3D mesh watermarking is an effective way to protect the copyright of 3D mesh models by imperceptibly embedding an ownership message into their geometric or topological information. The embedded watermark is supposed to be accurately extracted from the watermarked meshes after suffering various malicious or accidental geometric distortions during transmission, such as adding noise and smoothing. However, existing works rely on the geometric information of mesh models for watermark synchronization and extraction, resulting in poor robustness to geometric distortions. To address this issue, we propose a novel robust mesh watermarking scheme that is completely independent of geometric information. A watermark synchronization scheme is designed based on the Fiedler vector of the Kirchhoff matrix, upon which a watermark encoding strategy is introduced. Moreover, leveraging the vertex valence distribution of closed triangular meshes, we develop two topological embedding techniques to embed the watermark imperceptibly. Extensive experiments show that the proposed method achieves stronger robustness against geometric distortions, including similarity transformations, adding noise, smoothing, quantization, and even combination distortions, while maintaining the watermarked mesh quality compared with state-of-the-art methods.

Constructing the cover image with strong resistance to embedding distortion across multiple domains holds great promise for secure adaptive steganography. Nevertheless, existing CNN-based cover enhancement schemes, while effective against their target steganalyzers, struggle to resist detection beyond their target steganalyzer. Moreover, existing immune-based cover enhancement schemes focus solely on optimizing the spatial characteristics of the original cover. However, covers that can only resist spatial embedding distortion are insufficient to ensure security for steganography in other domains. In this paper, we propose a universal immunized (Uimm) cover construction scheme based on artificial immune evolution to enhance steganographic security, which is universal across multiple domains. Inspired by the similarity between steganography and immune response, we treat the original cover as the organism, the embedding distortion as the pathogen, and the universal immunoprocessing (UIP) applied to the original cover as the antibody. Moreover, the filter banks are meticulously designed to adaptively limit the universal immunoprocessing region (UIPR) and intensity of UIP, thereby fully considering the texture characteristics of the cover in the spatial domain. Under the constraints of the UIPR and UIP intensity, antibodies are guided to evolve towards improving the JPEG steganography security, yielding the optimal universal immunoprocessing policy. By considering intrinsic characteristics in both the spatial and JPEG domains, optimal universal immunoprocessing performed on the original cover will enhance its resistance to embedding distortion across multiple domains. Experimental results demonstrate that the proposed Uimm cover significantly improves the holistic security of adaptive steganography in both spatial and JPEG domains, effectively resisting handcrafted and CNN-based steganalysis. This represents a conceptual leap from single-domain immunity to cross-domain general immunity. It also achieves superior performance compared to related schemes.

With the widespread adoption of digital office systems, electronic documents have become the primary medium for information storage and dissemination. However, PowerPoint (PPT) documents are vulnerable to unauthorized photography during presentations, leading to potential sensitive information leakage and copyright infringement. Traditional document security measures, such as password protection and access control, fail to prevent offline photography attacks, while existing digital watermarking techniques face challenges due to PPT's dynamic characteristics and cross-channel distortion. To address these issues, this paper proposes a camera shooting resilient watermarking for PowerPoint presentation. The scheme first selects the master layout backdrop as the watermark carrier according to PPT's compositional features, enabling globally consistent watermark deployment without interfering with document editing functions. By systematically analyzing distortion characteristics during camera-shooting, we identify an optimal frequency band for watermark embedding. Furthermore, a cost-constrained adaptive embedding strategy is adopted to enhance watermark robustness while maintaining imperceptibility. Experimental results demonstrate that the proposed solution effectively resists interference in various complex photography scenarios, providing a viable approach for protecting PPT documents.

The performance of deep neural networks in Face Forgery Detection (FFD) is impressive, but they are obtained with static models incapable of adapting their behavior over time. However, in a dynamic world, FFD systems deployed on Internet of Things (IoT) devices operate on vast streams of face data with ever-changing distributions, leading to catastrophic forgetting and significant declines in detection performance. To address this issue, we propose a novel Rehearsal-Free Domain Lifelong Learning (RF-DLL) framework for FFD, termed SDProtoL. This framework adopts the prompt-based incremental learning paradigm and steers prototypes to further mitigate catastrophic forgetting and improve generalization to unseen data. Specifically, to alleviate the forgetting of previous domains without using previous data, we employ a Gaussian Mixture Model (GMM) to derive GMM-based Hierarchical Static Prototypes (GHSP) with non-convex and anisotropic characteristics, in order to fit the complex distribution of face forgery data. Furthermore, for better generalization, we propose the Dynamic Prototype-oriented Asymmetric Contrastive Regularization (DPACR), which improves generalization ability for unseen data by accumulating and transferring previous knowledge. Simulating practical RF-DLL scenarios, we establish a challenging Lifelong Face Forgery Detection (LFFD) benchmark and construct three protocols referencing real-world scenarios. Extensive experimental results demonstrate that our proposed method significantly alleviates catastrophic forgetting while exhibiting superior generalization performance in unseen domains.

Given the widespread dissemination of digital audio and the advancements in speech synthesis technologies, protecting audio copyright has become a critical issue. Although watermarks play an important role in copyright verification and forensic analysis, they are insufficient to proactively defend against malicious speech synthesis. To address this issue, we introduce a novel adversarial speech synthesis watermarking mechanism (ASSMark), which simultaneously traces the audio copyright and disrupts the speech synthesis models by embedding robust adversarial watermarks in a one-time manner. Specifically, we design a unified training framework that models the embedding of watermarks and adversarial perturbations as collaborative tasks. This approach allows for the fine-tuning of any robust watermark into an adversarial watermark, resulting in watermarked audio that can effectively defend against unauthorized speech synthesis attacks. Experimental results demonstrate that ASSMark achieves over 90% protection rate even to unknown black-box models. Compared to simplistic two-step protection methods, it not only effectively resists synthesis attacks but also achieves superior watermark extraction accuracy and speech quality, offering an outstanding solution for protecting audio copyright.

Adversarial steganography presents a significant challenge in digital media security, leveraging adversarial perturbations to obscure steganographic patterns and evade detection by traditional steganalysis methods. This paper proposes a universal detection method based on Directional Adversarial Noise Search (DANS) to improve the robustness of steganalysis against adversarial steganography. Specifically, an augmentation network is employed to generate adversarial noise distributions, guided by a novel DANS loss function that enables precise optimization of noise distribution. A threshold constraint is further applied to ensure the controllability of the generated noise. By embedding the generated perturbations into both cover and stego images, a robust augmented dataset is constructed to improve detection performance. Experimental results demonstrate that the proposed method improves detection accuracy across various adversarial steganography techniques, achieving up to a 17.14% increase. Moreover, it exhibits superior generalization ability in cross-dataset evaluations, highlighting its effectiveness and robustness in diverse scenarios.

With the widespread adoption of smart devices and social networking platforms, the development of robust image steganography techniques for public lossy channels has become increasingly crucial. Among JPEG-resistant steganographic methods, PMAS (Postprocessing and precise dither Modulation based robust Adaptive Steganography) has demonstrated superior performance by utilizing high-quality images and maintaining resilience against aggressive compression. This method achieves remarkable concealment in user-shared images, presenting substantial challenges to public communication security. To counter this threat, we propose a specialized lightweight Scale-Free Network for mining Clues in downward JPEG-resistant steganography (SF-ClueNet), specifically designed to identify vulnerabilities in PMAS despite its sophisticated anti-detection mechanisms. Departing from conventional approaches that depend on high-pass filter residuals, SF-ClueNet extracts comprehensive global statistical features, enabling effective detection of dispersed steganographic artifacts. When integrated with lightweight residual feature miner, our method maintains pattern recognition capabilities as image dimensions increase, ensuring consistent detection performance. Experimental results demonstrate that SF-ClueNet significantly enhances detection accuracy, exhibits robust performance against data distribution shifts with minimal transfer loss, and supports direct analysis of high-resolution images. These advanced capabilities position SF-ClueNet as a viable and efficient solution for practical steganalysis applications across diverse operational environments.

