The widespread application of AIGC contents has brought not only unprecedented opportunities, but also potential security concerns, e.g., audio-visual deepfakes. Therefore, it is of great importance to develop an effective and generalizable method for multi-modal deepfake detection. Typically, the audio-visual correlation learning could expose subtle cross-modal inconsistencies, e.g., audio-visual misalignment, which serve as crucial clues in deepfake detection. In this paper, we reformulate the correlation learning with variational Bayesian estimation, where audio-visual correlation is approximated as a Gaussian distributed latent variable, and thus develop a novel framework for deepfake detection, i.e., Forgery-aware Audio-Visual Adaptation with Variational Bayes (FoVB). Specifically, given the prior knowledge of pre-trained backbones, we adopt two core designs to estimate audio-visual correlations effectively. First, we exploit various difference convolutions and a high-pass filter to discern local and global forgery traces from both modalities. Second, with the extracted forgery-aware features, we estimate the latent Gaussian variable of audio-visual correlation via variational Bayes. Then, we factorize the variable into modality-specific and correlation-specific ones with orthogonality constraint, allowing them to better learn intra-modal and cross-modal forgery traces with less entanglement. Extensive experiments demonstrate that our FoVB outperforms other state-of-the-art methods in various benchmarks.

Image forgery localization, which aims to find suspicious tampered regions by splicing, copy-move, or removal manipulations, has attracted increasing attention. Although considerable progress has been made, most of the existing methods are still far from satisfactory in terms of generalization, e.g., cross-dataset evaluation, and show less robustness against lossy distortion, e.g., post-processing attacks or transmission over online social networks. To address these challenges, a general and robust image forgery localization framework using a diffusion probabilistic model is proposed in this paper. First, aDiffusion-drivenForgery-awareLocalization network (DFL) is presented. In specific, the task of image forgery localization is formulated as the one of mask reconstruction with the forgery-aware features as conditional prior, which introduces forgery-related knowledge into the diffusion process to gradually recover the ground-truth mask from the noisy one. AnEdge-guidedDiffusionRestoration network (EDR) is then developed and integrated into the DFL network, leading to theEDR-DFLmodel, boosting the performance against various post-processing attacks. With EDR-DFL, the distorted forged images could be restored using the EDR model in an edge feature preserving way, which allows the DFL model to localize the tampered regions in the restored images effectively. Extensive experimental results demonstrate that the proposed method significantly outperforms other state-of-the-art methods for cross-dataset evaluation and exhibits superior robustness against various challenging post-processing attacks.

Image forgery detection and localization (IFDL) is of vital importance as forged images can spread misinformation that poses potential threats to our daily life. However, previous methods still struggled to effectively handle forged images processed with diverse forgery operations in real-world scenarios. In this paper, we propose a novel Reinforced Multi-teacher Knowledge Distillation (Re-MTKD) framework for the IFDL task, structured around an encoder-decoder ConvNeXt-UperNet along with Edge-Aware Module, named Cue-Net. First, three Cue-Net models are separately trained for the three main types of image forgeries, i.e., copy-move, splicing and inpainting, which then serve as the multi-teacher models to train the target student model with Cue-Net through self-knowledge distillation. A Reinforced Dynamic Teacher Selection (Re-DTS) strategy is developed to dynamically assign weights to the involved teacher models, which facilitates specific knowledge transfer and enables the student model to effectively learn both the common and specific natures of diverse tampering traces. Extensive experiments demonstrate that, compared with other state-of-the-art methods, the proposed method achieves superior performance on several recently emerged datasets comprised of various kinds of image forgeries.

With the increasing reliance on the biometric-based authentication systems, such as face recognition, in applications within the IoT and edge networks, guaranteeing proper service functionality while safeguarding individual biometric privacy has become a critical concern. However, most existing face privacy protection approaches mainly focus on preserving the machine-recognizable identity information, inadvertently compromising individual privacy. To tackle this challenge, a novel Face Privacy-Enhancing Network (FPE-Net) is proposed, which consists of two primary stages: biometric encryption and face reconstruction. Specifically, a linear encryption module is designed in the first stage for obfuscating the original identity information, which is later integrated into the depth features of the target face via an identity injector. Notably, the identity encryption process operates independently of the deep generative network, enabling greater flexibility and efficiency for key configuration. Then in the second stage, a face decoder is utilized to synthesize the photo-realistic face. Moreover, such face not only prevents cross-matching with biometric databases but also preserves recognition utility, owing to the linear encryption mechanism and loss design. Extensive quantitative and qualitative experimental results demonstrate the feasibility of FPE-Net model, which outperforms existing state-of-the-art approaches in terms of privacy protection.

Reversible Adversarial Example (RAE) could be used to protect the privacy and copyright of images on social networks (SONs) by exploring the adversarial examples to disrupt the access of malicious AI models while ensuring recoverability with authorized users. Existing RAE methods add adversarial perturbations in spatial images which do not apply to JPEG images, the most widely adopted image format for image storage and transmission. To tackle this issue, we propose the first Reversible Adversarial Example (JPEG-RAE) generation framework for JPEG images, which consists of two primary components, i.e., JPEG-AE and G-RDH. JPEG-AE crafts the adversarial perturbations in the JPEG domain of images by leveraging chain rule of gradient propagation, so that they could effectively mislead the AI models in spatial domain when they are JPEG decompressed. And G-RDH adopts a gradient-directed bi-directional histogram shifting scheme for efficient reversible hiding of adversarial perturbations and location data in JPEG domain, where the histogram shifting is in sync with the sign of back-propagated gradients to further boost the performance of adversarial attacks. Experimental validation demonstrates that, although confined to the JPEG format such as the amount and intensity of alterable DCT coefficients, the proposed JPEG-RAE could still show superior or comparable performance, in terms of attack ability and recover ability, to its counterparts in spatial domain.

Existing Text Image Forgery Localization (T-IFL) methods often suffer from poor generalization due to the limited scale of real-world datasets and the distribution gap caused by synthetic data that fails to capture the complexity of real-world tampering. To tackle this issue, we propose Fourier Series-based Tampering Synthesis (FSTS), a structured and interpretable framework for synthesizing tampered text images. FSTS first collects 16,750 real-world tampering instances from five representative tampering types, using a structured pipeline that records human-performed editing traces via multi-format logs (e.g., video, PSD, and editing logs). By analyzing these collected parameters and identifying recurring behavioral patterns at both individual and population levels, we formulate a hierarchical modeling framework. Specifically, each individual tampering parameter is represented as a compact combination of basis operation–parameter configurations, while the population-level distribution is constructed by aggregating these behaviors. Since this formulation draws inspiration from the Fourier series, it enables an interpretable approximation using basis functions and their learned weights. By sampling from this modeled distribution, FSTS synthesizes diverse and realistic training data that better reflect real-world forgery traces. Extensive experiments across four evaluation protocols demonstrate that models trained with FSTS data achieve significantly improved generalization on real-world datasets. Dataset is available at \href{https://github.com/ZeqinYu/FSTS}{Project Page}.

