Detecting and localizing image splicing has become essential to fight against malicious forgery. A major challenge to localize spliced areas is to discriminate between authentic and tampered regions with intrinsic properties such as compression artifacts. We propose CAT-Net, an end-to-end fully convolutional neural network including RGB and DCT streams, to learn forensic features of compression artifacts on RGB and DCT domains jointly. Each stream considers multiple resolutions to deal with spliced object's various shapes and sizes. The DCT stream is pretrained on double JPEG detection to utilize JPEG artifacts. The proposed method outperforms state-of-the-art neural networks for localizing spliced regions in JPEG or non-JPEG images.

Seam carving is a representative content-aware image retargeting approach to adjust the size of an image. To preserve visually prominent content, seam-carving algorithms first calculate the connected path of pixels, referred to as the seam, according to a defined cost function and then adjust the size of an image by removing or duplicating repeatedly calculated seams. Seam carving is actively exploited to overcome diversity in the resolution of images between applications and devices; hence, detecting the distortion caused by seam carving has become important in image forensics. In this paper, we propose a convolutional neural network (CNN)-based approach to classifying seam-carving forgery. To attain the ability to learn low-level features, we designed a convolutional neural network (CNN) architecture comprising five types of network blocks specialized in capturing local artifacts caused by seam carving. An ensemble module is further adopted to both enhance performance and comprehensively analyze the features in the local areas. To validate the effectiveness of our work, extensive experiments based on various CNN-based baselines were conducted. Compared to the baselines, our work exhibits state-of-the-art performance in terms of three-class classification (original, seam inserted, and seam removed). The experimental results also demonstrate that our model with the ensemble module is robust for various unseen cases.

Content-aware image resizing is the process of adjusting the size of an image while preserving its important content. Image resizing is used to overcome diversity in resolutions between modules, such as display devices and applications, and can thus be deliberately exploited to distort or remove original content; therefore, detecting such tampering has become an important topic in forensics. This paper proposes a deep neural network architecture to capture subtle local artifacts caused by seam-based image resizing. Unlike past approaches that only classified two classes, our approach is the first attempt to solve a given forensic task with three-class classification: original, seam insertion, and seam carving. The experimental results show that our work performs better than the handcrafted feature-based method and networks designed for different forensic tasks.

Double compression (DC) involves compressing a video twice with the video codec. DC usually occurs when a compressed video is manipulated, because most videos are saved in the compression format after manipulating. Given this principle, if we can detect DC traces in videos, we can determine whether a video has been manipulated. Prior works have proposed methods for detecting DC of H.264 videos; however, they have limitations in that they can detect DC only for a limited dataset and set of compression parameters. To overcome these limitations, we propose a two-stream neural network that incorporates two components that analyze intra-coded frames and predictive frames. We explain why the proposed two-stream neural network can detect DC traces in videos more accurately than the prior method, and we demonstrate that the proposed network can detect DC videos that have various contents and parameters through extensive experiments.

Image recapture is one of the simplest methods of content leakage, but is very difficult to address with the rapid development and dissemination of digital devices. This paper proposes novel secure display schemes to prevent image recapture on a polarized 3D system without additional hardware. Using commercial polarized 3D characteristics, the secret image to be protected from recapturing is encrypted on a pixel-by-pixel basis and decrypted with polarized glasses (PGs). Moreover, the proposed secure display schemes allow for the restricted viewing of secret images that only authorized people may see or the selective viewing of secret images and public images that anyone may see. Quantitative and qualitative analyses of the experimental results show that the proposed schemes can prevent content leakage from recaptured still or time axis images via a digital camera because the secret image is statistically encrypted, similar to random noise. As such, the restricted viewing or the selective viewing is possible depending on the presence of PGs. Therefore, the proposed schemes not only prevent image recapture using digital devices but also implement a secure display using PGs.

