基于量化表不一致性的JPEG图像篡改盲检测

利用篡改后JPEG图像量化表不一致的特性,提出一种针对JPEG图像的篡改盲检测新方法。通过智能选取若干图像块,迭代估计出待测图像的原始量化表,大致定位出篡改区域。然后用估计出的原始量化表对篡改区域再进行一次JPEG压缩,由压缩前后图像像素值的差值最终确定篡改位置。实验结果表明,提出的估计量化表算法复杂度小,精度高。检测算法能有效地检测出多种篡改类型的JPEG图像,且对篡改区域和未篡改区域压缩因子相差较小的JPEG合成类篡改,检测正确率更高。     

基于量化表估计的JPEG篡改图像盲检测

传统JPEG篡改图像盲检测算法采用不同质量因子对待测图像进行尝试性再压缩.为了缩短检测时间,减少计算量,提出一种基于量化表估计的JPEG篡改图像盲检测有效算法.首先通过DCT系数直方图估计出篡改图像某一部分量化表,然后通过计算小块DCT系数中分别是对应量化步长整数倍的个数进一步判定该小块是否属于篡改区域.实验结果表明,提出的量化步长估计算法复杂度小,正确率高;检测算法能准确检测出copy-move和合成类图像篡改.此外,改进方法不仅对标准量化表压缩的JPEG篡改图像具有非常好的检测结果,而且对非标准量化表压缩的JPEG篡改图像检测也有效.     

基于量化噪声的JPEG图像篡改检测

JPEG图像篡改引入的双重压缩会导致篡改区域的原始压缩特性发生改变,因此可以利用篡改区域压缩特性的不一致性来检测图像的篡改。利用该原理,提出了一种基于量化噪声的JPEG图像篡改检测算法。算法对待检测图像进行分块,计算每块的量化噪声,求取图像块的量化噪声服从均匀分布和高斯分布的概率,从而检测出篡改过的双重压缩区域。实验结果表明：该算法能有效检测双重压缩的JPEG图像篡改,并能定位出篡改区域。     

基于伪Zernike矩和Contourlet变换的抗几何攻击图像水印算法

在研究Contourlet变换和伪Zernike矩理论的基础上,提出了一种新颖的抗几何攻击数字图像水印算法。该算法首先 用Contourlet变换处理 原始图像,提取出图像的低频区域;然后依据人眼的视觉特性以及嵌入水印前、后系数的相关性,采取量化调制伪Zernike矩的幅值将数字水印信息嵌入图像的低频区域中。其无须借助于原始图像,就能进行水印提取,实现真正的盲检测。实验结果表明,该算法对平移变换的抵抗力最强,对旋转和缩放也有很好的抵抗力,而且经过JPEG压缩之后,数字水印也不会失真。     

基于DSP的JPEG图像压缩编码的实现

针对JPEG压缩编码在各种嵌入式系统中的广泛应用,提出基于通用DSP,采用软件方法实现图像JPEG压缩编码的方法;文中首先给出了JPEG基本顺序过程的算法实现流程,同时重点研究了该方法中FDCT变换的快速算法和Huffman编码的代码实现;最后给出了彩色原始图像由不同的量化表得出的实验结果,并给出实验评价,实验结果表明这种方法有效地减小了图片存储所需空间,提高了系统的实时性和应用价值。     

JPEG图像篡改的盲检测技术

数字图像被动认证技术是一门不依赖数字水印或者签名等图像先验知识而对图像来源和内容真实性进行认证的新兴技术,JPEG图像篡改的盲检测已成为当前被动认证的研究热点。详细分析了三种基于JPEG压缩的盲检测算法：JPEG压缩历史检测和量化表的估计,块效应不一致性的篡改区域检测,JPEG二次压缩的检测,系统阐明了现有算法的基本特征和优缺点,最后展望了未来的研究方向。     

基于量化表的图像篡改检测算法

数字图像盲取证技术是近年来研究的一个热点问题。提出一种利用量化表来进行定位和检测数字图像双重JPEG压缩的篡改检测算法。首先对数字图像进行压缩消除图像本身的噪声,然后利用图像压缩模型来描述图像首次压缩与第二次压缩之间的关系进而估计首次压缩量化表,最后提出一种高效率的方法通过利用量化表来定位图像篡改区域。实验结果表明,该算法能够有效地对双重JPEG压缩的图像进行检测和定位,并具有很好的鲁棒性。     