The proliferation of video applications has exacerbated digital piracy issues, notably evidenced by unauthorized video digital editing and camcording processes. While existing research has introduced robust watermarking methods to safeguard video copyrights, these methods often address specific attack scenarios, limiting their overall efficacy. To address this problem, we propose a robust blind video watermarking scheme based on Frequency-Spherical Cavity Transformation (FSCT), offering a comprehensive solution for both digital editing and camcording processes. Our approach treats the spatial and temporal aspects of the video as a 3D cube, utilizing FSCT to ensure temporal translational invariance and resilience against spatial attacks. To mitigate artifacts induced by motion characteristics, we analyze the properties of FSCT and introduce a visual quality optimization strategy, enhancing imperceptibility while ensuring robustness. Simultaneously, during extraction process, the watermark can be successfully retrieved from video camcording with only a specified time interval, eliminating the need for temporal synchronization. Through extensive experimentation, the proposed method exhibits superior robustness against digital editing and camcording compared to existing methods.

Transition videos play a crucial role in media production, enhancing the flow and coherence of visual narratives. Traditional methods like morphing often lack artistic appeal and require specialized skills, limiting their effectiveness. Recent advances in diffusion model-based video generation offer new possibilities for creating transitions but face challenges such as poor inter-frame relationship modeling and abrupt content changes. We propose a novel training-free Transition Video Generation (TVG) approach using video-level diffusion models that addresses these limitations without additional training. Our method leverages Gaussian Process Regression ($\mathcal {GPR}$) to model latent representations, ensuring smooth and dynamic transitions between frames. Additionally, we introduce interpolation-based conditional controls and a Frequency-aware Bidirectional Fusion (FBiF) architecture to enhance temporal control and transition reliability. Evaluations of benchmark datasets and custom image pairs demonstrate the effectiveness of our approach in generating high-quality smooth transition videos. The project is provided inhttps://sobeymil.github.io/tvg.com.

In watermarking schemes, deriving the geometric invariants of the multimedia content is crucial for modern watermarking against geometric deformations. The invariants based on orthogonal moments can effectively describe the semantic content of multimedia due to their excellent mathematical properties. Modulating watermarked signals into these invariants can yield satisfactory geometric robustness. However, with the emerging risks posed by generative large models, the current theoretical analysis of the relationship between invariants and watermarking is still limited, and the intrinsic connection between them is neglected to varying degrees when designing moments-based watermarking schemes. To bridge this gap, we propose a set of design principles, including the texture complexity priority principle, uniform zeros distribution priority principle, and invariants conjugation priority principle, and reveal the critical influence of the mathematical properties of moments on video watermarking. Based on the above principles, we also propose a texture-aware adaptive video watermarking scheme based on orthogonal moments. Extensive experiments show that the proposed video watermarking scheme outperforms state-of-the-art watermarking algorithms regarding imperceptibility, robustness, and computational complexity.

Coverless generative steganography is a highly secure method of information hiding. With the advent of the AI-generated content (AIGC) era, the widespread dissemination of generative content on the internet provides an excellent hiding environment for generative steganographic images. Generative steganographic images do not require the participation of carrier images, making existing steganalysis methods expired. However, there are currently no detection methods specifically targeting generative steganographic content. To address this gap, we propose a steganalysis method for generative steganographic images. Our approach focuses on the intrinsic differences between generative steganographic images and ordinary generative images. Through comparative analysis, we propose optimizing the detection model using mutual information estimation. We hypothesize about the distribution characteristics of steganographic signals and design a feature discrimination loss function to further guide the model’s optimization. In addition to designing a feature extraction network to extract features from different image regions, we also incorporate an image classification model pretrained on a large dataset to extract classification features for the final classification. Experimental results in various training and testing scenarios demonstrate that the proposed model not only possesses excellent detection capability but also exhibits reliable generalization compared to other models. Furthermore, we provide necessary descriptions and analysis to validate the rationale behind the network design.

The misuse of AI-generated techniques in face forgery has raised significant concerns, driving advancements in detection methods. However, existing algorithms struggle with generalization in cross-domain scenarios due to domain shifts, limiting their practical applications. Prompt tuning, which learns soft prompts while freezing the backbone, enables the generalizable Vision-Language Models (VLMs) pre-trained on large-scale datasets to adapt to downstream tasks. Though effective, prompt tuning confronts challenges in face forgery detection, where its performance is sensitive to initialization and may undermine the generalizability of pre-trained VLMs. To address this issue, we propose a novel Gradient-aware Adaptive Meta-Prompt Learner (GAMP-Learner). The core idea is to learn a meta-general gradient from multiple source domains through the Direction-shared Gradient Pruning Module (DGPM) for efficient initialization in the inner-loop, while addressing gradient conflicts via the Adaptive Gradient Calibration Module (AdaGCM) to enhance generalization in the outer-loop. Notably, our GAMP-Learner can be seamlessly integrated into any prompt-based fine-tuning VLM in a model-agnostic way. Additionally, to capture fine-grained forgery clues, we design a Multi-Granularity Conditional Prompt Generator (MGCP), which constructs instance-level prompts by incorporating multi-scale content-style feature representations. Simulating practical scenarios, we devise three protocols which evaluate generalization performance trained on multiple source domains. Extensive experiments demonstrate that the proposed framework achieves competitive cross-domain detection performance compared to state-of-the-art methods.

The prevalence of digital content leakage via screen capture highlights the urgent need for robust watermarking solutions capable of withstanding cross-media transmission. Current approaches primarily focus on developing watermarking techniques resilient to screen-shooting distortions, where distinguishing the watermark signal from these distortions is paramount. In contrast, our study addresses an inverse problem by investigating the generation patterns of noise during screen-shooting and considering them as feasible representations of watermark signals. Leveraging Moiré patterns as one of the distortion signals naturally generated by the interaction between electronic screens and camera sensors, we propose Moiré-watermark, presenting watermark information encoded into meticulously crafted Moiré patterns within images. To enhance the naturalness of Moiré-watermark amidst the irregularities of screen-shooting Moiré patterns, we encode watermark signals using gratings at different angles. A corresponding angle-based decoding method facilitates effective blind extraction of watermarks. Comprehensive experimental evaluations under diverse conditions of distance, angle, lighting, and across various capturing and display devices, alongside comparisons with existing methods, validate the superior performance of Moiré-watermark.

In this article, the resilient secure synchronization of complex networks (CNs) that are subject to denial-of-service (DoS) attacks is studied. In contrast to existing logic processors, a new processor in which more essential information on DoS attacks, such as the number of sampling instants being attacked and attack moment being detected, can be captured is designed. Based on the benefits of the logic processor, a switching sampled-data (SD) control protocol in which different feedback gains are chosen for different attack cases is proposed. In contrast to existing control schemes, the switching SD control protocol is more flexible. Subsequently, according to different attack cases, a switching Lyapunov-Krasovskii function (LKF) that can effectively fulfil the switching SD control protocol is founded. New resilient secure synchronization criteria that can successfully counteract the effects of DoS attacks are then established for CNs based on the switching SD control protocol and switching LKF. Finally, a complex Chua’s circuit system is used to give evidence of the feasibility and superiority of the proposed method.

The widespread adoption of smartphones has introduced new challenges to document copyright protection, prompting the emergence of Screen-Shooting Resilient Document Watermarking (SSRDW) technology. In recent years, underpainting-based SSRDW techniques have proven to be highly effective. However, after careful study, we find that existing methods fail to simultaneously meet four essential criteria for SSRDW: high imperceptibility, strong robustness, adaptability to text processing, and high efficiency. In this paper, we introduce an enhanced underpainting-based SSRDW approach capable of satisfying all four requirements. Our approach enhances imperceptibility by employing underpainting embedding methods independent of text content. Additionally, we introduce a fast resynchronization mechanism to improve time efficiency. Furthermore, we propose an enhanced watermark extraction method that enhances robustness and enables watermark retrieval even in scenarios involving text processing. Extensive experimental validation underscores the superior performance of our enhanced SSRDW method.

