Identifying Shifted Double JPEG Compression Artifacts for Non-intrusive Digital Image Forensics

A337 improved forgery detection and tampering localization method is proposed for double JPEG compression images based on first digit feature of discrete cosine transform (DCT) coefficients and kernel-based nonlinear representor (KNR). First, a to-be-checked JPEG image is divided into overlapping blocks of size $64\times 64$ , and the first digit (1∼9) features of alternating current (AC) DCT coefficients at the first nine positions of every $8\times 8$ blocks are obtained in each image block, followed by principal component analysis (PCA) transform to form compact features. And then the KNR classifier is used to judge whether the corresponding image block has been re-compressed. Finally, the test results of double compression are used to locate the tampered area of JPEG image. Experimental results show that in comparison with representative algorithms, the improved algorithm achieves better results, and is robust to operations such as rotation, resizing and feathering. Moreover, KNR classifier outperforms classical SVM classifier in recognition effect and efficiency.

This paper proposes a JPEG anti-forensic method, which aims at removing from a given image the footprints left by JPEG compression, in both the spatial domain and DCT domain. With reasonable loss of image quality, the proposed method can defeat existing forensic detectors that attempt to identify traces of the image JPEG compression history or JPEG anti-forensic processing. In our framework, first because of a total variation-based deblocking operation, the partly recovered DCT information is thereafter used to build an adaptive local dithering signal model, which is able to bring the DCT histogram of the processed image close to that of the original one. Then, a perceptual DCT histogram smoothing is carried out by solving a simplified assignment problem, where the cost function is established as the total perceptual quality loss due to the DCT coefficient modification. The second-round deblocking and de-calibration operations successfully bring the image statistics that are used by the JPEG forensic detectors to the normal status. Experimental results show that the proposed method outperforms the state-of-the-art methods in a better tradeoff between the JPEG forensic undetectability and the visual quality of processed images. Moreover, the application of the proposed anti-forensic method in disguising double JPEG compression artifacts is proven to be feasible by experiments.

From the digital forensics point of view, image forgery is considered as evidence that could provide a major breakthrough in the investigation process. Additionally, the development of storage device technologies has increased storage space significantly. Thus a digital investigator can be overwhelmed by the amount of data on storage devices that needs to be analysed. In this paper, we propose a model for classifying bulk JPEG images produced by the data carving process or other means into three different classes to solve the problem of identifying forgery quickly and effectively. The first class is JPEG images that contain errors or corrupted data, the second class is JPEG images that contain forged regions, and the third is JPEG images that have no signs of corruption or forgery. To test the proposed model, some experiments were conducted on our own dataset in addition to CASIA V2 image forgery dataset. The experiments covered different types of forgery technique. The results yielded around 88% accuracy rate in the classification process using five different machine learning methods on CASIA V2 dataset. It can be concluded that the proposed model can help investigators to automatically classify JPEG images, which reduce the time needed in the overall digital investigation process.

In recent years, with the development of the information society, digital images can be easily uploaded and downloaded. Against this background, instances of plagiarism and tampering with images have been increasing. Therefore, it is necessary to develop robust watermarking for the copyright protection of digital images. In this study, we propose a blind digital image watermarking method based on dual-tree complex discrete wavelet transform and quantization. We apply this wavelet transform to the original image and embed two watermarked images of different sizes using quantization index modulation and relative modulation. Quantization index modulation is robust against geometric attacks. Relative modulation is robust against noise addition. The results of our experiments show that our method is robust against Gaussian noise, salt and pepper noise, cropping, and Jpeg compression. The experimental results show that the watermark can be extracted by Gaussian noise, salt and pepper noise, cropping, Jpeg compression. In addition, we visually check the extracted watermarked images. Furthermore, we compare three methods: one using dual-tree complex discrete wavelet transform and sliding window. The second method is based on discrete wavelet transform, discrete cosine transform, and singular value decomposition. The third method is based on Discrete Cosine Transform and RM, QIM. By comparing these three methods with the existing methods, we confirmed the significance and performance of our method.