基于块效应不一致的JPEG图像篡改区域检测

利用JPEG图像经篡改后块效应不一致的特性,提出一种根据JPEG篡改图像块效应值的分布检测篡改区域的方法。改进一种通过参照DCT系数的功率谱直方图估计量化表的方法,根据量化表求得图像每个8×8块的块效应,然后根据块效应值的分布规律检测出图像篡改区域。实验表明算法的应用范围广,具有一定的实践意义。     

一种求JPEG初始量化矩阵的方法*

针对目前双重压缩研究的问题,提出一种确定初始矩阵的新方法（a method of calculating primary quantization matrix of JPEG,简称CM）。在未知双重压缩系数大小关系的条件下,CM方法通过剪切JPEG图像的方式获得与原始图像近似的特性,然后再通过对这些特性进行分析来确定初始量化矩阵。实验结果表明,CM方法能够确定初始量化矩阵,它是一种非常有效的方法。     

基于相关性的JPEG高压缩图像峰值信噪比盲估计

量化噪声是JPEG图像压缩过程中产生的主要失真,在图像质量评估时需要对其大小进行估计。针对峰值信噪比(PSNR)估计方法在实际应用时终端用户无法获取原始图像的限制,基于图像空域相关性,引入JPEG压缩图像的斜率均方差和最大非零离散余弦变换系数的位置作为可测特征量,利用训练图像进行机器学习得到训练模型,将测试图像进行解析得到图像PSNR盲估计结果。实验结果表明,该PSNR估计结果优于目前流行的峰值信噪比盲估计方法。     

基于HVS模型的JPEG2000数字图像水印算法

文章提出了一种基于HVS模型的JPEG2000数字图像水印算法,该算法根据计算原始图像嵌入水印的能力限制和数据量化方法进行水印的嵌入与提取。实验证明,该算法对于常用的图像处理方法以及JPEG2000编码在保持了较好质量的基础上有较强的鲁棒性。     

A multi-objective evolutionary approach to image quality/compression trade-off in JPEG baseline algorithm

The JPEG algorithm is one of the most used tools for compressing images. The main factor affecting the performance of the JPEG compression is the quantization process, which exploits the values contained in two tables, called quantization tables. The compression ratio and the quality of the decoded images are determined by these values. Thus, the correct choice of the quantization tables is crucial to the performance of the JPEG algorithm. In this paper, a two-objective evolutionary algorithm is applied to generate a family of optimal quantization tables which produce different trade-offs between image compression and quality. Compression is measured in terms of difference in percentage between the sizes of the original and compressed images, whereas quality is computed as mean squared error between the reconstructed and the original images. We discuss the application of the proposed approach to well-known benchmark images and show how the quantization tables determined by our method improve the performance of the JPEG algorithm with respect to the default tables suggested in Annex K of the JPEG standard.     

基于奇偶量化的Contourlet变换域指纹图像水印算法

提出了一种基于奇偶量化的Contourlet变换域指纹图像水印算法。原始指纹图像经过Contourlet变换后，分解为一系列多尺度、局部化、方向性的子带图像，选择在低频子带中嵌入经过二维Arnold置乱加密后的水印。嵌入水印时，采用奇偶量化算法修改低频子带的系数。水印提取时，不需要原始图像，实现了盲提取。实验结果表明，提出的指纹图像水印算法能够较好地抵抗JPEG有损压缩、叠加噪声、剪裁等攻击，具有较好的不可见性和鲁棒性，提高了指纹识别的可靠性。     

基于DWT的彩色图像鲁棒盲水印新算法

本文提出了一种基于小波域低频系数量化的彩色图像鲁棒盲水印新算法.首先将RGB彩色宿主图像转换为YCbCr颜色空间,然后对亮度分量Y进行三层小波分解,采用适当的量化间隔对低频逼近子带LL3进行均值量化,根据量化系数的奇偶特性再与二值图像置乱转化后的水印序列进行异或运算,从而把水印嵌入到亮度分量的逼近子带中.仿真实验结果表明,本文提出的算法对JPEG/JPEG2000压缩、添加噪声、滤波、剪切、像素移位等多种攻击有较强的鲁棒性,透明性好,并且能实现水印的盲提取.     