Median filtering is used as an anti-forensic technique to erase processing history of some image manipulations such as JPEG, resampling, etc. Thus, various detectors have been proposed to detect median filtered images. To counter these techniques, several anti-forensic methods have been devised as well. However, restoring the median filtered image is a typical ill-posed problem, and thus it is still difficult to reconstruct the image visually close to the original image. Also, it is further hard to make the restored image have the statistical characteristic of the raw image for the anti-forensic purpose. To solve this problem, we present a median filtering anti-forensic method based on deep convolutional neural networks, which can effectively remove traces from median filtered images. We adopt the framework of generative adversarial networks to generate images that follow the underlying statistics of unaltered images, significantly enhancing forensic undetectability. Through extensive experiments, we demonstrate that our method successfully deceives the existing median filtering forensic techniques.

As computer graphics technology advances, it is becoming increasingly difficult to determine whether a given picture was taken by camera or via computer graphics. In this work, we propose a method to using simple CNN structures to identify photorealistic computer graphics (PRCG) using convolutional neural networks (CNN). This network trained to identify the source of image patches. We showed the network without pooling layer showed 98.2% accuracy, which is 2.1% higher than the result of using conventional object-recognition network. Testing random patches from image, the accuracy of identifying image reached 98.5%. Furthermore, it is possible to detect the photograph-PRCG synthesized regions from the image.

As technological developments have enabled high-quality fingerprint scanning, sweat pores, one of the Level 3 features of fingerprints, have been successfully used in automatic fingerprint recognition systems (AFRS). Since the pore extraction process is a critical step for AFRS, high accuracy is required. However, it is difficult to extract the pore correctly because the pore shape depends on the person, region, and pore type. To solve the problem, we have presented a pore extraction method using deep convolutional neural networks and pore intensity refinement. The deep networks are used to detect pores in detail using a large area of a fingerprint image. We then refine the pore information by finding local maxima to identify pores with different intensities in the fingerprint image. The experimental results show that our pore extraction method performs better than the state-of-the-art methods.

Hue modification is a common strategy used to distort the true meaning of a digital image. In order to detect this kind of image forgery, we proposed a robust forensics scheme for detecting hue modification. First, we pointed out that photo response nonuniformity (PRNU) separated by a color filter array forms a pattern independent of others, since the position of each PRNU pixel means that they do not overlap. Using PRNUs from each color channel of an image, we designed a forensic scheme for estimating hue modification. We also proposed an efficient estimation scheme and an algorithm for detecting partial manipulation. The results confirmed that the proposed method distinguishes hue modification and estimates the degree of change; moreover, it is resistant to the effects of common image processing.

Because they will impact so many areas, copyright issues will inevitably arise as 3D printing expands into the content industry. The problem is that protections based on conventional methods are not effective, because the 3D printing process disables those protections. In this paper, we propose a robust and blind watermarking scheme that is able to protect content not only when the 3D model is shared in the digital world, but also when the 3D digital content is converted to analog content by 3D printing. First, we base our proposed watermark on a component that is unchanging to the printing direction for robustness against the printing process. The printing artifacts, instead of being regarded as severe distortion, are treated as a template that provides orientation information to the watermark detector. To achieve this, we also propose a blind estimation algorithm for the printing direction that starts from an analysis of the layering artifact. Using the results from a proposed estimator, the watermark from the printed-and-scanned model is synchronized with the original orientation. With the results of our tests with various 3D mesh models and attacks, we experimentally verified that the proposed method does not lose embedded patterns during the 3D print-scan process, especially with low-cost printers. Moreover, our method provides a new solution for estimating the printing direction that will be useful in a variety of fields.

We propose a new protection scheme for displaying a static binary image on a screen. The protection is achieved by a visual cryptography algorithm that divides the target images into several divisions. The visual difference between the text and the background is induced by exploiting the temporal responsibilities of the human eye. With the results of our user study, we demonstrate that encrypted visual information was mentally recovered by the human visual system. Moreover, the images captured from our scheme do not provide any meaningful information to the human eye, so that our method provides a strong security measure against screenshot piracy.