There is an increasing number of crime cases involving computers and multimedia files such as child pornographic photos, recorded voices and CCTV videos. It is not uncommon that suspects can manage to erase the files or even destroy the whole file system before seized by the law enforcement officer. Reconstructing fragmented files is still a challenge in digital forensics for both damaged and integrity files, especially without the file system information.
\n
\nFirstly, this work focuses on JPEG files, one of the most popular photograph formats, and proposes techniques for recovering partially-damaged standalone JPEG fragments by reconstructing pseudo headers. The techniques deal with missing Huffman tables and sub-sampling factors, estimate the resolution of JPEG fragments, assess the image quality of JPEG files with incorrect quantization tables, and create quantization tables that are very close to the correct quantization tables in a reasonable amount of time. Extensive experiments with real camera pictures demonstrate that the techniques can recover standalone fragments accurately and efficiently.
\n
\nSecondly, this study focuses on JPEG file carving for, in particular, heavily fragmented JPEG files when file system information damaged. Due to the fact that, it is a critical challenge to identify encrypted or compressed unknown data streams without sufficient corresponding decoding information. To begin with, we propose a technology to display and identify the JPEG file fragmentation piece, which is a partial content of a file and consist of one or more physically consecutive data blocks, from unknown data streams. A new similarity matching metric (CED) is proposed to evaluate the difference between data blocks in large scale. By evaluating the pattern of similarity distribution in fragmented file, we can identify the occurrence of fragmentation point in advance. Then, a fragmentation piece-based _le carving methodology is presented to recover heavily fragmented JPEG files. By comparing with state-of-the-art photo recovery technical, the proposed techniques can automatically and successfully recover most of fragmented files from real case.
\n
\nMoreover, with the popularity of smart phone, voice chat of instant message (IM) applications are getting popular. However, huge amount of manpower is required to listen, analyze, and identify relevant chat files of IM apps during current forensic investigation tools and techniques. This study proposes a semi-automatic integrated framework to deal with audio forensic investigation for IM apps by applying modern technologies. The main objective is to reduce the amount of manpower in the investigation. This is the first work that applies speech to text technology in voice chat of IM apps forensic. Both text and audio features are extracted to reconstruct the dialog conversation. Experiments with real case data show that the framework is promising. And it is able to translate dialog into readable text and improve the efficiency during investigation with reconstructed conversation.

Double JPEG image compression detection, or more specifically, double quantization detection, is an important digital image forensic method to detect the presence of image forgery or tampering. In this paper, we introduce an improved double quantization detection method to improve the accuracy of JPEG image tampering detection. We evaluate our detection method using the publicly available CASIA authentic and tampered image data set of 9501 JPEG images. We carry out 20 rounds of experiments with stringent parameter setting placed on our detection method to demonstrate its robustness. Each round of classifier is generated from a unique, non-overlapping and small subset composing of 1/20 of the tampered and 1/72 of the authentic images, to obtain a training data set of about 100 images per class, with the rest of the 19/20 of the tampered and 71/72 of the authentic images used for testing. Through the experiments, we show an average improvement of 40.31% and 44.85% in the true negative (TN) rate and true positive (TP) rate, respectively, when compared with the current state-of-the-art method. The average TN and TP rates obtained from 20 rounds of experiments carried out using our detection method, are 90.81% and 76.95%, respectively. The experimental results show that our JPEG image forensics method can support a reliable large-scale digital image evidence authenticity verification with consistent good accuracy. The low training to testing data ratio also indicates that our method is robust in practical applications even with a relatively limited or small training data set available.

The artifacts by JPEG recompression have been demonstrated to be useful in passive image authentication. In this paper, we focus on the shifted double JPEG problem, aiming at identifying if a given JPEG image has ever been compressed twice with inconsistent block segmentation. We formulated the shifted double JPEG compression (SD-JPEG) as a noisy convolutive mixing model mostly studied in blind source separation (BSS). In noise free condition, the model can be solved by directly applying the independent component analysis (ICA) method with minor constraint to the contents of natural images. In order to achieve robust identification in noisy condition, the asymmetry of the independent value map (IVM) is exploited to obtain a normalized criteria of the independency. We generate a total of 13 features to fully represent the asymmetric characteristic of the independent value map and then feed to a support vector machine (SVM) classifier. Experiment results on a set of 1000 images, with various parameter settings, demonstrated the effectiveness of our method.