With wide applications of image editing tools, forged images (splicing, copy-move, removal and etc.) have been becoming great public concerns. Although existing image forgery localization methods could achieve fairly good results on several public datasets, most of them perform poorly when the forged images are JPEG compressed as they are usually done in social networks. To tackle this issue, in this paper, a self-supervised domain adaptation network, which is composed of a backbone network with Siamese architecture and a compression approximation network (ComNet), is proposed for JPEG-resistant image forgery detection and localization. To improve the performance against JPEG compression, ComNet is customized to approximate the JPEG compression operation through self-supervised learning, generating JPEG-agent images with general JPEG compression characteristics. The backbone network is then trained with domain adaptation strategy to localize the tampering boundary and region, and alleviate the domain shift between uncompressed and JPEG-agent images. Extensive experimental results on several public datasets show that the proposed method outperforms or rivals to other state-of-the-art methods in image forgery detection and localization, especially for JPEG compression with unknown QFs.

In recent years, the popularity of digital media sharing, especially high-quality images through online social networks (OSNs) has spurred an increasing demand for digital rights management (DRM) with watermarking. Although the most recent watermarking schemes with deep networks have exhibited considerable performance improvement, they still fall short in resisting multiple attacks with high-fidelity watermarking. To tackle this issue, a customized framework with encoder/decoder structure is proposed in this letter, aiming to consistently improve the robustness performance against multiple attacks. In specific, theMulti-scaleSalientFeatureAttentionBlock(MSFABlock) is exploited to effectively extract the robust image features with the encoder and decoder by taking advantage of the salient features, e.g., the image features obtained with difference of Gaussian (DoG) and other gradient operators. In addition, an adaptive squared Hinge function is developed as message loss to encourage adaptive watermark embedding. Experimental results demonstrate excellent performance in terms of robustness and perceptual fidelity as well as high efficiency of the proposed scheme in comparison to other SOTA methods.

With the advancement of deepfake generation techniques, the importance of deepfake detection in protecting multimedia content integrity has become increasingly obvious. Recently, temporal inconsistency clues have been explored to improve the generalizability of deepfake video detection. According to our observation, the temporal artifacts of forged videos in terms of motion information usually exhibits quite distinct inconsistency patterns along horizontal and vertical directions, which could be leveraged to improve the generalizability of detectors. In this paper, a transformer-based framework forDiffusion Learning ofInconsistencyPattern (DIP) is proposed, which exploits directional inconsistencies for deepfake video detection. Specifically, DIP begins with a spatiotemporal encoder to represent spatiotemporal information. A directional inconsistency decoder is adopted accordingly, where direction-aware attention and inconsistency diffusion are incorporated to explore potential inconsistency patterns and jointly learn the inherent relationships. In addition, the SpatioTemporal Invariant Loss (STI Loss) is introduced to contrast spatiotemporally augmented sample pairs and prevent the model from overfitting nonessential forgery artifacts. Extensive experiments on several public datasets demonstrate that our method could effectively identify directional forgery clues and achieve state-of-the-art performance.

Hyper-realistic avatars in the metaverse have already raised security concerns about deepfake techniques; deepfakes involving generated video “recording” may be mistaken for a real recording of the people it depicts. As a result, deepfake detection has drawn considerable attention in the multimedia forensic community. Though existing methods for deepfake detection achieve fairly good performance under the intra-dataset scenario, many of them gain unsatisfying results in the case of cross-dataset testing with more practical value, where the forged faces in training and testing datasets are from different domains. To tackle this issue, in this article, we propose a novel Domain-Invariant and Patch-Discriminative feature learning framework—DI&PD. For image-level feature learning, a single-side adversarial domain generalization is introduced to eliminate domain variances and learn domain-invariant features in training samples from different manipulation methods, along with the global and local random crop augmentation strategy to generate more data views of forged images at various scales. A graph structure is then built by splitting the learned image-level feature maps, with each spatial location corresponding to a local patch, which facilitates patch representation learning by message-passing among similar nodes. Two types of center losses are utilized to learn more discriminative features in both image-level and patch-level embedding spaces. Extensive experimental results on several datasets demonstrate the effectiveness and generalization of the proposed method compared with other state-of-the-art methods.

The recent advances in generative image steganography have drawn increasing attention due to their potential for provable security and bulk embedding capacity. However, existing generative steganographic schemes are usually tailored for specific tasks and are hardly applied to applications with practical constraints. To address this issue, this paper proposes a generic generative image steganography scheme called Steganography StyleGAN (StegaStyleGAN) that meets the practical objectives of security, capacity, and robustness within the same framework. In StegaStyleGAN, a novel Distribution-Preserving Secret Data Modulator (DP-SDM) is used to achieve provably secure generative image steganography by preserving the data distribution of the model inputs. Additionally, a generic and efficient Secret Data Extractor (SDE) is invented for accurate secret data extraction. By choosing whether to incorporate the Image Attack Simulator (IAS) during the training process, one can obtain two models with different parameters but the same structure (both generator and extractor) for lossless and lossy channel covert communication, namely StegaStyleGAN-Ls and StegaStyleGAN-Ly. Furthermore, by mating with GAN inversion, conditional generative steganography can be achieved as well. Experimental results demonstrate that, whether for lossless or lossy communication channels, the proposed StegaStyleGAN can significantly outperform the corresponding state-of-the-art schemes.

As manipulating images may lead to misinterpretation of the visual content, addressing the image forgery detection and localization (IFDL) problem has drawn serious public concerns. In this work, we propose a simple assumption that the effective forensic method should focus on the mesoscopic properties of images. Base on the assumption, a novel two-stage self-supervised framework leveraging the diffusion model for IFDL task, i.e., DiffForensics, is proposed in this paper. The DiffForensics begins with self-supervised denoising diffusion paradigm equipped with the module of encoder-decoder structure, by freezing the pre-trained encoder (e.g., in ADE-20K) to inherit macroscopic features for general image characteristics, while encour-aging the decoder to learn microscopic feature represen-tation of images, enforcing the whole model to focus the mesoscopic representations. The pre-trained model as a prior, is then further fine-tuned for IFDL task with the customized Edge Cue Enhancement Module (ECEM), which progressively highlights the boundary features within the manipulated regions, thereby refining tampered area local-ization with better precision. Extensive experiments on several public challenging datasets demonstrate the effectiveness of the proposed method compared with other state-of-the-art methods. The proposed DiffForensics could significantly improve the model's capabilities for both accurate tamper detection and precise tamper localization while con-currently elevating its generalization and robustness.

With wide applications of image editing tools, forged images have become a great public concern. Although existing methods for image forgery localization (IFL) could achieve fairly good results on several public datasets, most of them perform unsatisfactorily for cross-dataset evaluation and online social network applications. To tackle this issue, a novel coarse-to-fine framework using Diffusion probabilistic model for Image Forgery Localization (Diff-IFL) is proposed in this paper, which consists of a coarse localization module and a mask diffusion module. The coarse localization module employs a transformer-based architecture to represent the tampered images and generate coarse masks. While the mask diffusion module formulates IFL as a mask reconstruction task, it relies on the extracted forgery feature representations as the conditional prior to gradually recover the clean ground-truth mask from the noisy mask. Extensive experiments demonstrate that Diff-IFL outperforms other SOTA methods and exhibits superior robustness against social media forgery.