With the rapid development of mobile Internet and video applications, robust video watermarking technology has become a focal area of research for protecting and tracking intellectual property rights in digital media. An important characteristic of video is that it has both spatial and temporal properties. Previous studies in video watermarking have primarily focused on either spatial or temporal distortions and did not uniformly consider all types of video distortion, which restricts the robustness of video watermarking. In this paper, a novel robust blind video watermarking is proposed by exploring consistent spatio-temporal distortion and stable 3-D DCT coefficients. Our method achieves stronger robustness by uniformly treating the spatial and temporal distortions of the video. The properties of the stable 3-D DCT coefficients are mathematically proved, which makes the scheme generalizable. Extensive experiments have demonstrated that our method is resistant to video compression (H.264/AVC, H.265/HEVC) attacks, geometric attacks, and temporal domain attacks, and outperforms the current state-of-the-art video watermarking schemes.

Existing video watermarking embeds robust watermarks in each frame of the video for copyright protection and tracking. However, just as any content written on a blank paper is easily perceived, embedding watermarks in the texture-poor frames impairs imperceptibility. Common geometric attacks such as scaling and rotation pose a significant challenge to the existing video watermarking. Image watermarking based on moments is robust against geometric attacks. However, moment-based watermarking is difficult to migrate to the video due to its lack of perceptual guarantee and high computational cost. In this paper, we propose an adaptive video watermarking scheme by exploring the relationship between moments and video textures, which can adaptively select texture-rich frames to embed watermarks for perceptual guarantee. Furthermore, we utilize the properties of moment calculation in videos to optimize efficiency. Extensive experiments show that the proposed method can achieve better imperceptibility than existing methods while maintaining strong robustness.

Song voice conversion tools have gained more and more popularity in the recent past. People have been uploading their self-made forgery songs on video websites, and these songs have been converted in timbre. However, singing voice conversion technology may cause copyright infringement of the songs. In order to protect the copyright of songs, the method of singing voice conversion detection needs to be investigated. We propose Rhythm and Pitch Aware Songs Conversion Detection (RPA-SCD), a dual-branch network for song voice conversion detection. RPA-SCD can predict forged song fragments through rhythm and pitch which are the global and local information of music. To evaluate the proposed method, we contribute a multilingual song conversion detection(MSCD) dataset. Our proposed model achieves the EER of 2.30% in the original domain of MSCD, which is lower than other benchmarks for speech forgery detection. The experiments show that our approach achieves state-of-the-art performance on the song conversion detection task. The MSCD dataset can be found at https://drive.google.com/file/d/1rFsvMYihVtk81uFbL7UpyUEs-qBgsX6H/view?usp=drive_link. The code can be found at https://github.com/Samantha-Du/RPA-SDD.

Malicious video tampering and unauthorized distribution on social networks pose significant challenges in distinguishing authentic from altered content and raise copyright infringement concerns. Although numerous watermarking schemes have been developed to protect copyright, tamper detection and localization have not been adequately addressed. To this end, we propose RDFMark, a dual-functional solution offering a unified framework for copyright protection and tamper detection. RDFMark introduces a new paradigm in video tamper detection by embedding a robust watermark that survives diverse attacks, enabling the extraction of copyright information and tampering cues. Extensive experiments validate the effectiveness of RDF-Mark against common video tampering scenarios encountered during social network transmissions, such as masking, barrage insertion, and cropping. Furthermore, it demonstrates resilience to geometric distortions, strong compression, re-compression, and composite attacks on social networks.

Neural-based natural language watermarking (NLW) shows promise for generating context-aware lexical substitutions, minimizing semantic loss in watermarked text. However, existing works confront two primary challenges: 1) the reliance and sensitivity on textual context during substitutes generation hinders text reversibility, and 2) strict synchronization constraints on the generation order of substitutes from both original and watermarked text blocks out some suitable substitutes, limiting watermark capacity. This paper puts forward a reversible neural NLW approach with improved capacity and text quality. Specifically, we construct a novel lexical substitution system (LSS), utilizing prompt learning for candidates generation and comprehensive assessment features for candidates ranking. A reversible watermarking scheme is then presented by ingeniously screening recoverable positions and enabling multi-bit substitutions via the proposed LSS. Experiments validate that our method achieves complete reversibility while enhancing watermark payload and text fidelity compared to prior arts.

Abstract Blind image watermarking is regarded as a vital technology to provide copyright of digital images. Due to the rapid growth of deep neural networks, deep learning-based watermarking methods have been widely studied. However, most existing methods which adopt simple embedding and extraction structures cannot fully utilize the image features. In this paper, we propose a novel Single-Encoder-Dual-Decoder (SEDD) watermarking architecture to achieve high imperceptibility and strong robustness. Precisely, the single encoder utilizes normalizing flow to realize watermark embedding, which can effectively fuse the watermark and cover image. For watermark extraction, we introduce a parallel dual-decoder to improve the imperceptibility and extracting ability. Extensive experiments demonstrate that better watermark robustness and imperceptibility are obtained by SEDD architecture. Our method achieves a bit error rate less than 0.1% under most attacks such as JPEG compression, Gaussian blur and crop. Besides, the proposed method also obtains strong robustness under combined attacks and social platform processing.

Supervisory Control and Data Acquisition (SCADA) systems are the core of industrial control systems and an important part of critical infrastructure. With the deployment of 5G networks around the world, SCADA systems are no longer a relatively secure and physically isolated system like in the past, but are facing huge network virus threats. In order to solve the problem that existing models ignore the communication between nodes in the system, we propose an industrial virus transmission model SELBR based on immune knowledge by simulating the function of T cells in the immune system. By introducing E node, the model is used to realize the function of information transfer between nodes. What’s more, we fit the numerical simulation results with the actual data set to verify the existence of the model, and verify the effectiveness of the model for controlling the spread of industrial viruses through model comparison experiments. Numerical results show that the model can effectively control the spread of the virus. Finally, on the basis of parameter sensitivity analysis, preventive suggestions are put forward to further strengthen the security of SCADA system.

The popularity of digital devices and the importance of information which lies in text documents lead to the overflowing of screen-shooting attacks for text documents nowadays. So the robust screen-shooting watermarking for text documents is in great demand. However, the existing watermarking of text documents in traditional channels which is widely researched is not available under the huge distortion from screen-shooting attacks. To tackle this problem, in this letter, watermarking based on text document structures which can persist in both print-camera and screen-shooting processes is developed. Our method has two main advantages: 1) Extensive robustness. We define partial brightness integration accompanied with stroke separation method to embed watermark information on characters. For various documents with different fonts' styles, fonts' sizes and different languages' fonts, it shows great performances. Besides, for different shooting conditions such as different shooting distances and angles, the accuracy is also guaranteed. 2) Great imperceptibility. By introducing special divisional method and adaptive intensity, high visual ability is achieved. Both of these two merits are verified by massive experiments.

With the proliferation of smartphones, preserving the confidentiality of digital documents has become increasingly challenging. Document formats, such as PowerPoint (PPT), are particularly susceptible to unauthorized capture and leakage through smartphone cameras. In this letter, we propose a camera shooting resilient watermarking for PowerPoint presentation based on the PPT master layout backdrop, which exhibits two key properties. 1) Imperceptibility. We employ a difference-based embedding method on the PPT master layout backdrop to ensure good visual quality of the watermarked PPT. 2) Robustness. We design a distance-based selection method for embedding coefficients, ensuring the watermark's robustness under various shooting conditions. The experimental results demonstrate the robustness of our scheme against screen and projection shooting of PPT presentation, across diverse shooting distances, angles, and coverage levels of the PPT master layout backdrop.