基于JPEG双量化效应的图像盲取证

JPEG图像的双量化效应为JPEG图像的篡改检测提供了重要线索.根据JPEG图像被局部篡改后,又被保存为JPEG格式时,未被篡改的区域(背景区域)的离散余弦变换(DCT)系数会经历双重JPEG压缩,篡改区域的DCT系数则只经历了1次JPEG压缩.而JPEG图像在经过离散余弦变换后其DCT域的交流(AC)系数的分布符合一个用合适的参数来描述的拉普拉斯分布,在此基础上提出了一种JPEG图像重压缩概率模型来描述重压缩前后DCT系数统计特性的变化,并依据贝叶斯准则,利用后验概率表示出图像篡改中存在的双重压缩效应块和只经历单次压缩块的特征值.然后设定阈值,通过阈值进行分类判断就可以实现对篡改区域的自动检测和提取.实验结果表明,该方法能快速并准确地实现篡改区域的自动检测和提取,并且在第2次压缩因子小于第1次压缩因子时,检测结果相对于利用JPEG块效应不一致的图像篡改盲检测算法和利用JPEG图像量化表的图像篡改盲检测算法有了明显的提高.     

Forensics for partially double compressed doctored JPEG images

Digital image forensics is required to investigate unethical use of doctored images by recovering the historic information of an image. Most of the cameras compress the image using JPEG standard. When this image is decompressed and recompressed with different quantization matrix, it becomes double compressed. Although in certain cases, e.g. after a cropping attack, the image can be recompressed with the same quantization matrix too. This JPEG double compression becomes an integral part of forgery creation. The detection and analysis of double compression in an image help the investigator to find the authenticity of an image. In this paper, a two-stage technique is proposed to estimate the first quantization matrix or steps from the partial double compressed JPEG images. In the first stage of the proposed approach, the detection of the double compressed region through JPEG ghost technique is extended to the automatic isolation of the doubly compressed part from an image. The second stage analyzes the doubly compressed part to estimate the first quantization matrix or steps. In the latter stage, an optimized filtering scheme is also proposed to cope with the effects of the error. The results of proposed scheme are evaluated by considering partial double compressed images based on the two different datasets. The partial double compressed datasets have not been considered in the previous state-of-the-art approaches. The first stage of the proposed scheme provides an average percentage accuracy of 95.45%. The second stage provides an error less than 1.5% for the first 10 DCT coefficients, hence, outperforming the existing techniques. The experimental results consider the partial double compressed images in which the recompression is done with different quantization matrix.     

No-reference image quality assessment based on localized discrete cosine transform for JPEG compressed images

In this paper, we propose a new no-reference image quality assessment for JPEG compressed images. In contrast to the most existing approaches, the proposed method considers the compression processes for assessing the blocking effects in the JPEG compressed images. These images have blocking artifacts in high compression ratio. The quantization of the discrete cosine transform (DCT) coefficients is the main issue in JPEG algorithm to trade-off between image quality and compression ratio. When the compression ratio increases, DCT coefficients will be further decreased via quantization. The coarse quantization causes blocking effect in the compressed image. We propose to use the DCT coefficient values to score image quality in terms of blocking artifacts. An image may have uniform and non-uniform blocks, which are respectively associated with the low and high frequency information. Once an image is compressed using JPEG, inherent non-uniform blocks may become uniform due to quantization, whilst inherent uniform blocks stay uniform. In the proposed method for assessing the quality of an image, firstly, inherent non-uniform blocks are distinguished from inherent uniform blocks by using the sharpness map. If the DCT coefficients of the inherent non-uniform blocks are not significant, it indicates that the original block was quantized. Hence, the DCT coefficients of the inherent non-uniform blocks are used to assess the image quality. Experimental results on various image databases represent that the proposed blockiness metric is well correlated with the subjective metric and outperforms the existing metrics.

     

基于JPEG 2000实时量化水印和指纹识别的身份认证系统

提出了新的JPEG2000实时量化水印算法,并将其用于改进的基于指纹识别和数字水印的银行养老金发放系统。系统客户端,量化水印在JPEG2000压缩过程中实时嵌入指纹图像,压缩比特流传送到服务端;系统服务端,水印在JPEG2000解压缩过程中实时提取,使用解压缩的指纹图像和水印进行身份认证。实验表明典型指纹图像压缩到1/4~1/20的时候,嵌入的水印能够无损提取,指纹图像虽不能完全恢复但识别率没有明显降低。因而在低网络带宽条件下,新系统有更好的交互性能,在电子商务中有很好的应用前景。