The mission of cloth-changing person re-identification (CC-ReID) is to discover cloth-invariant and identity-related cues, while traditional person ReID methods rely on appearance features that are biased to cloth-related cues. To tackle this cloth-biased problem, many CC-ReID methods introduced auxiliary body shape information to extract cloth-invariant features, such as 2D sketch images or the 3D Skinned Multi-Person Linear (SMPL) model. However, 2D auxiliary information lacks 3D spatial features, while the 3D SMPL model encounters challenges in capturing features at a finer granularity due to manually defined parameters. To extract fine-grained 3D shape features, we estimate depth maps that contain richer shape information and propose a Fine-grained Depth feature Mining and Distillation (FDMD) framework. We introduce a depth branch and design a fine-grained local feature interaction module to mine fine-grained 3D body shape knowledge from estimated depth maps by exploring the context of semantic-aware local body-part features. To integrate cloth-invariant depth knowledge into the appearance features, the fine-grained 3D shape features are transferred to an appearance branch by feature-space-aligned distillation. Extensive experiments demonstrate that FDMD can achieve state-of-the-art performance on three widely used CC-ReID benchmarks PRCC, Celeb-reID and LaST.

Nowadays, the abuse of AI-generated content (AIGC), especially the facial images known as deepfake, on social networks has raised severe security concerns, which might involve the manipulations of both visual and audio signals. For multimodal deepfake detection, previous methods usually exploit forgery-relevant knowledge to fully finetune Vision transformers (ViTs) and perform cross-modal interaction to expose the audio-visual inconsistencies. However, these approaches may undermine the prior knowledge of pretrained ViTs and ignore the domain gap between different modalities, resulting in unsatisfactory performance. To tackle these challenges, in this paper, we propose a new framework, i.e., Forgery-aware Audio-distilled Multimodal Learning (FRADE), for deepfake detection. In FRADE, the parameters of pretrained ViT are frozen to preserve its prior knowledge, while two well-devised learnable components, i.e., the Adaptive Forgery-aware Injection (AFI) and Audio-distilled Cross-modal Interaction (ACI), are leveraged to adapt forgery relevant knowledge. Specifically, AFI captures high-frequency discriminative features on both audio and visual signals and injects them into ViT via the self-attention layer. Meanwhile, ACI employs a set of latent tokens to distill audio information, which could bridge the domain gap between audio and visual modalities. The ACI is then used to well learn the inherent audio-visual relationships by cross-modal interaction. Extensive experiments demonstrate that the proposed framework could outperform other state-of-the-art multimodal deepfake detection methods under various circumstances.

Recent achievements have shown that model-based steganographic schemes hold promise for better security than heuristic-based ones, as they can provide theoretical guarantees on secure steganography under a given statistical model. However, it remains a challenge to exploit the correlations between DCT coefficients for secure steganography in practical scenarios where only a single compressed JPEG image is available. To cope with this, we propose a novel model-based steganographic scheme using the Conditional Random Field (CRF) model with four-element cross-neighborhood to capture the dependencies among DCT coefficients for JPEG steganography with symmetric embedding. Specifically, the proposed CRF model is characterized by the delicately designed energy function, which is defined as the weighted sum of a series of unary and pairwise potentials, where the potentials associated with the statistical detectability of steganography are formulated as the KL divergence between the statistical distributions of cover and stego. By optimizing the constructed energy function with the given payload constraint, the non-independent distortion cost corresponding to the least detectability can be accordingly obtained. Extensive experimental results validate the effectiveness of our proposed scheme, especially outperforming the previous independent art J-MiPOD.

Abstract In the area of reversible data hiding (RDH), one of the most popular techniques is prediction-error expansion (PEE), which hides data in the prediction errors with well-preserved image fidelity. The key to a successful PEE-based RDH implementation usually lies in prediction algorithms with high accuracy. Existing PEE-based RDH works often employ one single prediction algorithm, which is usually globally optimized, but with less consideration of the pixel distribution characteristics within local neighborhoods. In this manuscript, the technique of pattern adaptive prediction is proposed for pixel estimation according to the type of local binary pattern (LBP), which is obtained from the pixel’s eight neighborhood. Theoretically speaking, pattern-based predictors can be designed for each and every LBP patterns to create multiple prediction-error histograms (PEHs). However, the process of performance optimization with multiple PEHs requires extremely heavy computing power. To speed up the optimization process, LBP patterns are classified into various groups based on the degree of histogram concentration. Experiments demonstrate that the prediction accuracy is obviously improved and the image fidelity is well preserved.

This letter concentrates on the challenges of deep learning-based robust image watermarking against print-scanning, print-camera, and screen-shooting attacks for “physical channel transmission”. Given the excellent performance demonstrated by wavelet domain watermarking, in this paper, we incorporate the wavelet integrated convolutional neural networks (CNNs) and propose a Deep Wavelet-domain Watermarking (DWW) model, which is dedicated to embedding watermarks in the wavelet domain rather than the spatial domain of the previous arts. In addition, a frequency-domain enhanced mask loss is developed to increase the loss weight in the high-frequency regions of the image during back-propagation, thereby encouraging the model to embed the message in low-frequency components with priority so as to improve the robustness performance. Experiment results show that the proposed DWW consistently outperforms other state-of-the-art (SOTA) schemes by a clear margin in terms of embedding capacity, imperceptibility, and robustness.

Deep learning-based watermarking frameworks have received extensive research attention in recent years. The main structure of this framework consists of an encoder, a noise layer and a decoder (Encoder-NoiseLayer-Decoder). However, such a framework has the major drawback that it requires visible markers to locate a watermarked image, which compromises the imperceptibility of watermarking. To address this restriction, a novel Lite localization network based on Lite-HRNet is proposed. In order to generate high-quality watermarked image, we designed the Double U-Net Encoder (DUE), which can better hide the watermarking information in image pixels that are invisible to the human eye. Meanwhile, to improve robustness, two bicubic interpolation operations are added to the noise layer to increase the type of distortion. In addition, to further enhance the performance of the watermarking algorithm, the novel WGAN-GP loss function based on discriminator is designed to guide the training of the model. Numerous experiments demonstrate the superior performance of our proposed scheme in terms of localization function, visual quality, and robustness. The proposed scheme shows better results compared to state-of-the-art algorithms.

Recent advances in content-adaptive Audio Steganography in Temporal Domain (ASTD) suggest that modification of micro-amplitude samples may compromise its security. To prevent the micro-amplitude samples from being modified, a targeted Large Amplitude First (LAF) rule was adopted in some audio steganographic schemes, e.g., DFR. However, it is observed that the results with LAF rule are often unstable across different datasets, we thus propose a new Micro-Amplitude Suppression (MAS) rule in this paper following the design philosophy of wet paper coding. Unlike DFR where the audio steganographic performance heavily depends on the adopted heuristic filters, we propose to evaluate the embedding cost of cover audio with the Generalized Audio Intrinsic Energy (GAIE), which is obtained by calculating the weighted sum of squared DCT coefficients for each segmented audio clip with carefully designed weights. Extensive experimental results demonstrate that the proposed MAS rule tends to be more general and consistent than the LAF rule, and the proposed GAIE also shows better empirical security performance and audio quality compared to the advanced AAC and DFR_res (a variant of DFR). In addition, by preventing the micro-amplitude samples from being modified, the proposed GAIE_MAS can not only outperform other hand-crafted audio steganographic schemes but also the recently emerged deep learning-based schemes, e.g., IAA.