The cover image with strong resistance against embedding distortion has promise for improving the holistic security of steganography. However, existing methods use the vulnerability of Convolutional Neural Network (CNN) to construct the enhanced cover which can only deceive the target CNN-based steganalyzer, but is not more suitable for steganography. When resisting steganalysis outside the target steganalyzer, its performance drops significantly. In this paper, we propose an immune-cover construction scheme via Artificial Immune System (AIS). By the association between the steganography and immune theory, we regard the cover as the organism, the distortion introduced by steganography as the pathogenic factor, and the immunoprocessing for optimizing the original cover as the antibody. Based on AIS, the optimal immunoprocessing is dynamically searched and performed on the original cover to construct an immune-cover which is most suitable for steganography. Besides, the proposed method carefully selects the immunoprocessing region to prevent artifacts, and guarantees the visual quality of the immune-cover through the constraint of the immunoprocessing intensity. Extensive experimental results demonstrate that the proposed immune-cover has much stronger resistance against embedding distortion compared with the related methods, thus significantly improving the holistic security of the adaptive steganography evaluated on both traditional and CNN-based steganalyzers.

Recent advancements in image steganography demonstrate that reasonable cover enhancement approaches can effectively improve the security performance of steganography. However, the existing proposals based on adversarial steganography against targeted steganalyzers are insufficient in terms of eliminating or reducing anomalous pixels caused by subsequent embedding. This limitation impedes the full potential of cover enhancement schemes for enhancing steganographic security. In this article, we present Immucover, a novel immunized cover image construction method that leverages fuzzy enhancement and an artificial immune system (AIS) to incorporate texture region- and edge-region-adaptive enhancement. Specifically, we first design a parameterized method to adaptively enhance the texture region of the given cover image using a distortion function. Then, Immucover detects and enhances the edge region of the cover image using a fuzzy-parameterized approach based on an optimized smallest univalue segment assimilating nucleus for edge detection. Finally, a powerful AIS module acts as an optimizer to optimize the parameters that affect the texture and edge area, i.e., the area where the secret information is suitably embedded. In this way, a so-called immunized cover image is generated. In addition, we develop a novel affinity metric to assess the antibody quality within the AIS module, which guides the generation of the immunized cover with higher security. Comprehensive experiments conducted on widely used datasets demonstrate that our Immucover provides significantly improved resistance to steganalysis and enhances the security of steganography.

Performing post-processing on the stego image has promise for improving the steganography security. Nevertheless, the existing post-processing schemes neglect the characteristics of the stego image, which lack strong theoretical interpretability. Moreover, existing schemes do not fully consider the holistic steganography security against both CNN-based and handcrafted steganalyzers. In this paper, we propose a steganography immunoprocessing (IP) framework based on Artificial Immune System (AIS) that is universal for the stego images from the same steganographic process to further enhance the security. Based on the natural relationship between immune theory and steganography, we regard the immunoprocessing policy as the antibody, and the performance of the anti-steganalysis for stego images protected by antibody as the antibody affinity. By the immune dynamic optimization process, the optimal immunoprocessing policy is dynamically searched and performed on the stego image to achieve further optimization. In addition, we enhance the resistance of the stego against the target CNN-based steganalyzer by limiting the immunoprocessing direction. Performing the optimal immunoprocessing on stego images will enhance the holistic security of steganography. Experimental results demonstrate that the proposed immunoprocessing can significantly improve the holistic security of adaptive steganography against both CNN-based and handcrafted steganalyzers, and achieve better performance than related schemes.

In moment-based watermarking schemes, the accuracy of the moments is crucial for constructing robust watermarking schemes. The robustness of the watermarking scheme relies heavily on the proper representation of the moments. Despite the importance, current theoretical research on accuracy is very limited in watermarking techniques. To this end, we propose a novel robust image watermarking scheme based on accurate polar harmonic Fourier moments (PHFMs). Specifically, the accurate PHFMs computation based on polar pixel tiling with nearest neighbor interpolation (PPTN) is designed. This computation is general and used for embedder and extractor. This ingenious design eliminates geometric and numerical integration errors and also avoids the distortion interaction caused by watermarks. Also, an improved quantization strategy is applied to the embedding process, and satisfactory imperceptibility is obtained. The watermark is extracted without the host image. The experimental results show the excellent robustness of the proposed watermarking scheme to common image processing attacks, geometric attacks, and some kinds of compound attacks. The proposed scheme is superior to the state-of-the-art image watermarking schemes.

Under the framework of multiagent systems (MASs), this article is focused on the leaderless consensus of multiple Euler–Lagrange (EL) systems subject to denial-of-service (DoS) attacks. First, a new joint control protocol, which combines the distributed sampled-data (SD) control and adaptive distributed control (DC), is designed. The distributed SD control is equipped with logic processors, which can capture key information of DoS attacks, and the adaptive DC is conducive to relaxing some generally required constraint conditions. Second, a new property of a non-negative differentiable function is proposed, which is very helpful in solving the distributed SD control issue of multiple EL systems. Third, by setting up a$\textbf{W}$-dependent Lyapunov–Krasovskii functional (LKF) and utilizing the property of the non-negative differentiable function, new leaderless consensus results are derived for multiple EL systems with DoS attacks. The obtained leaderless consensus results are in the shape of linear matrix inequalities (LMIs) and can efficiently counter the influence of the DoS attacks. Ultimately, the effectiveness of the derived results is inspected by an MAS with multiple two-linked robot manipulators.

Existing image manipulation detection and localization methods tend to detect the trail of manipulation and achieve decent performance. So far, however, there has been little discussion about the category imbalance. In this paper, we propose a parallel attention mechanism based network to localize tampered regions, which is inclined to have better generalization, while it possesses higher model capacity. In addition, we designed a trainable parameter constrained shifted-window dual attention module to strengthen the manipulated features. Finally, to settle the difficulty of category imbalance, we conduct a category-based normalization loss function, which allows the model to pay more attention to the manipulated regions and further improves the generalization capability. Extensive experimental results demonstrate that the proposed approach can effectively reconcile the weights of different categories during training and produce state-of-the-art performance in various benchmark datasets.

Noise feature modules are commonly used in image manipulation localization. However, different noise learning modules can only target limited tampering methods. In actual image tampering localization tasks, the tampering methods are unknown; it is difficult for a single noise feature module to match all tampering methods. In this paper, we explore the ability of different noise feature modules to localize different manipulation types. Furthermore, we propose a Multi-Channel Noise Feature (MCNF) representation model to describe the noise feature of the tampered images. MCNF contains multiple noise feature modules, which can automatically discriminate the noise feature modules that are practical for localization results. Experiments on several standard image tampering datasets show that our MCNF model achieves state-of-the-art performance compared to alternative methods.

Recently, adversarial perturbations have been used to reassign cost which can enhance the security of steganography, called as adversarial embedding. However, existing methods selected costs to be modified by self-defined rules which were hard to achieve the optimal security against steganalyzers. In this paper, we propose an automatic adversarial embedding scheme called RLAE (deep Reinforcement Learning-based content-adaptive Adversarial Embedding). In RLAE, an agent network utilizes a generative network which generates an embedding policy for cost reassignment automatically according to a basic steganography cost map. Then, an environment network employs a steganalyzer as an attack target that offers rewards for optimizing the agent network. To provide more comprehensive information, we design a joint reward by considering both the adversarial perturbations calculated from the environment network and noise residual signal representing image textures. Experimental results show that the security of the proposed RLAE is superior than state-of-the-art works, especially steganography with for the large payloads.