In this letter, we propose a simple reversible watermarking method for JPEG images using the characteristics of JPEG compression. The method embeds a watermark into a JPEG image, and it extracts the watermark from the watermarked JPEG image and at the same time can recover the watermarked JPEG image to an original unwatermarked JPEG image. Moreover we investigate the number of embeddable blocks, the quality of watermarked images, and the increase of file-size by embedding a watermark.

Millions of digital images are captured by imaging devices on a daily basis. The way imaging devices operate follows an integral process from which the information of the original scene needs to be estimated. The estimation is done by inverting the integral process of the imaging device with the use of optimization techniques. This linear inverse problem, the inversion of the integral acquisition process, is at the heart of several image processing applications such as denoising, deblurring, inpainting, and super-resolution. We describe in detail the use of linear inverse problems in these applications. We review and compare several state-of-the-art optimization algorithms that invert this integral process. Linear inverse problems are usually very difficult to solve. Therefore, additional prior assumptions need to be introduced to successfully estimate the output signal. Several priors have been suggested in the research literature, with the Total Variation (TV) being one of the most prominent. In this thesis, we review another prior, the l0 pseudo-norm over the gradient domain. This prior allows full control over how many non-zero gradients are retained to approximate prominent structures of the image. We show the superiority of the l0 gradient prior over the TV prior in recovering genuinely piece-wise constant signals. The l0 gradient prior has shown to produce state-of-the-art results in edge-preserving image smoothing. Moreover, this general prior can be applied to several other applications, such as edge extraction, clip-art JPEG artifact removal, non-photorealistic image rendering, detail magnification, and tone mapping. We review and evaluate several state-of-the-art algorithms that solve the optimization problem based on the l0 gradient prior. Subsequently we apply the l0 gradient prior to two applications where we show superior results as compared to the current state-of-the-art. The first application is that of single-image reflection removal. Existing solutions to this problem have shown limited success because of the highly ill-posed nature of the problem. We show that the standard l0 gradient prior with a modified data-fidelity term based on the Laplacian operator is able to sufficiently remove unwanted reflections from images in many realistic scenarios. We conduct extensive experiments and show that our method outperforms the state-of-the-art. In the second application of haze removal from visible-NIR image pairs we propose a novel optimization framework, where the prior term penalizes the number of non-zero gradients of the difference between the output and the NIR image. Due to the longer wavelengths of NIR, an image taken in the NIR spectrum suffers significantly less from haze artifacts. Using this prior term, we are able to transfer details from the haze-free NIR image to the final result. We show that our formulation provides state-of-the-art results compared to haze removal methods that use a single image and also to those that are based on visible-NIR image pairs.

In this paper a method is presented for copyright protection in digital images. Copyright protection is achieved by embedding an invisible signal, known as digital signature or watermark, in the digital image. The method proposed in this paper casts the signature in the frequency domain by slightly modifying the values of randomly selected DC coefficients of the Discrete Cosine Transform (DCT) of the image. The same method is applied also on the Subband or Wavelet Transform coefficients. An adaptive method is proposed also based on perceptual criteria that guarantees the invisibility of the watermark and avoids the deterioration of the image. Signature detection is done via hypothesis testing, without to use any information from the original image. The watermarks embedded by the proposed method are very resistant to JPEG and other frequently used compression. Experimental results using real image data verify the effectiveness of the method.