Video models on federated learning (FL) enable continual learning of the involved models for video tasks on end-user devices while protecting the privacy of end-user data. As a result, the security issues on FL, e.g., the backdoor attacks on FL and their defense have increasingly become the domains of extensive research in recent years. The backdoor attacks on FL are a class of poisoning attacks, in which an attacker, as one of the training participants, submits poisoned parameters and thus injects the backdoor into the global model after aggregation. Existing backdoor attacks against videos based on FL only poison RGB frames, which makes it that the attack could be easily mitigated by two-stream model neutralization. Therefore, it is a big challenge to manipulate the most advanced two-stream video model with a high success rate by poisoning only a small proportion of training data in the framework of FL. In this paper, a new backdoor attack scheme incorporating the rich spatial and temporal structures of video data is proposed, which injects the backdoor triggers into both the optical flow and RGB frames of video data through multiple rounds of model aggregations. In addition, the adversarial attack is utilized on the RGB frames to further boost the robustness of the attacks. Extensive experiments on real-world datasets verify that our methods outperform the state-of-the-art backdoor attacks and show better performance in terms of stealthiness and persistence.

Though existing methods for face forgery detection achieve fairly good performance under the intra-dataset scenario, few of them gain satisfying results in the case of cross-dataset testing with more practical value. To tackle this issue, in this paper, we propose a novel domain-invariant feature learning framework - DIFL for face forgery detection. In the framework, an adversarial domain generalization is introduced to learn the domain-invariant features from the forged samples synthesized by various algorithms. Then a center loss in fractional form (CL) is utilized to learn more discriminative features by aggregating the real faces while separating the fake faces from the real ones in the embedding space. In addition, a global and local random crop augmentation strategy is utilized to generate more data views of forged facial images at various scales. Extensive experimental results demonstrate the effectiveness and generalization of the proposed method compared with other state-of-the-art methods.

Conventional image steganography is assumed to transmit the message, in the most securest way possible for a given payload, over lossless channels, and the associated steganographic schemes are generally vulnerable to active attacks, e.g., JPEG re-compression, and scaling, as seen on social networks. Although considerable progress has been made on robust steganography against JPEG re-compression, there exist few steganographic schemes capable of resisting scaling attacks due to the tricky inverse interpolations involved in algorithm design. To tackle this issue, a framework for robust image steganography resisting scaling with general interpolations either in std form with fixed interpolation block, or pre-filtering-based anti-aliasing implementation with variable block, is proposed in this paper. And the task of robust steganography can be formulated as one of constrained integer programming aiming at perfectly recovering the secret message from the stego image while minimizing the difference between cover and stego images and the embedding distortion between scaled cover and scaled stego images. By introducing a metric - the degree of pixel involvement (dPI) to identify the modifiable pixels in the cover image, the optimization problem above could be effectively solved using the branch and bound algorithm (B&B). Extensive experiments demonstrate that the proposed scheme could not only resist scaling attacks with various interpolation techniques at arbitrary scaling factors (SFs), but also outperform the prior art in terms of security between the cover and stego images by a clear margin. In addition, the application of the proposed method in LinkedIn against the joint attacks of scaling and JPEG re-compression also shows its effectiveness on social network in real-world scenarios.

The recent success of deep image watermarking has demonstrated the potential of deep learning for watermarking, which has drawn increasing attention to deep video watermarking with the objective to improve its robustness and perceptual quality. Compared to images, video watermarking is much more challenging due to the rich structures of video data and the diversity of attacks in video transmission pipeline. The existing deep video watermarking schemes are far from satisfactory in dealing with temporal attacks, e.g., frame averaging, frame dropping and transcoding. To this end, a novel deep framework for Robustness Enhanced Video watermarking (REVMark) is proposed in this paper, aiming at improving the overall robustness, especially in dealing with H.264/AVC compression, while maintaining good visual quality. REVMark has an encoder/decoder structure with a pre-processing block (TAsBlock) to effectively extract the temporal-associated features on aligned frames. To ensure the end-to-end robust training, a distortion layer is integrated into the REVMark to resemble various attacks in real-world scenarios, among which, a new differentiable simulator of video compression, namely DiffH264, is developed to approximately simulate the process of H.264/AVC compression. In addition, the mask loss is incorporated to guide the encoder to embed the watermark in the human-imperceptible regions, thus improving the perceptual quality of the watermarked video. Experimental results demonstrate that the proposed scheme can outperform other SOTA methods while achieving 10X faster inference.

Abstract In prediction error-based reversible data hiding, multiple histograms modification (MHM) is well known for high image quality and thus has received wide attention in recent years. However, the computational cost for performance optimization in MHM is too high, which is particularly critical for real-time applications. This manuscript aims to reduce the computational complexity of MHM by presenting two techniques, including fast performance optimization and adaptive pixel distribution. Fast performance optimization provides a two-stage process for optimal bin selection by exploiting the concept of per-bit distortion of data embedding within a prediction error histogram (PEH). In fast performance optimization, the distribution characteristics of the per-bit distortion are investigated to significantly narrow down the solution space of optimal bin selection. The second technique is adaptive pixel distribution, which tries to nonuniformly allocate pixels into multiple PEHs to further reduce the time complexity. Extensive experiments show that the computational complexity of MHM is significantly reduced while well preserving the image quality.

As encryption masks the content of the original image and thus statistical characteristics of the original image cannot be used to compress the encrypted version, compressing an encrypted image efficiently remains a significant challenge today. In this study, a novel encryption-then-lossy-compression (ETLC) scheme was developed using nonuniform downsampling and a customized deep network. Specifically, the nonuniform downsampling method integrates both uniform and random sampling to achieve an arbitrary compression ratio for an encrypted image. Lossy reconstruction from the decrypted and decompressed image is described as a constrained optimization problem, and an ETLC-oriented customized deep neural network (ETCNN) is elaborately designed to solve this problem. ETCNN contains three parts: channel-wise non-local attention including residual group and non-local sparse attention, a residual content supplementation (RCS), and a downsampling constraint (DC), where RCS and DC are customized modules exploiting specific features of the downsampling-based ETLC system. Extensive experimental simulations show that the proposed scheme outperforms the state-of-the-art ETLC methods remarkably, indicating the feasibility and effectiveness of the proposed scheme exploiting the nonuniform downsampling and ETCNN-based reconstruction. Code is available athttps://github.com/hujuanzp/ETCNN.