3D printing content is a new form of content being distributed in digital as well as analog domains. Therefore, its security is the biggest technical challenge of the content distribution service. In this paper, we analyze the 3D print-scan process, and we organize possible distortions according to the processes with respect to 3D mesh watermarking. Based on the analysis, we propose a circular shift coding structure for the 3D model. When the rotating disks of the coding structure are aligned in parallel to the layers of the 3D printing, the structure preserves a statistical feature of each disk from the layer dividing process. Based on the circular shift coding structure, we achieve a 3D print-scan resilient watermarking scheme. In experimental tests, the proposed scheme is robust against such signal processing, and cropping attacks. Furthermore, the embedded information is not lost after 3D print-scan process.

In digital image forensics, previous methods for hue forgery detection cannot be used after common image processing such as resizing and JPEG compression. In this paper, we suggest a robust forensics scheme for estimating hue modification of images. To achieve this goal, we use sensor pattern noise from each color channel of un-tampered images as the ground truth. Since we know the unique characteristics of each color channel, we can estimate a hue modification by testing suspicious images for all hue changes. The results confirms that the proposed method distinguishes hue modification and estimates the changed degree; moreover, it provides robustness against resizing and JPEG compression.

Bio-cryptographic systems provide security by securing a cryptographic key with biometric data. To secure the cryptographic key, biometric data should be significantly consistent. However, the uncertainty of biometrics makes it complicated to achieve highly accurate authentication. In this paper, we propose a novel bio-cryptographic authentication scheme that improves the security of Public Key Infrastructure. To confirm authenticity of a public key and make authentication information using biometric features consistent, we created a one-way, bio-hash function by using a public-key bitstream and Zernike moments extracted from fingerprint images. Experimental results and analysis show that the proposed method prevents attackers from modifying public-key certificates and getting access to biometric data. Moreover, the results prove that the proposed method achieves very high accuracy.

This paper proposes a forensic technique to localize duplicated image regions based on Zernike moments of small image blocks. We exploit rotation invariance properties to reliably unveil duplicated regions after arbitrary rotations. We devise a novel block matching procedure based on locality sensitive hashing and reduce false positives by examining the moments' phase. A massive experimental test setup benchmarks our algorithm against state-of-the-art methods under various perspectives, examining both pixel-level localization and image-level detection performance. By taking signal characteristics into account and distinguishing between “textured” and “smooth” duplicated regions, we find that the proposed method outperforms prior art in particular when duplicated regions are smooth. Experiments indicate high robustness against JPEG compression, blurring, additive white Gaussian noise, and moderate scaling.

Copyright protection issues for personal user's User Generated Contents (UGC) have been arising with the rapid growth of Social Network Services (SNS), blogs, and video sharing services. In this situation, Creative Commons License (CCL) has been established as a convenient method to protect copyrights of UGCs. However, the lack of technical method to protect CCL causes illegal manipulations and sharing of CCL applied UGCs. In this paper, we propose a dual watermarking system which protects copyrights of CCL applied UGCs. Experimental results show that manipulations such as rescaling, cropping, clipping, temporal manipulation, and frame rate changing are detected by semi-fragile watermarks. Simultaneously, robust watermarks notify the CCL information after above manipulations. To prevent interference of two watermarks, robust watermarks are embedded at spatial domain while fragile watermarks are embedded at block based DCT domain.

Color visual cryptography (VC) encrypts a color secret message into n color halftone image shares. Previous methods in the literature show good results for black and white or gray scale VC schemes, however, they are not sufficient to be applied directly to color shares due to different color structures. Some methods for color visual cryptography are not satisfactory in terms of producing either meaningless shares or meaningful shares with low visual quality, leading to suspicion of encryption. This paper introduces the concept of visual information pixel (VIP) synchronization and error diffusion to attain a color visual cryptography encryption method that produces meaningful color shares with high visual quality. VIP synchronization retains the positions of pixels carrying visual information of original images throughout the color channels and error diffusion generates shares pleasant to human eyes. Comparisons with previous approaches show the superior performance of the new method.