According to the Buckets effect, the capacity of a bucket depends on the length of the shortest board. This principle also applies to social networking platform resilient (SNPR) image watermarking, which should be comprehensive and free from significant shortcomings. In the frequency domain, the watermarked region is formed using log-polar coordinate mapping (LPM) and has a ring-like structure. However, this structure cannot be stretched or compressed, and it causes a streaking effect at the edges of the watermarked image. These issues have been addressed in the proposed method. Specifically, an adaptive optimization framework is used to adjust the embedding strength and range of the watermark, and multiple synchronization strategies are adopted to correct flip and aspect ratio. Compared with state-of-the-art works, the proposed method significantly improves the imperceptibility of the watermarked image and its robustness to various distortions and lossy transmission on social networking platforms (SNPs).

The emergence of artificial intelligence-generated content (AIGC) has raised concerns about the authenticity of multimedia content in various fields. However, existing research for forgery content detection has focused mainly on binary classification tasks of complete videos, which has limited applicability in industrial settings. To address this gap, we propose UMMAFormer, a novel universal transformer framework for temporal forgery localization (TFL) that predicts forgery segments with multimodal adaptation. Our approach introduces a Temporal Feature Abnormal Attention (TFAA) module based on temporal feature reconstruction to enhance the detection of temporal differences. We also design a Parallel Cross-Attention Feature Pyramid Network (PCA-FPN) to optimize the Feature Pyramid Network (FPN) for subtle feature enhancement. To evaluate the proposed method, we contribute a novel Temporal Video Inpainting Localization (TVIL) dataset specifically tailored for video inpainting scenes. Our experiments show that our approach achieves state-of-the-art performance on benchmark datasets, including Lav-DF, TVIL, and Psynd, significantly outperforming previous methods. The code and data are available at https://github.com/ymhzyj/UMMAFormer/.

Digital images are widely used nowadays, carrying a great deal of information. However, copyright infringement of digital images occurs from time to time which poses a threat to the rights and interests of copyright owners and the security of information content. To deal with such problems, in this paper, we propose an adaptive robust watermarking scheme to protect the copyright of color digital images. Spread spectrum watermark and two kinds of synchronization information are embedded in the luminance and chrominance channels of the color image, respectively. The synchronization information helps to restore geometric distortions. We also design a pre-quantization strategy to mitigate the distortions caused by social platforms. To achieve better imperceptibility, we propose an adaptive strategy based on the peculiarities of Human Visual System to adjust the strength of watermark residuals. Experimental results demonstrate that the proposed method is equipped with comprehensive and balanced robustness against common image processing, geometric distortions, unknown processing from social platforms and keeps the watermarked images with excellent imperceptibility simultaneously.

Convolutional Neural Networks (CNNs) have made remarkable progress in steganalysis. However, they struggle to detect adversarial steganography accurately which merges adversarial samples and steganography. While handcrafted models show limited vulnerability to adversarial steganography, their accuracy pales in comparison to that of CNN analyzers. To address these limitations head-on, we propose TStegNet, an innovative two-stream CNN steganalyzer designed to detect adversarial steganography. TStegNet leverages confidence artifacts and pixel artifacts, enabling a comprehensive analysis of hidden information. Specifically, we design a confidence loss function and apply backpropagation to amplify the confidence artifacts, which enhances the performance of our model. Additionally, we use the feature similarity function to minimize the impact of adversarial perturbation. Extensive experiments reveal that proposed TStegNet outperforms existing state-of-the-art methods, representing a significant milestone in the fight against adversarial steganography.

The need for cancelable biometric techniques has seen a progressive rise due to the rapid deployment of biometric authentication systems. These techniques prevent compromising biometric data by generating and using their corresponding cancelable templates for user authentication. However, the non-invertible distance preserving transformation methods employed in various schemes are often vulnerable to information leakage since matching is performed in the transform domain. This paper proposed a non-invertible distance preserving scheme based on vector permutation and shift-order process. First, the dimension of feature vectors is reduced using kernelized principal component analysis before randomly permuting the extracted vector features. A shift-order process is then applied to the generated features to achieve non-invertibility and combat similarity correlation-based attacks. The generated hash codes are resilient to various security and privacy attacks such as ARM, masquerade, and brute-force preimage. Experimental evaluations conducted on eight fingerprint datasets from FVC2002, FVC2004, and FVC2006 reveal a high matching performance of the proposed method with better recognition accuracy than other existing state-of-the-art. The scheme also fulfills the revocability and unlinkability requirements of cancelable biometrics.

Lots of non-additive cost assignment methods designed for image steganography have improved the security of stego images, but surprisingly there are only a few such non-additive cost assignment methods for video steganography. In this paper, we first analyze the distortion propagation by decomposing it into inner-block, inter-block, and inter-frame distortion drifts. Then, we determine the inner-block distortion drift (caused by the embedding modifications) that induces the inter-block and the inter-frame distortion drifts, using prediction. Based on the findings, we compose a joint distortion for all transform coefficients in each transform block. Finally, we propose a joint distortion-based non-additive cost assignment (NACA) method to reduce the inner-block distortion drift by distortion compensation. This allows us to further reduce both intra-frame (inter-block) and inter-frame distortion drifts, and achieve enhanced security. We conduct extensive experiments to evaluate the performance of NACA, in terms of security and coding performance. The evaluation results demonstrate that NACA achieves improved security and visual stego video quality, and maintains a very marginal increase in bit-rate, in comparison to four other competing additive cost assignment approaches.

In recent years, online video sharing platforms have been widely available on social networks. To protect copyright and track the origins of these shared videos, some video watermarking methods have been proposed. However, their robustness performance is significantly degraded under geometric deformations, which destroy the synchronization between the watermark embedding and extraction. To this end, we propose a novel robust blind video watermarking scheme by embedding the watermark into low-order recursive Zernike moments. To reduce the time complexity, we give an efficient computation method by exploring the characteristics of video and moments. The moment accuracy is greatly improved due to the introduction of a recursive computation method. Furthermore, we design an optimization strategy to enhance visual quality and reduce distortion drift of watermarked videos by analyzing the radial basis function. The robustness of the proposed scheme is verified by different attacks, including geometric deformations, length-width ratio changes, temporal synchronization attacks, and combined attacks. In practical applications, the proposed scheme effectively resists processing from video sharing platforms and screenshots taken with smartphones and PC monitors. The watermark is extracted without the host video. Experimental results show that our proposed scheme outperforms other state-of-the-art schemes in terms of imperceptibility and robustness.

Adaptive image steganography is the process of embedding secret messages into undetectable regions of a cover image through the design of a distortion function by a steganographer. Since the state-of-the-art steganalyzers are mainly based on image residual analysis, it is reasonable to modify stego image for withstanding steganalysis by reducing or eliminating the image residual distance between cover and stego image. However, simply modifying stego images may lead to message extraction failure and the introduction of additional detectable artifacts. In this paper, we propose a novel secure steganography strategy by constructing immunized stego-image via an artificial immune system, called ISteg, which ensures the accurate extraction of hidden data while enhancing the security against steganalyzers. Inspired by the biological immune system, we use an artificial immune system (AIS) to build ISteg. Specifically, ISteg generates the immunized stego-image by automatically modifying the stego to maximize the affinity of the antibody. The affinity is developed to evaluate antibody quality according to the Euclidean distance between the residual co-occurrence matrix features of the cover image and the modified stego image. In this manner, the so-called immunized stego-image is generated. Extensive experimental results demonstrate that the proposed ISteg strategy can effectively improve the security performance of existing steganography.

This article focuses on designing an event-triggered impulsive fault-tolerant control strategy for the stabilization of memristor-based reaction-diffusion neural networks (RDNNs) with actuator faults. Different from the existing memristor-based RDNNs with fault-free environments, actuator faults are considered here. A hybrid event-triggered and impulsive (HETI) control scheme, which combines the advantages of event-triggered control and impulsive control, is newly proposed. The hybrid control scheme can effectively accommodate the actuator faults, save the limited communication resources, and achieve the desired system performance. Unlike the existing Lyapunov-Krasovskii functionals (LKFs) constructed on sampling intervals or required to be continuous, the introduced LKF here is directly constructed on event-triggered intervals and can be discontinuous. Based on the LKF and the HETI control scheme, new stabilization criteria are derived for memristor-based RDNNs. Finally, numerical simulations are presented to verify the effectiveness of the obtained results and the merits of the HETI control method.