Nowadays it has still remained as a big challenge to efficiently compress color images in the encrypted domain. In this paper we present a novel deep-learning-based approach to encryption-then-lossy-compression (ETC) of color images by incorporating the domain knowledge of the encrypted image reconstruction process. In specific, a simple yet effective uniform down-sampling is utilized for lossy compression of images encrypted with a modulo-256 addition, and the task of image reconstruction from an encrypted down-sampled image is then formulated as a problem of constrained super-resolution (SR) reconstruction. A customized residual dense spatial network (RDSN) is proposed to solve the formulated constrained SR task by taking advantage of spatial attention mechanism (SAM), global skip connection (GSC), and uniform down-sampling constraint (UDC) that is specific to an ETC system. Extensive experimental results show that the proposed ETC scheme achieves significant performance improvement compared with other state-of-the-art ETC methods, indicating the feasibility and effectiveness of the proposed deep-learning based ETC scheme.

Nowadays, convolutional neural network (CNN)-based JPEG steganalyzers have demonstrated much better performance than the conventional steganalysis methods based on hand-crafted feature sets. By incorporating the selection-channel aware (SCA) knowledge, the performance of the deep learning-based approach could be further improved. For prac-tical applications, however, the SCA knowledge is usually u-navailable to steganalyzers. In this paper, a novel CNN model is proposed by including extra non-linear kernels in the first network layer to enhance stego signal and exploring the resid-ual channel-spatial attention (CSA) module which plays the same role as SCA to further improve the performance. In addition, instead of global average pooling, spatial pyramid pooling (SPP) is adopted to better preserve the hierarchical feature representation at various scales for JPEG steganaly-sis. Experimental results show that the proposed CNN model with extra non-linear kernels, CSA and SPP outperforms other state-of-the-art deep learning-based approaches for JPEG steganalysis.

“Physical” watermarking, i.e., print-scanning, print-camera, and screen-shooting resilient watermarking, has drawn great attention these years. Recent studies show that watermarking with deep networks, e.g., StegaStamp, is particularly suitable for these tasks with strong robustness against the attacks from “physical transmissions” in the wild, although it usually has a relatively small amount of embedding capacity. Recognizing the conventional CNNs are vulnerable to input noise inter-ruptions, the wavelet-based CNNs are adopted in our work by replacing their down-sampling (pooling) and up-sampling layers with Discrete Wavelet Transform (DWT) and Inverse Discrete Wavelet Transform (IDWT), respectively, to learn the stable feature representation from noise-corrupted sam-ples. A new residual regularization loss function incorporating the texture complexity is also proposed to significantly improve the visual quality of watermarked images. Exper-imental results show that the proposed wavelet-based CNN model significantly outperforms the state-of-the-art StegaS-tamp in terms of embedding capacity, imperceptibility, and robustness.

In this letter, a novel steganographic scheme in motion vector domain (MV) for H.264 video is presented, which can significantly improve the security performance against the newly emerged powerful multi-domain feature set MVC (motion vector consistency). By taking into account both the consistency degree of motion vectors for sub-blocks within a macroblock (MB) or sub-macroblock (sub-MB), and the MV statistics, the corresponding distortion function called dMVC is proposed. The proposed dMVC is also shown to be capable of integrating with existing methods to resist the joint steganalytic attacks of both MVC feature and local optimality features, e.g., NPELO, in the framework of minimal distortion embedding. Compared with other state-of-the-art MV-based steganographic schemes, experimental results on YUV sequences at various embedding rates and QPs show that the proposed method gains significant performance improvement while maintaining good coding efficiency.

In this letter, we propose a general digital image operation anti-forensic framework based on generative adversarial nets (GANs), called dual-domain generative adversarial network (DDGAN). To tackle the issue of image operation detection, the proposed framework incorporates both operation specific forensic features and machine-learned knowledge to ensure that the generated images exhibit better undetectability performance against various detectors. The DDGAN consists of a generator and two discriminators working on different domains, i.e., the operation-specific feature domain which helps to conceal the artifacts from the perspective of forensic analysis for the target task, and the spatial domain which facilitates to take advantage of machine-learned features from the scratch as a supplementary. Through the experiments on median filtering and JPEG compression anti-forensics, we show the superior performance of the proposed DDGAN compared with state-of-the-art anti-forensic methods in terms of undetectability and visual quality.

With wide applications of image editing tools, forged images (splicing, copy-move, removal and etc.) have been becoming great public concerns. Although existing image forgery localization methods could achieve fairly good results on several public datasets, most of them perform poorly when the forged images are JPEG compressed as they are usually done in social networks. To tackle this issue, in this paper, a self-supervised domain adaptation network, which is composed of a backbone network with Siamese architecture and a compression approximation network (ComNet), is proposed for JPEG-resistant image forgery localization. To improve the performance against JPEG compression, ComNet is customized to approximate the JPEG compression operation through self-supervised learning, generating JPEG-agent images with general JPEG compression characteristics. The backbone network is then trained with domain adaptation strategy to localize the tampering boundary and region, and alleviate the domain shift between uncompressed and JPEG-agent images. Extensive experimental results on several public datasets show that the proposed method outperforms or rivals to other state-of-the-art methods in image forgery localization, especially for JPEG compression with unknown QFs.

With wide applications of facial manipulation technology, fake images and videos are becoming a great public concern. Although existing methods for face forgery detection could achieve fairly good results on public database, most of them perform poorly when the fake images/videos are compressed as they are usually done in social networks. To tackle this issue, a self-supervised decoupling network (SSDN), that incorporates compression irrelevance, is proposed in this paper. The proposed model learns two separate feature representations for the suspect videos, i.e., authenticity and compression. A joint self-supervised strategy is then utilized for feature decoupling, in which, the similarity decoupling is carried out by similarity learning on authentic features, whereas for adversarial decoupling, the proposed SSDN model is trained in an adversarial manner for robust feature learning. Experimental results show that the SSDN outperforms the state-of-the-art methods for deepfake detection against compression attacks on public datasets, e.g., FaceForensics++.

Recent advances on adaptive steganography show that the performance of image steganographic communication can be improved by incorporating the non-additive models that capture the dependencies among adjacent pixels. In this paper, a Gaussian Markov Random Field model (GMRF) with four-element cross neighborhood is proposed to characterize the interactions among local elements of cover images, and the problem of secure image steganography is formulated as the one of minimization of KL-divergence in terms of a series of low-dimensional clique structures associated with GMRF by taking advantages of the conditional independence of GMRF. The adoption of the proposed GMRF tessellates the cover image into two disjoint subimages, and an alternating iterative optimization scheme is developed to effectively embed the given payload while minimizing the total KL-divergence between cover and stego, i.e., the statistical detectability. Experimental results demonstrate that the proposed GMRF outperforms the prior arts of model based schemes, e.g., MiPOD, and rivals the state-of-the-art HiLL for practical steganography, where the selection channel knowledges are unavailable to steganalyzers.

Batch steganography aims at properly allocating a large payload to multiple covers, so as to keep the whole covert communication at a satisfactory level of security. JPEG is currently one of the most widely used formats for image storage and transmission. This paper presents an efficient JPEG batch steganographic scheme, which allocates the payload in a linear manner w.r.t. a new heuristic measure - the intrinsic energy of JPEG image contents, in which more concerns are with the high frequency components, and the proposed measure could also be easily generalized to cover selection in batch steganographic applications. And a calibration strategy is elaborately designed to balance the security level when JPEG covers of various QFs are involved in JPEG batch steganography. In this way, the proposed scheme can effectively resolve the problem that the statistical undetectability fluctuates dramatically w.r.t. the size and quality factor when the batch set is involved with various image parameters, and consequently maintains the overall security of the practical JPEG batch steganographic system. Experimental results show that the proposed method exhibits security performance superior or comparable to the state-of-the-art batch schemes while maintaining a low computational cost.

