JPEG compression, etc., and desynchronization attacks such as rotation, scaling, translation (RST), cropping, local random bend, and print-scan, etc.     

量化水印算法分析

对量化水印算法稳健性与抗攻击能力问题,给出了公有水印算法稳健性充分必要条件和抗攻击能量上限,弥补了抗攻击能力评估上没有精确界限的缺陷。分析了载体系数的改变幅度及固定保真度优化中附加模板的取值范围,部分解决了保真度优化中出现的保护水印问题。指出了量化信息编码与一般变量信息承载之间的区别,明确了在信息容量计算中的差别原因。构造了优化视觉效果的嵌入方法和与其等价的其它10个嵌入方法。在同等的载体改变范围内,提出了比基于mod运算的水印嵌入方法抗攻击能力高一倍的量化水印算法,弥补了基于mod运算的水印嵌入方法的抗攻击能力比较低的缺陷。构造了嵌入灰度值的水印算法,避免了量化嵌入算法中把灰度值转换为二进制嵌入后所带来的缺陷。实验验证了此算法及理论分析结果。     

用于图像篡改定位和恢复的分层半脆弱数字水印算法

设计并实现了一种基于分层的半脆弱数字水印算法,用于图像内容完整性认证,篡改定位和恢复。在嵌入方,将水印信息嵌入到原始图像所有像素的2bit最低有效比特,其中认证水印由奇偶校验码和块与子块之间灰度平均值的比较组成;恢复水印为环面自同构映射块的平均灰度值。在检测方,采用分层思想,分3层对水印进行篡改定位和恢复。实验结果表明,该算法对局部发生的篡改可以有效地检测,精确定位并进行内容恢复,同时能够有效的抵抗矢量量化攻击。     

Digital fragile watermarking scheme for authentication of JPEG images

It is a common practice in transform-domain fragile watermarking schemes for authentication purposes to watermark some selected transform coefficients so as to minimise embedding distortion. The author points out that leaving most of the coefficients unmarked results in a wide-open security gap for attacks to be mounted on them. A fragile watermarking scheme is proposed to implicitly watermark all the coefficients by registering the zero-valued coefficients with a key-generated binary sequence to create the watermark and involving the unwatermarkable coefficients during the embedding process of the embeddable ones. Non-deterministic dependence is established by involving some of the unwatermarkable coefficients selected according to the watermark from a nine-neighbourhood system in order to thwart different attacks, such as cover-up, vector quantisation and transplantation. No hashing is needed in establishing the non-deterministic dependence.     

A feature-based robust digital image watermarking scheme

A robust digital image watermarking scheme that combines image feature extraction and image normalization is proposed. The goal is to resist both geometric distortion and signal processing attacks. We adopt a feature extraction method called Mexican hat wavelet scale interaction. The extracted feature points can survive a variety of attacks and be used as reference points for both watermark embedding and detection. The normalized image of an image (object) is nearly invariant with respect to rotations. As a result, the watermark detection task can be much simplified when it is applied to the normalized image. However, because image normalization is sensitive to image local variation, we apply image normalization to nonoverlapped image disks separately. The disks are centered at the extracted feature points. Several copies of a 16-bit watermark sequence are embedded in the original image to improve the robustness of watermarks. Simulation results show that our scheme can survive low-quality JPEG compression, color reduction, sharpening, Gaussian filtering, median filtering, row or column removal, shearing, rotation, local warping, cropping, and linear geometric transformations.     

A novel watermarking scheme for H.264/AVC video authentication

As the H.264/AVC-based video products become more and more popular, issues of copyright protection and authentication that are appropriate for this standard will be very important. In this paper, a content-based authentication watermarking scheme for H.264/AVC video is proposed. Considering the new feature of H.264/AVC, the content-based authentication code for spatial tampering is firstly generated using the reliable features extracted from video frame blocks. The authentication code, which can detect malicious manipulations but allow recompression, is embedded into the DCT coefficients in diagonal positions using a novel modulation method. Spatial tampering can be located by comparing the extracted and the original feature-based watermarks. In addition, combining ECC and interleaving coding, the frame index of each video frame is used as watermark information and embedded in the residual coefficients. Temporal tampering can be detected by the mismatch between the extracted and the observed frame index. Experimental results show that the proposed scheme can discriminate the malicious tampering from the mild signal processing. The tampered location can also be approximately determined according to the glide window and the predefined threshold.     

Secure reversible visible image watermarking with authentication

This paper proposes a secure reversible visible watermarking approach. The proposed pixel mapping function superposes a binary watermark image on a host image to create an intermediate visible watermarked image. Meanwhile, an almost inverse function generates the recovery data for restoring the original pixels. To prevent unauthorized users from approximating the original pixels in the watermarked region, this method adds an integer sequence in the intermediate watermarked image. The sequence is composed of integers generated by two random variables having normal distributions with zero means and distinct variances. The variances facilitate a trade-off between the watermark transparency and the noise generated by unauthorized users. The proposed method also uses Lagrange multipliers to find the optimized variances for the trade-off. Finally, this method uses reversible data embedding to embed the recovery data and hash value for reversibility and authentication, respectively. Experimental results show the watermark visibility for test images along with the watermark transparency for different variances. Using the optimized variances, the watermarked image is at the balance between the watermark transparency and the unauthorized-user-generating noise.