This article focuses on the design of a mode- dependent adaptive event-triggered control (AETC) scheme for the stabilization of Markovian memristor-based reaction-diffusion neural networks (RDNNs). Different from the existing works with completely known transition probabilities, partly unknown transition probabilities (PUTPs) are considered here. The switching conditions and values of memristive connection weights are all correlated with Markovian jumping. A mode-dependent AETC scheme is newly proposed, in which different adaptive event-triggered mechanisms will be applied for different Markovian jumping modes and memristor switching modes. For each given mode, the corresponding event-triggered mechanism can efficiently reduce the number of transmission signals by adaptively adjusting the threshold. Thus, the mode-dependent AETC scheme can effectively save the limited network communication resources for the considered system. Based on the proposed control scheme, a new stabilization criterion is set up for Markovian memristor-based RDNNs with PUTPs. Meanwhile, a memristor-dependent AETC scheme is devised for memristor-based RDNNs. Finally, simulation results are presented to verify the effectiveness and superiority of the analysis results.

In recent years, with the significant improvement of Gener-ative Adversarial Networks (GANs), fake images generated by GAN become hardly distinguishable from real ones, thus threatening the authentication of digital images. To resolve this issue, several fake image detectors based on supervised binary classification have been designed. However, current methods remain vulnerable when testing samples are gener-ated by an unknown GAN model. In this work, an unsuper-vised domain adaptation strategy is introduced to improve the performance in the generalization of GAN-generated image detection by using a small number of unlabeled images from the target domain. Self-Attention block and novel loss function have been constructed to optimize the domain adaptation process, thus getting a better generalization. Experimental results demonstrate that the proposed scheme achieves high detection accuracy with few unlabeled images in the target domain, which shows that unsupervised methods can be used for the detection of GAN-generated images.

It is increasingly common for people to share videos on social platforms such as TikTok, YouTube and bilibili. The shared videos, however, is usually affected by lossy channel compression, thus a robust video watermarking based on discrete sine transform (DST) is proposed. Firstly, We divide each video frame into blocks and perform DST in embedding region, then the high-frequency component of DST is selected for watermark embedding. For further strengthening the robustness of lossy channel compression, a signed odd-even interval quantization method is designed by improving the odd-even quantization method. As a consequence of uncomplicated feature of DST, the computational efficiency of this algorithm is efficient, making it ideal for application in real-time scenarios such as video live broadcast and video conference. Moreover, the watermark is extracted without the host video. Experimental results show that the proposed algorithm not only ensures excellent video quality, but also significantly improves the robustness of lossy channel compression, and the robustness to different video contents is stable.

Existing deepfake detection methods have achieved high accuracy under the intra-database scenario, but most of them suffer from a significant performance drop when evaluated on cross-database experiments. In this paper, we regard the deepfake detection task as domain generalization problem and design a Facial Decomposition-based Domain Generalization framework to learn a more generalized feature representation. To optimize our framework, we experimentally analysis that current CNN-based detectors tend to overfit the facial semantic content but neglect the common shared traces of deepfake, thus a facial semantic content decomposition based dual-branch network is designed to reduce the representation discrepancies among multiple source domains in a specific feature space to be domain invariant. Besides, based on the consideration that the distribution discrepancies are much larger among the fake faces than the real ones, a hybrid loss function is designed to enlarge the distance among samples from different categories even in cross-domain scene. Extensive experiments demonstrate that the proposed method can achieve better performance compared with state-of-the-art methods, especially on cross-database evaluation.

Designing an efficient synchronization method for video watermarking under complex hybrid attacks is an urgent task due to the development of social networks and multimedia software. However, few existing watermarking methods can resist various complex attacks on the social network, such as scaling, frame rate conversion, and other strong hybrid attacks. To address this issue, we first analyze the relationship between the spatial and discrete cosine transform (DCT) domain and propose a robust video watermarking based on residual synchronization (RS) mechanism. In the RS mechanism, the spread-spectrum encoded RS unit and sync frames are generated according to the change rules obtained from the analysis to achieve robustness. On the other hand, the high-frequency energy average distribution (HFEAD) mechanism is designed by analyzing the effects of distortion drifts and texture features for watermark embedding to achieve good visual quality. Extensive experimental results show that the RS mechanism can effectively synchronize watermarks under various attacks, e.g., asymmetric cropping, resizing, scaling, etc., hybridized with strong recompression attacks. Moreover, the proposed scheme is resistant to screenshots and social platform transcoding.

The cost function is crucial to the security of adaptive image steganography, However, some existing cost functions are heuristically designed and hard to be optimal in the undetectability against the evolving steganalyzer. In this letter, we propose a cost reassignment algorithm for adaptive steganography based on artificial immune system. Under the scenario of minimizing additive distortion, this method reassigns the cost by adjusting the modification probability distribution obtained by the cost function, and dynamically optimizes the adjustment mode through the immunity-based information hiding model, so that the modified pixels are more concentrated in the regions that are difficult to be detected. The experimental results show that the proposed method is suitable for a variety of the state-of-the-art cost functions and can achieve better performance on resisting the steganalysis.

Recently, researchers have been dedicated to discovering Generative Adversarial Network (GAN) artifacts and using them to identify generated images. However, current approaches exhibit restricted performance when testing against unseen GAN models, which is also known as the cross-domain scenario. To overcome this limitation, we propose a novel GAN-generated image detection framework by estimating artifact similarity, which is inspired by relation network. The proposed method consists of two stages, including representation learning and representation comparison. For representation learning, ResNet-50 equipped with Instance Normalization in the Shallow layers (ResNet-INS) is constructed as the embedding network to extract generalized features. For representation comparison, Category and Domain-Aware loss function (CDA loss) is designed by leveraging both category and domain information efficiently, which can enlarge inter-class discrepancy of different categories (GAN-generated or pristine images) and improve intra-class compactness from different domains (source attributions) in the same category. Extensive experiments are conducted which consider various cross-domain scenarios to verify the generalization of the proposed method. Besides, our method exhibits satisfying robustness against common post-processings, even when data augmentation is not considered during the training stage.

Audio fingerprinting methods can compress audio contents into compact signatures so that we can save storage and reduce query time. This technology is widely used in many fields, such as audio retrieval, music information retrieval and audio authentication. However, most of the existing methods cannot balance the recognition accuracy, query speed and storage size well. This letter presents a novel self-supervised learning scheme called asymmetric contrastive learning to generate binary hash fingerprints of audio segments. Meanwhile, we design a new loss function named bidirectional asymmetric pairwise loss to minimize the loss of information. Experimental results show that our scheme can achieve a high top-1 hit rate on both music and speech datasets. Furthermore, the proposed scheme outperforms the previous work of real-value fingerprinting in query speed and storage size.

Cost assignment plays a key role in coding performance and security of video steganography. Existing cost assignment methods (for adaptive video steganography) are designed for specific transform coefficients rather than all transform coefficients. In addition, existing video steganographic frameworks do not allow Syndrome-Trellis Codes (STCs) to modify all transform coefficients in both intra-coded and inter-coded frames at the same time. To address these limitations, in this article, we first propose a novel video steganographic framework. Then, we give a theoretical analysis of distortion drift in both intra- and inter-coding procedures. Based on the analysis, we design a Distortion Drift-Based Cost Assignment method, hereafter referred to as DDCA. DDCA considers the inner-block, inter-block and inter-frame distortion costs in order to improve the coding performance and the security of stego videos when the embedding payload is fixed. We conducted extensive experiments using two video datasets to evaluate the proposed video steganographic framework and DDCA, in terms of the coding performance and the security. Our experiments show that the proposed framework outperforms three recent state-of-the-art methods, for example the coding performance and the security of stego videos can benefit from DDCA by making full use of all nonzero transform coefficients.