In this paper, we apply cross-layer intersection mechanism to dense u-net for image forgery detection and localization. We first train DenseNet for binary classification. Spatial rich model (SRM) filters are adopted for capturing residual signals in the detected images. Then we propose a new approach to preserve complete feature maps of fully connected layer and consider them as the spatial decision information for image segmentation. In addition, these features in downsampling path are transferred more effectively and densely to upsampling path through multiscale upsampling and concatenation. A multi-stage training scheme is then applied to improve the convergence of the network. The experimental results show that the proposed method works well on several standard datasets.

In recent years, many forensic methods are proposed for camera model identification (CMI). These methods expose the acquisition devices of the images according to the traces left during the imaging process. However, such traces could be easily affected by common image operations, e.g., JPEG compression, which make the camera model identification of post-processed images very difficult. In this paper, we propose a GAN based decoupling network (Decoupling-GAN) to boost the performance of CNN-based model detectors for JPEG compressed images by alleviating the effects of JPEG compression on the camera model identification. Through the adversarial training, we rebuilt the consistency of extracted feature maps between the original images and JPEG compressed ones. Experimental results show that Decoupling-GAN exhibits good robustness performance and outperforms prior arts in terms of detection accuracy under JPEG attacks.

Abstract Reversible data hiding (RDH) with contrast enhancement (RDH-CE) is a special type of RDH in improving the subjective visual perception by enhancing the image contrast during the process of data embedding. In RDH-CE, data hiding is achieved via pairwise histogram expansion, and the embedding rate can be increased by performing multiple cycles of histogram expansions. However, when embedding rate gets high, human visible image degradation is observed. Previous work designed an upper bound of the embedding level for RDH-CE, which effectively avoids image over-sharping but offers limited embedding capacity. In this paper, a better tunable bound is designed to enhance the embedding capacity of RDH-CE by exploiting the characteristics of histogram distribution. Furthermore, the objective distortion introduced by histogram pre-shifting is minimized when the embedding level is no more than the upper bound, and the human visible degradation is minimized when the embedding level exceeds the limitation of the proposed upper bound. Experimental results validate that the proposed method provides appropriate upper bound of the embedding level, increases the effective embedding capacity and offers better image contrast.

Reversible data hiding (RDH) has unique advantage in copyright and integrity protection for multimedia contents. As a typical RDH scheme, histogram shifting technique (HS) has found wide applications due to its high quality of marked image. At present, most existing HS-based RDH schemes rely on single histogram generated from cover image to hide data. Since the single histogram-based approach (SH_RDH) commonly employs smooth regions in the cover image for data hiding, it might not well utilize the cover image and exploit the correlations among image contents of different texture characteristics. In this paper, a novel RDH general framework using multiple histograms modification (MH_RDH) is proposed, which involves two key issues as follows: 1) the construction of multiple histograms based on optimized multi-features and 2) the rate allocation among multiple histograms is formulated as the one of rate-distortion optimization and solved with evolutionary algorithms. The experimental results show that the proposed method could considerably increase the payload of current MH_RDH-based embedding (ranging from 0.2 to 0.7 bpp for most test images) and outperform the other state-of-the-art SH_RDH and MH_RDH schemes.

Nowadays, convolutional neural network (CNN) is appied to different types of image classification tasks and outperforms almost all traditional methods. However, one may find it difficult to apply CNN to JPEG steganalysis because of the extremely low SNR (embedding messages to image contents) in the task. In this paper, a selection-channel-aware CNN for JPEG steganalysis is proposed by incorporating domain knowledge. Specifically, instead of random strategy, kernels of the first convolutional layer are initialized with hand-crafted filters to suppress the image content. Then, truncated linear unit (TLU), a heuristically-designed activation function, is adopted in the first layer as the activation function to better adapt to the distribution of feature maps. Finally, we use a generalized residual learning block to incorporate the knowledge of selection channel in the proposed CNN to further boost its performance. J-UNIWARD, a state-of-the-art JPEG steganographic scheme, is used to evaluate the performance of the proposed CNN and other competing JPEG steganalysis methods. Experiment results show that the proposed CNN steganalyzer outperforms other feature-based methods and rivals the state-of-the-art CNN-based methods with much reduced model complexity, at different payloads.

In this paper, the uniform embedding for JPEG steganography is generalized to the motion vector (MV) domain of H.264 video (UED_H.264). By taking into account the several key issues in video steganography, e.g., the MV correlations, the local optimality and the degradation of the reconstructed video frames, the comprehensive distortion function is proposed to incorporate the minimal distortion embedding framework. The proposed scheme uses MVs as cover elements, which includes both forward and backward MVs, and the embedding is deliberately carried out with ternary syndrome-trellis codes (STC) according to the proposed scheme. Experimental results show that the proposed scheme can improve the security performance on resisting the steganalytic attacks while preserving the coding efficiency within acceptable level.

Nowadays, JPEG steganographic schemes, e.g., J-UNIWARD, which take into account the effects of embedding in the spatial domain tend to exhibit higher security and introduce less artifacts that can be captured by the prevalent steganalyzers. Following the paradigm, this letter proposes a new design of the distortion measure for JPEG steganography by incorporating the statistics of both the spatial and discrete cosine transform (DCT) domains. The spatial statistics of the decompressed JPEG images are first well characterized with distortion measures of some efficient steganographic schemes in the spatial domain, e.g., HILL, and the resulting embedding entropies of spatial blocks in alignment with DCT blocks are then transformed into the DCT domain to obtain the distortion measures for JPEG steganography. Experimental results show that the proposed method outperforms considerably other state-of-the-art JPEG steganographic schemes, i.e., J-UNIWARD and UERD, for the most effective feature set GFR at present, and rivals them for other feature sets, e.g., DCTR and CC-JRM.

Nowadays, the most prevailing approach to steganography is the minimal embedding distortion framework, which includes an optimizable distortion function for each cover element and an encoding method to minimize the distortion. With the emergence of Syndrome-Trellis Code, the distortion function plays an increasingly important role in modern adaptive image steganography. In this letter, a new distortion function called generalized uniform embedding distortion (GUED) is proposed for JPEG steganography. The proposed GUED consists of the new distortion measures for both Alternating Current (AC) mode and Discrete Cosine Transform (DCT) block, which are represented in a more general exponential model, aiming to flexibly allocate the embedding data so as to minimize the global changes of the statistics of quantized DCT coefficients after embedding. In addition, an empirical rule is developed to determine the parameters of the exponential function according to the payload and quality factor. By exploring the statistics of both DCT and spatial domains, the proposed GUED is shown to be more consistent with the objective of generalized uniform embedding strategy, i.e., maintaining the relative changes of DCT coefficients to be proportional to their coefficients of variations. Extensive experiments demonstrate that the proposed GUED gains significant performance improvements when compared with its original UERD, and outperforms the state-of-the-art J-UNIWARD with markedly reduced computation time.