Estimating printing source is applicable in many forensic situations. In this paper, we propose an electrophotographic printer identification scheme from its printed material, in which imperceptible halftone patterns are contained inherently. The halftone textures in each channel of CMYK domain are analyzed. We construct a histogram from angle values of linear features extracted by Hough transform. By averaging the histograms from multiple images, a printer's reference pattern is identified. The soure printer is determined by a maximum correlation value between the reference patterns and the histogram of given image. Experiments are performed on 9,000 images made by 9 printers. The result supports that the presented scheme clearly recognizes different halftone textures.

As digital images have been propagated all over the world, identification techniques for image sources become more important. In this paper, we analyze properties of the scanner and find the spectral noise appeared in the frequency domain. Based on this observation, we propose a new method for identifying scanner models from scanned images. To enhance characteristics of the spectral noise, a refining process is proposed and employed to scanned images. We make reference patterns with the refined noise for each scanner. The proposed method uses the Euclidean distance to calculate a similarity between each reference pattern and the extracted spectral noise from the test image. The experimental results show that the proposed method achieves high accuracy with various types of images.

This paper investigates the digital forensic problem of determining whether an image has been produced by a specific digital camera. We employ the binary hypothesis testing scheme to detect the presence of photo-response non-uniformity( PRNU) in the image. The main challenge of this scheme is the extremely weak amount of PRNU in the observed noise residual. We propose to extract from the noise residual the significant regions with higher signal quality and discard those regions heavily deteriorated by irrelevant noises. Experimental results demonstrate that the proposed algorithm can improve the identification performance in the sense of decreasing the false rejection rate, which is a critical measure in practical applications.

Many illegal copies of digital video productions for cinema release can be found on the Internet before their official release. During the illegal copying of cinema footage, composite geometric distortions commonly occur due to the angle of the camcorder relative to the screen. We propose a novel video watermarking based on spread spectrum way that satisfies the requirements for protecting digital cinema. It enables the detector to not only extract the embedded message but also estimate the position where the camcorder recording is made. It is sure that the proposed position estimating model (PEM) can judge the seat in a theater with a mean absolute error (MAE) of (33.84, 9.53, 50.38) cm. Experimental results using various types of films show that the presented method provides the mathematical model for detecting and investigating the position of the pirate.

This paper introduces a color visual cryptography encryption method that produces meaningful color shares via visual information pixel (VIP) synchronization and error diffusion halftoning. VIP synchronization retains the positions of pixels carrying visual information of original shares throughout the color channels and error diffusion generates shares pleasant to human eyes. Comparisons with previous approaches show the superior performance of the new method.

Color laser printers are nowadays abused to print or forge official documents and bills. Identifying color laser printers will be a step for media forensics. This paper presents a new method to identify color laser printers with printed color images. First, 39 noise features of color printed images are extracted from the statistical analysis of the HH sub-band on discrete wavelet transform. Then, these features are applied to train and classify the support vector machine for identifying the color laser printer. In the experiment, 9 models of 4 brands, Xerox, Konica, HP, Canon, are tested to classify the brand of color laser printer, the color toner, and the model of color laser printer. The results prove that the presented identification method performs well using the noise features of color printed images.

In this paper, we propose a blind watermark detection scheme which is robust against composite geometric distortions. The composite geometric distortions mean that an image or a video is rotated, perspective-projected, and resized at the same time. It commonly happens during capturing a movie in a theater by a camcorder illegally. Local auto-correlation function (LACF) with a novel estimation model is utilized for calculating composite geometric distortions. We demonstrate robustness of the presented scheme against composite geometric distortions for digital cinema.