This article focuses on the design of fuzzy secure control for a class of coupled systems, which are modeled by a nonlinear$N$-dimensional ($N$-D) parabolic partial differential equation (PDE) subsystem and an ordinary differential equation (ODE) subsystem. Under stochastic deception attacks, a fuzzy secure control scheme is designed, which is effective to tolerate the attacks and ensure the desired performance for the considered systems. A new fuzzy-dependent Poincare–Wirtinger’s inequality (PWI) is proposed. Compared with the traditional Poincare’s inequality, the fuzzy-dependent PWI is more flexible and less conservative. Meanwhile, an augmented Lyapunov–Krasovskii functional (LKF) is newly constructed, which strengthens the correlations of the PDE subsystem and ODE subsystem. Then, on the ground of the fuzzy-dependent PWI and the augmented LKF, new exponential stabilization criteria are set up for the PDE-ODE coupled systems. Finally, a hypersonic rocket car is presented to verify the effectiveness and less conservatism of the obtained results.

This study focuses on the quasisynchronization problem for reaction–diffusion neural networks (RDNNs) in the presence of deception attacks. Under deception attacks, a time–space sampled-data (TSSD) control mechanism is proposed for RDNNs. Compared with traditional control strategies, the proposed control mechanism can not only save network bandwidth but also improve the cybersecurity of communications. Inspired by Halanay’s inequality, a new inequality is proposed, which can be effectively applied to the quasisynchronization problem for dynamical systems. Then, by using this inequality and the Lyapunov functional approach, quasisynchronization criteria are set for RDNNs. The desired control gain is gained from solving a group of linear matrix inequalities. Moreover, in the absence of deception attacks, the exponential synchronization problem is studied for RDNNs. In the end, simulation results are given to demonstrate the usefulness of the theoretical analysis.

Abstract Data hiding aims to embed a secret message into a digital object such as image by slightly modifying the object content without arousing noticeable artefacts. The resultant object containing hidden information will be sent to a desired receiver via some insecure channels, e.g. images transmitted through noisy channel, social networks are vulnerable to unknown pollution or compression by a third party, which may lead the transmitted objects to be attacked such that the reconstructed message has a significant error rate. It therefore requires us to use robust embedding strategies for data hiding to realise reliable message retrieval. To this end, in this paper, a metric model to estimate the robustness of data hiding for noisy transmission based on the statistical characteristics of cover and embedding operation is presented, the former is mainly reflected by spatial frequency and texture feature, and the latter embedding operation is mainly reflected by embedding modification. The goal is to ensure that both statistical characteristics and embedding operation can be used to maximise the embedding robustness. To the best knowledge, it is the first time to estimate robustness before data hiding by a special metric model. Experimental results show that, by combining the proposed metric model in three classical data hiding methods, i.e. BPS, DE and QIM, the robustness can be significantly improved, which demonstrates its superiority and applicability.

Nowadays, verifying the integrity of digital videos is significant especially for applications about multimedia communication. In video forensics, detection of double compression can be treated as the first step to analyze whether a suspicious video undergoes any tampering operations. In the last decade, numerous detection methods have been proposed to address this issue, but most existing methods design a universal detector which is hard to handle various recompression settings efficiently. In this work, we found that the statistics of different Coding Unit (CU) types have dissimilar properties when original videos are recompressed by the increased and decreased bit rates. It motivates us to propose a two-stage cascaded detection scheme for double HEVC compression based on temporal inconsistency to overcome limitations of existing methods. For a given video, CU information maps are extracted from each short-time video clip using our proposed value mapping strategy. In the first detection stage, a compact feature is extracted based on the distribution of different CU types and Kullback–Leibler divergence between temporally adjacent frames. This detection feature is fed into the Support Vector Machine classifier to identify abnormal frames with the increased bit rate. In the second stage, a shallow convolutional neural network equipped with dense connections is designed carefully to learn robust spatiotemporal representations, which can identify abnormal frames with the decreased bit rate whose forensic traces are less detectable. In experiments, the proposed method can achieve more promising detection accuracy compared with several state-of-the-art methods under various coding parameter settings, especially when the original video is recompressed with a low quality (e.g., more than 8%).

With the increasing popularity of digital video communication, video data hiding has become an active research topic in covert communication and privacy protection. Traditional video data hiding methods often use quantized discrete cosine transform (QDCT) coefficients to carry a sufficient payload. However, since QDCT coefficients expose texture features and motion characteristics of the present video frame heavily, data embedding with QDCT coefficients may lead to significant intra-frame distortion and inter-frame distortion drift. To avoid obvious visual artifacts and keep bit-rate within a satisfactory level of the marked video, data embedding in QDCT coefficients should take into account both the intra-frame and inter-frame distortion impacts. It motivates the authors to propose an efficient cost assignment-based video data hiding method in this paper. The proposed cost assignment method aims to accurately evaluate the data embedding distortion. Specifically, the proposed scheme considers intra-frame changes and intra-frame distortion drift, for which the texture and motion changes of frames can be measured. The frame position is also used to reflect a cumulative distortion difference of multiple frames. For data embedding, syndrome-trellis code (STC) is adopted to minimize the overall distortion. Experimental results show that the proposed method significantly outperforms existing works in terms of payload-distortion performance.

Detection of double compression is regarded as one primary step in analyzing the integrity of digital videos, which is of prominent importance in video forensics. However, current methods are vulnerable with the severe lossy quantization in the recompression process such that it is challenging to obtain reliable frame-wise detection results, especially for the high efficiency video coding (HEVC) standard. In view of these issues, in this paper, a hybrid neural network is proposed to reveal abnormal frames in HEVC videos with double compression by learning robust spatio-temporal representations from coding information in the compression domain. Based on the statistical analysis of Coding Units (CUs), it is interesting to find that HEVC video streams contain “rich” coding information that could be leveraged to identify abnormal traces caused by double compression. Two types of coding information maps, including CU Size Map (CSM) and CU Prediction mode Map (CPM), are exploited. In contrast with the conventional paradigm relying on pixel-level representations of decoded frames, CSMs and CPMs of a short-time video clip are treated as the input, aiming to achieve high robustness against recompression of low quality. In our hybrid neural network, an attention-based two-stream residual network is proposed to learn hierarchical representations from CSM and CPM, which are then jointly optimized by the attention-based fusion module. Finally, the temporal variation is modeled by Long Short-Term Memory (LSTM) to obtain frame-wise detection results. We have conducted extensive experiments considering various video content and coding parameters, such as bitrates and sizes of Group of Picture. Experimental results show that our approach can obtain state-of-the-art performance compared with conventional methods, especially when videos are recompressed in the low bitrate coding scenarios.

Bitrate is generally regarded as an important criterion of video quality. However, with sophisticated video editing software, forgers can create fake bitrate videos by up-converting the bitrate of original videos with lower video quality to attract more viewers on video sharing websites. In this work, we first model the generation process of fake bitrate videos and analyze the dominant sources of information loss. It is found that the recompression error generated by the proposed one-step-further recompression operation is an efficient measurement to expose distinguishable quality variation tendencies between true and fake bitrate videos. Based on this analysis, we propose a detection method for fake bitrate videos using a hybrid deep-learning network from recompression error. For an input video, the patch-wise recompression errors are first calculated to increase the learning capability of the network. To learn robust representations of recompression errors in local regions with different degrees of predictability, a hybrid deep-learning network that contains two branches with heterogeneous structures is designed. For noise-like recompression errors, the first branch has a shallow CNN structure initialized with an Inception-like module using multisize convolutional kernels. For zero-element clustered recompression errors, the second branch has a multi-layer perceptron structure equipped with a unique layer that extracts the histogram of zero-element clustered square regions. The output vectors of different branches are concatenated and then jointly optimized to obtain the patch-wise detection results. Finally, the majority voting (local-to-global) strategy is applied to obtain the final detection result. Extensive experiments are conducted to evaluate the detection performance under various coding parameter settings, such as different bitrates, rate-distortion optimization strategies and so on. The experimental results demonstrate the superiority of the proposed method compared with several state-of-the-art methods to provide more fine-grained forensic clues.