Similar to conventional compression with the original, unencrypted image as the input, the recently emerged compression on encrypted images generally exploits statistical correlation of natural images to improve compression efficiency. Most of these compression schemes in the literature leverage statistical correlation at the content-owner or service-provider side, which would either increase the computational burden on the content-owner or disclose statistical distributions to the service-provider and thus probably hinder their practical applications. Through analysis on properties of the compression system for encrypted data, we believe that it is more preferable to exploit statistical correlation of natural images at the receiver side with both encryption key and sufficient computational capability, which in turn would improve compression efficiency while achieving low computational complexity and sufficient security for the content owner and the service provider. In light of this, we use the Markov random field (MRF) to characterize binary images in the spatial domain and represent it with a factor graph. The constructed MRF representation of the binary image in the factor graph is then integrated seamlessly with the factor graph for low-density parity check (LDPC)-based decompression, yielding a joint factor graph for binary image reconstruction. By deriving message update equations for the joint factor graph, we develop a new lossless compression scheme for encrypted binary images, which involves stream-cipher-based encryption, LDPC-based compression, and factor-graph-based image reconstruction. Preferable parameters for the proposed scheme are first determined numerically on a specific binary image and then applied to other binary images. Extensive simulations show that significant improvements in terms of compression bit rate over the state of the art are achieved, demonstrating the feasibility and effectiveness of the proposed scheme.

Recently, a number of high-fidelity reversible data hiding algorithms have been developed based on prediction-error expansion (PEE) and pixel sorting. In PEE, prediction is made using either a full-enclosed or a half-enclosed predictor. While in PEE with pixel sorting, the local complexity (LC), which is usually assumed to be proportional to the magnitude of prediction-error (PE), is exploited to reduce the embedding distortion. However, this assumption may not always hold in all conditions. In this letter, a directional enclosed predictor is proposed to detect the locations where LC is not proportional to PE. And, a directionally enclosed prediction and expansion (DEPE) scheme is then developed for efficient reversible data hiding. With DEPE, data embedding is restricted to pixels where LC correlates to PE with a proportional relationship. Experimental results show that, compared to the full-enclosed or half-enclosed prediction schemes, DEPE significantly improves the image fidelity while providing a considerable payload.

Histogram shifting (HS) embedding as a typical reversible data hiding scheme is widely investigated due to its high quality of stego-image. For HS-based embedding, the selected side information, i.e., peak and zero bins, usually greatly affects the rate and distortion performance of the stego-image. Due to the massive solution space and burden in distortion computation, conventional HS-based schemes utilize some empirical criterion to determine those side information, which generally could not lead to a globally optimal solution for reversible embedding. In this paper, based on the developed rate and distortion model, the problem of HS-based multiple embedding is formulated as the one of rate and distortion optimization. Two key propositions are then derived to facilitate the fast computation of distortion due to multiple shifting and narrow down the solution space, respectively. Finally, an evolutionary optimization algorithm, i.e., genetic algorithm is employed to search the nearly optimal zero and peak bins. For a given data payload, the proposed scheme could not only adaptively determine the proper number of peak and zero bin pairs but also their corresponding values for HS-based multiple reversible embedding. Compared with previous approaches, experimental results demonstrate the superiority of the proposed scheme in the terms of embedding capacity and stego-image quality.

Nowadays, the prevailing detectors of steganographic communication in digital images mainly consist of three steps, i.e., residual computation, feature extraction, and binary classification. In this paper, we present an alternative approach to steganalysis of digital images based on convolutional neural network (CNN), which is shown to be able to well replicate and optimize these key steps in a unified framework and learn hierarchical representations directly from raw images. The proposed CNN has a quite different structure from the ones used in conventional computer vision tasks. Rather than a random strategy, the weights in the first layer of the proposed CNN are initialized with the basic high-pass filter set used in the calculation of residual maps in a spatial rich model (SRM), which acts as a regularizer to suppress the image content effectively. To better capture the structure of embedding signals, which usually have extremely low SNR (stego signal to image content), a new activation function called a truncated linear unit is adopted in our CNN model. Finally, we further boost the performance of the proposed CNN-based steganalyzer by incorporating the knowledge of selection channel. Three state-of-the-art steganographic algorithms in spatial domain, e.g., WOW, S-UNIWARD, and HILL, are used to evaluate the effectiveness of our model. Compared to SRM and its selection-channel-aware variant maxSRMd2, our model achieves superior performance across all tested algorithms for a wide variety of payloads.

Geometric transformations, such as resizing and rotation, are almost always needed when two or more images are spliced together to create convincing image forgeries. In recent years, researchers have developed many digital forensic techniques to identify these operations. Most previous works in this area focus on the analysis of images that have undergone single geometric transformations, e.g., resizing or rotation. In several recent works, researchers have addressed yet another practical and realistic situation: successive geometric transformations, e.g., repeated resizing, resizing-rotation, rotation-resizing, and repeated rotation. We will also concentrate on this topic in this paper. Specifically, we present an in-depth analysis in the frequency domain of the second-order statistics of the geometrically transformed images. We give an exact formulation of how the parameters of the first and second geometric transformations influence the appearance of periodic artifacts. The expected positions of characteristic resampling peaks are analytically derived. The theory developed here helps to address the gap left by previous works on this topic and is useful for image security and authentication, in particular, the forensics of geometric transformations in digital images. As an application of the developed theory, we present an effective method that allows one to distinguish between the aforementioned four different processing chains. The proposed method can further estimate all the geometric transformation parameters. This may provide useful clues for image forgery detection.

Histogram-shifting (HS) embedding as a most successful reversible data hiding (RDH) scheme is widely investigated. How to take advantage of covers' redundancyis one of the key issues for performance improvement of RDH. Among them, sorting technique isof practically importance, which uses in prioritysmooth areas with high correlation to hide message. For conventional schemes, usually a single empirical feature in the context of local cover object, such as local variance, is utilized as complexity measure for sorting, which may not effectively exploit the covers' correlations and thus lead to limited performance improvement. In this paper, a general framework for optimal construction of complexity measure with multi-feature is developed, which includes optimal feature selection and weight parameters determination based on an optimization model. In addition, the optimal truncation point for the top of the sorted cover is determined based on the payload constraint to construct the optimal histogram for further performance improvement of HS embedding. Compared with previous approaches, experimental results demonstrate the superiority of the proposed scheme.

Uniform embedding was first introduced in 2012 for non-side-informed JPEG steganography, and then extended to the side-informed JPEG steganography in 2014. The idea behind uniform embedding is that, by uniformly spreading the embedding modifications to the quantized discrete cosine transform (DCT) coefficients of all possible magnitudes, the average changes of the first-order and the second-order statistics can be possibly minimized, which leads to less statistical detectability. The purpose of this paper is to refine the uniform embedding by considering the relative changes of statistical model for digital images, aiming to make the embedding modifications to be proportional to the coefficient of variation. Such a new strategy can be regarded as generalized uniform embedding in substantial sense. Compared with the original uniform embedding distortion (UED), the proposed method uses all the DCT coefficients (including the DC, zero, and non-zero AC coefficients) as the cover elements. We call the corresponding distortion function uniform embedding revisited distortion (UERD), which incorporates the complexities of both the DCT block and the DCT mode of each DCT coefficient (i.e., selection channel), and can be directly derived from the DCT domain. The effectiveness of the proposed scheme is verified with the evidence obtained from the exhaustive experiments using a popular steganalyzer with rich models on the BOSSbase database. The proposed UERD gains a significant performance improvement in terms of secure embedding capacity when compared with the original UED, and rivals the current state-of-the-art with much reduced computational complexity.