Geometric attacks are the Achilles heel for many image watermarking schemes. Geometric attacks can be decomposed into two classes: global affine transforms and local geometrical distortions. Most countermeasures proposed in the literature only address the problem of global affine transforms (e.g. rotation, scaling and translation). In this paper, we propose a blind image watermarking algorithm robust to local geometrical distortions such as row or column removal, cropping, local random bend, etc. The robust feature points are adaptively extracted from digital images and local image regions (circular regions) that are invariant to geometric attacks are obtained according to the multi-scale space representation and image normalization. At each local image region, the watermark is embedded by quantizing the magnitudes of the pseudo-Zernike moments. By binding digital watermark with local image regions, resilience against local geometrical distortions can be readily obtained. Experimental results show that the proposed image watermarking is not only invisible and robust against common image processing operations, such as sharpening, noise adding, JPEG compression, etc, but also robust against geometric attacks such as rotation, translation, scaling, row or column removal, copping, local random bend, etc.

Abstract:- In order to improve the detection of malicious tampering images, it is necessary to decrease the fragility of hidden watermarks, even for digital images which have been incidentally distorted. In this paper, we propose a new invariant semi-fragile digital watermarking technique based on eigenvalues and eigenvectors of real symmetric matrix generated by the four pixel pairs. A signature bit for detecting the malicious tampering of an image is generated using the dominant eigenvector. And the multi-rings Zernike transform (MRZT) is proposed to achieve geometric invariance. The MRZT method is to the geometric distortions even when the image is under malicious attacks. Experimental results show that this algorithm can resist high quality JPEG compression, and improve the detection performance of various malicious tampering. Key-Words: fragile watermarking, Eigenvalue, Geometric invariance, Zernike moment 1

Digital image forgery and its forensics have emerged as a significant research domain. Digital forensics is required to examine the questioned images and classify them as authentic or tampered. This paper aims at image tamper detection using a novel energy deviation measure (EDM). The EDM is a measure of deviation in pixel intensity with respect to its immediate and distant neighbourhood. It is extracted by measuring the inter pixel intensity difference across and inside the DCT block boundary of a JPEG image. Features from EDM have been used for the classification of the authentic and tampered images. Support vector machine is used for image classification. The experimental results have shown that the proposed method performs better with fewer dimensions as compared to other state of the art methods. It gives improved accuracy and area under curve while classifying images. It is robust to noise and JPEG image compression quality factor.

This study aims to compare the integrity and reproducibility of measurements created from uncompressed and compressed digital images in order to implement compliance with 21 CFR Part 11 for image analysis studies executed using 21 CFR Part 58 compliant capture systems. Images of a 400-mesh electron microscope grid and H&E stained rat liver tissue were captured on an upright microscope with digital camera using commercially available analysis software. Digital images were stored as either uncompressed TIFFs or in one of five different levels of JPEG compression. The grid images were analyzed with automatic detection of bright objects while the liver images were segmented using color cube-based morphometry techniques, respectively, using commercially-available image analysis software. When comparing the feature-extracted measurements from the TIFF uncompressed to the JPEG compressed images, the data suggest that JPEG compression does not alter the accuracy or reliability to reproduce individual data point measurements in all but the highest compression levels. There is, however, discordance if the initial measure was obtained with a TIFF format and subsequently saved as one of the JPEG levels, suggesting that the use of compression must precede feature extraction. It is a common practice in software packages to work with TIFF uncompressed images. However, this study suggests that the use of JPEG compression as part of the analysis work flow was an acceptable practice for these images and features. Investigators applying image file compression to other organ images will need to validate the utility of image compression in their work flow. A procedure to digitally acquire and JPEG compress images prior to image analysis has the potential to reduce file archiving demands without compromising reproducibility of data.

Based on hyperchaos and discrete fractional Fourier transform (DFRFT) algorithm, a digital image watermarking algorithm based on hyperchaos sequence and the discrete fractional Fourier is presented in this paper. First, the original image and the watermarking image are transformed, respectively. At the same time, according to cipher sequence generated by the hyperchaotic array, the watermarking image's real part and imaginary part are encrypted, respectively. Second, the encryption watermark image's real part and imaginary part alternative embedded into the host FRFT coefficients. Then, the watermarking image is obtained after Fourier inversely transformed. Moreover, different attacks such as noise, JPEG compression and the cropping have been analyzed for the robustness of the watermark image. The results of the experiment have shown that the proposed algorithm is robust against common image manipulations.