There are various traditional video watermarking schemes robust to geometrical transformations. Most of them use synchronizing techniques or invariant transforms to resist geometric distortions. However, these algorithms are time-consuming or not resilient to mixed geometrical distortions. In this paper, we propose a video watermarking method that is robust to compound geometrical attacks and has low complexity. The proposed method embeds a reference pattern and a watermark pattern into the temporal axis of a video during each pre-defined frame interval. Even if the marked video is geometrically distorted, both the reference pattern and watermark one are always in sync. Thus, we can simply detect the watermark without additional synchronization step.

Reversible or lossless data hiding enables host media to be restored from marked media without any loss of host information. However, since most of existing lossless data hiding methods are fragile, hidden data cannot be extracted after marked media goes through alteration such as JPEG compression. In this paper, we present a robust (referred to as semi-fragile) lossless data hiding method that utilizes sub-sampling and block gravity center. Gravity center of sub-sampled part of each block is insensitive to alteration, so it achieves robustness. This technique can be applied to selective authentication for images. That is, if a marked image does not change at all, the hidden data is correctly extracted and at the same time an original image is recovered. After compression, the hidden data is still extracted without error. Experimental results prove that the presented scheme achieves both reversibility and robustness against the predefined distortion.

For difference-expansion (DE)-based reversible data hiding, the embedded bit-stream mainly consists of two parts: one part that conveys the secret message and the other part that contains embedding information, including the 2-D binary (overflow) location map and the header file. The first part is the payload while the second part is the auxiliary information package for blind detection. To increase embedding capacity, we have to make the size of the second part as small as possible. Tian's classical DE method has a large auxiliary information package. Thodi mitigated the problem by using a payload-independent overflow location map. However, the compressibility of the overflow location map is still undesirable in some image types. In this paper, we focus on improving the overflow location map. We design a new embedding scheme that helps us construct an efficient payload-dependent overflow location map. Such an overflow location map has good compressibility. Our accurate capacity control capability also reduces unnecessary alteration to the image. Under the same image quality, the proposed algorithm often has larger embedding capacity. It performs well in different types of images, including those where other algorithms often have difficulty in acquiring good embedding capacity and high image quality.

In this paper, we propose a blind watermark detection scheme for digital cinema which is robust against D-A/A-D conversion attack. D-A/A-D conversion attack commonly happens during capturing cinema with a camera and is composed of various geometrical distortions as well as signal processing distortions. In many video applications, geometric distortions have been compensated by utilizing auto-correlation function (ACF). Since the ACF cannot resist projective transform but affine transform, we present a robust watermark detection scheme utilizing local auto-correlation function (LACF) with a mathematical model. Also, we suggest a way to design watermark pattern to improve performance of the LACF. We demonstrate robustness of the presented scheme against D-A/A- D conversion attacks for digital cinema as well as common video processing attacks.

This paper presents a robust watermarking method for curves that uses informed-detection. To embed watermarks, the presented algorithm parameterizes a curve using the B-spline model and acquires the control points of the B-spline model. For these control points, 2D mesh are created by applying Delaunay triangulation and then the mesh spectral analysis is performed to calculate the mesh spectral coefficients where watermark messages are embedded in a spread spectrum way. The watermarked coefficients are inversely transformed to the coordinates of the control points and the watermarked curve is reconstructed by calculating B-spline model with the control points. To detect the embedded watermark, we calculate the difference between the original and watermarked mesh spectral coefficients with the same process for embedding. By calculating correlation coefficients between the detected and candidate watermark, we decide which watermark was embedded.

A blind watermarking algorithm for vector graphic images is presented. The algorithm is resilient to both global and local geometrical distortions. The polygonal line is represented by the wavelet descriptor. An additive watermarking scheme is used to embed the watermark by slightly modifying the wavelet descriptor, and that causes invisible distortions to the coordinates of the vertices. The invariant properties of the wavelet descriptor ensure that the presented algorithm is resilient against both global and local geometrical distortions. Using vector graphic images from contour maps, we demonstrate that the presented algorithm outperforms the algorithm based on the Fourier descriptor.