In this letter, we propose a blind and effective method to estimate the image rotation angle. It operates by exploiting the hidden periodicities in the rotated image using the 2-D spectrum of its second order statistics. The expected positions of rotation-related peaks as well as their trajectory paths in the 2-D spectrum are first derived. Based on the peaks found around those paths, the rotation angle is then estimated using a much simpler formula than those of the state-of-the-art 2-D method. Through both theoretical analyses and experimental results on a large set of natural images, the proposed scheme is demonstrated to give more accurate estimate than previous methods.

Steganography is the science and art of covert communication, which aims to hide the secret messages into a cover medium while achieving the least possible statistical detectability. To this end, the framework of minimal distortion embedding is widely adopted in the development of the steganographic system, in which a well designed distortion function is of vital importance. In this paper, a class of new distortion functions known as uniform embedding distortion function (UED) is presented for both side-informed and non side-informed secure JPEG steganography. By incorporating the syndrome trellis coding, the best codeword with minimal distortion for a given message is determined with UED, which, instead of random modification, tries to spread the embedding modification uniformly to quantized discrete cosine transform (DCT) coefficients of all possible magnitudes. In this way, less statistical detectability is achieved, owing to the reduction of the average changes of the first- and second-order statistics for DCT coefficients as a whole. The effectiveness of the proposed scheme is verified with evidence obtained from exhaustive experiments using popular steganalyzers with various feature sets on the BOSSbase database. Compared with prior arts, the proposed scheme gains favorable performance in terms of secure embedding capacity against steganalysis.

Recently, the median filtering (MF) detector as a forensic tool for the recovery of images' processing history has attracted wide interest. This paper presents a novel method for the blind detection of MF in digital images. Following some strongly indicative analyses in the difference domain of images, we introduce two new feature sets that allow us to distinguish a median-filtered image from an untouched image or average-filtered one. The effectiveness of the proposed features is verified with evidence from exhaustive experiments on a large composite image database. Compared with prior arts, the proposed method achieves significant performance improvement in the case of low resolution and strong JPEG post-compression. In addition, it is demonstrated that our method is more robust against additive noise than other existing MF detectors. With analyses and extensive experimental researches presented in this paper, we hope that the proposed method will add a new tool to the arsenal of forensic analysts.

Steganography is the art of covert communication. This paper presents an efficient JPEG steganography scheme based on the block entropy of DCT coefficients and syndrome trellis coding (STC). The proposed cost function explores both the block complexity and distortion effects due to flipping and rounding errors. The STC provides multiple solutions to embed messages to a block of coefficients. The proposed scheme determines the best one with minimal distortion effect. In this way, the total distortions are significantly reduced, which corresponds to less detectability of steganalysis. Compared with similar schemes, experiment results demonstrate the superior performance of the proposed scheme in terms of secure embedding capacity against steganalysis.

Achieving robustness, imperceptibility and high capacity simultaneously is of great importance in digital watermarking. This paper presents a new informed image watermarking scheme with high robustness and simplified complexity at an information rate of 1/64 bit/pixel. Firstly, a Taylor series approximated locally optimum test (TLOT) detector based on the hidden Markov model (HMM) in the wavelet domain is developed to tackle the problem of unavailability of exact embedding strength in the receiver due to informed embedding. Then based on the TLOT detector and the concept of dirty-paper code design, new HMM-based spherical codes are constructed to provide an effective tradeoff between robustness and distortion. The process of informed embedding is formulated as an optimization problem under the robustness and distortion constraints and the genetic algorithm (GA) is then employed to solve this problem. Moreover, the perceptual distance in the wavelet domain is also developed and incorporated into the GA-based optimization. Simulation results demonstrate that the proposed informed watermarking algorithm has high robustness against common attacks in signal processing and shows a comparable performance to the state-of-the-art scheme with a greatly reduced arithmetic complexity.

Histogram shifting technique, as a general lossless data hiding scheme, leads to a good quality for the stego-image but the embedding capacity is limited. To enhance the capacity, histogram shifting would be always operated by multiple embedding, which uses multiple pairs of peak and zero points at a time, or multi-layer embedding, which utilizes one pair of peak and zero point each time and then repeating many times based on the resulting image. Although the performance of multi-layer embedding approach outperforms that of multiple embedding, it is, however, much more time-consuming. A fast performance estimation scheme for multi-layer embedding is proposed in the paper, which provides a fast and accurate capacity and quality estimation based only on the histogram information. The proposed scheme provides a fast assessment whether the multi-layer embedding approach is an appropriate reversible watermarking method for the given secret data and host image.

The robust performance against geometrical manipulations is still one of major concerns in robust watermarking although significant improvement has been achieved in past decades. In this paper, we tackle the global geometrical attacks by designing a deformable multi-scale transform (DMST) that has joint shiftability in position, orientation, and scale. Via DMST, we both derive theoretically the principles for geometrical synchronization and develop a template-based scheme to efficiently estimate geometrical parameters. Also, the hidden Markov model in the standard wavelet domain is extended to the steerable wavelet domain and further used to improve the performance of watermark extraction. Experimental simulation demonstrates that the proposed watermarking scheme is quite robust to the common signal processing, geometrical attacks, and their joint attacks.

A new informed image watermarking algorithm is presented in this paper, which can achieve the information rate of 1/64 bits/pixel with high robustness. Firstly, a LOT (local optimal test) detector based on HMM in wavelet domain is developed to tackle the issue that the exact strength for informed embedding is unknown to the receiver. Then based on the LOT detector, the dirty-paper code for informed coding is constructed and the metric for the robustness is defined accordingly. Unlike the previous approaches of informed watermarking which take the informed coding and embedding process separately, the proposed algorithm implements a joint coding and embedding optimization for high capacity and robust watermarking. The genetic algorithm (GA) is employed to optimize the robustness and distortion constraints simultaneously. Experimental results show that the proposed algorithm achieves significant improvements in performance against JPEG, gain attack, low-pass filtering and etc.

A DWT-HMM (hidden Markov model in wavelet domain) image watermarking algorithm with content-adaptive approach is proposed in this paper to optimized the trade-off between robustness and visual quality, which is characterized as follows: the entropy mask proposed by Watson is constructed in wavelet domain; the entropy mask and the new developed integrated HVS are used as the measures to adaptively select image components for watermarking; repeat-accumulation (RA) code with erasure and error correction is employed to synchronize the watermarked image; and a posterior HMM is utilized in watermark detection. Considerable improvement in robustness performance with the proposed adaptive algorithm is obtained over the previous DWT-HMM watermarking algorithm with stochastic embedding.