Biometric image hashing techniques have been widely studied and seen progressive advancements. However, only a handful of available solutions provide two-factor cancelability while simultaneously satisfying the tradeoff among all criteria of template protection mechanisms. In this paper, we propose a novel scheme for generating a secure and robust hash from a fingerprint image using Fourier-Mellin transform and fractal coding. First, due to its invariance property, Fourier-Mellin transform is incorporated into the domain fingerprint minutiae blocks to provide feature alignment, therein generating a fixed-length minutiae representation for comparison. Then, dimensionality reduction and texture compression are exploited using fractal coding to generate a robust and compact hash for improved security and recognition. The experimental results demonstrate a favorable recognition performance on benchmarked state-of-the-art schemes from FVC2002 and FVC2004 fingerprint databases. The analyses prove our method's robustness and resiliency to security and privacy attacks. Our method also satisfies the revocability and unlinkability criteria of cancelable biometrics.

Recently, the success of generating fake images by Generative Adversarial Network (GAN) has threatened the authentication of digital images. To address this issue, several automated fake image detectors have been proposed. However, current methods remain vulnerable when testing samples undergo post-processing attacks. In this work, we employed residual signals of chrominance components from multi color spaces, including YCbCr, HSV and Lab, to learn robust deep representations via the well-designed shallow convolutional neural network (CNN). Then, the learned deep representations from different color spaces are concatenated and then fed into the Random Forest (RF), which is the widely used ensemble classifier, to obtain final detection results. Extensive experiments are conducted on the fake image dataset generated by the advanced GAN technique. Experimental results demonstrate the proposed scheme outperforms state-of-the-art methods and achieves the promising average detection accuracy (above 99%) under several post-processing attacks, such as Gaussian blurring and so on.

In this paper, we propose a novel scheme for generating a secure and robust hash using Fourier-mellin transform and fractal coding. First, Fourier-mellin transform is incorporated into the domain minutiae blocks due to its invariance property in order to improve performance under geometric operations, hence generating a fixed-length minutiae representation. Then the property of dimensionality reduction and texture compression is exploited using fractal coding in order to generate a robust and compact hash. To secure the system, encryption is performed on the extracted hash using a secret key. Experimental results demonstrated the robustness of our hashing scheme to a wide range of distortion manipulations. Additionally, performance of the proposed scheme is measured and compared with recent state-of-art techniques, and our scheme generally outperform the others.

The conventional reversible data hiding (RDH) algorithms often consider the host as a whole to embed a secret payload. In order to achieve satisfactory rate-distortion performance, the secret bits are embedded into the noise-like component of the host such as prediction errors. From the rate-distortion optimization view, it may be not optimal since the data embedding units use the identical parameters. This motivates us to present a segmented data embedding strategy for efficient RDH in this paper, in which the raw host could be partitioned into multiple subhosts such that each one can freely optimize and use the data embedding parameters. Moreover, it enables us to apply different RDH algorithms within different subhosts, which is defined as ensemble. Notice that, the ensemble defined here is different from that in machine learning. Accordingly, the conventional operation corresponds to a special case of the proposed work. Since it is a general strategy, we combine some state-of-the-art algorithms to construct a new system using the proposed embedding strategy to evaluate the rate-distortion performance. Experimental results have shown that, the ensemble RDH system could outperform the original versions in most cases, which has shown the superiority and applicability.

This paper presents a reference sharing mechanism-based self-embedding watermarking scheme. The host image is embedded with watermark bits including the reference data for content recovery and the authentication data for tampering location. The special encoding matrix derived from the generator matrix of selected systematic Maximum Distance Separable (MDS) code is adopted. The reference data is generated by encoding all the representative data of the original image blocks. On the receiver side, the tampered image blocks can be located by the authentication data. The reference data embedded in one image block can be shared by all the image blocks to restore the tampered content. The tampering coincidence problem can be avoided at the extreme. The maximal tampering rate is deduced theoretically. Experimental results show that, as long as the tampering rate is less than the maximal tampering rate, the content recovery is deterministic. The quality of recovered content does not decrease with the maximal tampering rate.

This paper proposes a blind authentication scheme to identify duplicated regions for copy-move forgery based on perceptual hashing and package clustering algorithms. For all fixed-size image blocks in suspicious images, discrete cosine transform (DCT) is used to obtain their DCT coefficient matrixes. Their perceptual hash matrixes and perceptual hash feature vectors are orderly addressed. Moreover, a package clustering algorithm is proposed to replace traditional lexicographic order algorithms for improving the detection precision. Similar blocks can be identified by matching the perceptual hash feature vectors in each package and its adjacent package. The experimental results show that the proposed scheme can locate irregular tampered regions and multiple duplicated regions in suspicious images although they are distorted by some hybrid trace hiding operations, such as adding white Gaussian noise and Gaussian blurring, adjusting contrast ratio, luminance, and hue, and their hybrid operations.

Reversible data hiding (RDH) into encrypted images is of increasing attention to researchers as the original content can be perfectly reconstructed after the embedded data are extracted while the content owner's privacy remains protected. The existing RDH techniques are designed for grayscale images and, therefore, cannot be directly applied to palette images. Since the pixel values in a palette image are not the actual color values, but rather the color indexes, RDH in encrypted palette images is more challenging than that designed for normal image formats. To the best knowledge of the authors, there is no suitable RDH scheme designed for encrypted palette images that has been reported, while palette images have been widely utilized. This has motivated us to design a reliable RDH scheme for encrypted palette images. The proposed method adopts a color partitioning method to use the palette colors to construct a certain number of embeddable color triples, whose indexes are self-embedded into the encrypted image so that a data hider can collect the usable color triples to embed the secret data. For a receiver, the embedded color triples can be determined by verifying a self-embedded check code that enables the receiver to retrieve the embedded data only with the data hiding key. Using the encryption key, the receiver can roughly reconstruct the image content. Experiments have shown that our proposed method has the property that the presented data extraction and image recovery are separable and reversible. Compared with the state-of-the-art works, our proposed method can provide a relatively high data-embedding payload, maintain high peak signal-to-noise ratio values of the decrypted and marked images, and have a low computational complexity.

We propose to utilize the prediction-error of prediction error (PPE) of a pixel to reversibly carry the secret data in this letter. In the proposed method, the pixels to be embedded are firstly predicted with their neighboring pixels to obtain the prediction errors (PEs). By exploiting the PEs of the neighboring pixels, the prediction of the PEs of the pixels to be embedded can be then determined. And, a sorting technique based on the local complexity of a pixel is used to collect the PPEs to generate an ordered PPE sequence so that, smaller PPEs will be processed first for data embedding. By reversibly shifting the PPE histogram (PPEH) with optimized parameters, the pixels corresponding to the altered PPEH bins can be finally modified to carry the entire secret data. Experimental results have implied that, the proposed algorithm can benefit from the prediction procedure, sorting technique as well as parameters selection, and therefore outperform some state-of-the-art works in terms of payload-distortion performance.

This paper presents a method for detection of motion vector-based video steganography. First, the modification on the least significant bit of the motion vector is modeled. The influence of the embedding operation on the sum of absolute difference (SAD) is illustrated, which allows us to focus on the difference between the actual SAD and the locally optimal SAD after the adding-or-subtracting-one operation on the motion value. Finally, based on the fact that most motion vectors are locally optimal for most video codecs, two feature sets are extracted and used for classification. Experiments are carried out on videos corrupted by various steganography methods and encoded by various motion estimation methods, in various bit rates, and in various video codecs. Performance results demonstrate that our scheme outperforms previous works in general, and is more favorable for real-world applications.