Current difference-expansion (DE) embedding techniques perform one layer embedding in a difference image. They do not turn to the next difference image for another layer embedding unless the current difference image has no expandable differences left. The obvious disadvantage of these techniques is that image quality may have been severely degraded even before the later layer embedding begins because the previous layer embedding has used up all expandable differences, including those with large magnitude. Based on integer Haar wavelet transform, we propose a new DE embedding algorithm, which utilizes the horizontal as well as vertical difference images for data hiding. We introduce a dynamical expandable difference search and selection mechanism. This mechanism gives even chances to small differences in two difference images and effectively avoids the situation that the largest differences in the first difference image are used up while there is almost no chance to embed in small differences of the second difference image. We also present an improved histogram-based difference selection and shifting scheme, which refines our algorithm and makes it resilient to different types of images. Compared with current algorithms, the proposed algorithm often has better embedding capacity versus image quality performance. The advantage of our algorithm is more obvious near the embedding rate of 0.5 bpp.

The paper introduces a robust image watermark based on an invariant image feature vector. Normalized Zernike moments of an image are used as the vector. The watermark is generated by modifying the vector. The watermark signal is designed with Zernike moments. The signal is added to the cover image in the spatial domain after the reconstruction process. We extract the feature vector from the modified image and use it as the watermark. The watermark is detected by comparing the computed Zernike moments of the test image and the given watermark vector. Rotation invariance is achieved by taking the magnitude of the Zernike moments. An image normalization method is used for scale and translation invariance. The robustness of the proposed method is demonstrated and tested using Stirmark 3.1. The test results show that our watermark is robust with respect to geometrical distortions and compression.

Guaranteed service will provide high quality services to real-time applications, e.g., audio or video, over packet networks such as the Internet. To support guaranteed service, a service discipline must guarantee a delay bound to each session. In addition, a preferred service discipline should achieve high network utilization and good scalability. The service disciplines studied so far have problems in achieving these two objectives at the same time. Generalized processor sharing (GPS) service disciplines can have low network utilization. Rate-controlled (RC) service disciplines have difficulty in scalability because of regulators. For service curve (SC) service disciplines, both the network utilization and the scalability depend on the adopted SC. To date, there have been no studies on an SC which can make an SC discipline achieve these two objectives. We propose a new service discipline based on SC service disciplines. The proposed discipline achieves these two goals in a variable-sized packet environment. We show that the discipline can achieve the network utilization achievable by the RC service disciplines. We further show that our SC requires O(1) complexity for deadline calculation. Different from the RC service disciplines, the SC service discipline with our SC does not need regulators at all. Thus, it has better scalability than the RC service disciplines and is work-conserving. We also show that the proposed SC makes SC service disciplines have strictly higher network utilization than the GPS service disciplines including the multi-rate service discipline.

Because the Internet traffic, that will be the major traffic of broadband integrated services digital networks, is bursty when cells are being switched within the multistage switching network, it has a higher possibility that multiple cells arriving simultaneously at a switching element through different incoming links may have to be forwarded along the same outgoing link. We propose a high-performance large-scale ATM switch dealing with such link contention problem. It is a new unbuffered augmented Banyan network using fully adaptive self-routing control: the deflection self-routing Banyan network. To utilize all the links of the network as alternate paths, we employ the deflection-routing algorithm in each switching element, such that cells failing to get selected for the intended link are sent along different links, in the hope that they later return, or detour the contended link and continue their journey to the destination. Cells are never dropped within the switching network, whereas the switch has no multiple cell buffers. The proposed routing is as simple as that of the generic Banyan network, and all the switch elements (SEs) have a uniform structure. To design the proposed network and its self-routing, we use the topological properties that all the SEs of the Banyan network are arranged in a regular pattern topologically. We formulate and prove these properties through an algebraic formalism. We also ran a performance analysis to provide quantitative comparison against the Banyan network and the replicated Banyan networks. As a result, we show that the new network has a far better performance and scalability than the other networks.