Learning adaptive embedding costs with deep learning has become an important direction for improving steganographic security. However, most existing approaches focus on spatial-domain images, while effective cost learning for JPEG images remains challenging due to the complexity introduced by block-wise discrete cosine transform and lossy quantization. In this paper, we present an enhanced GAN-based framework that learns asymmetric embedding costs for JPEG images from scratch, improving both content adaptivity and training stability. Specifically, we introduce an adaptive modification loss that directly links the generator’s predicted embedding probabilities to the actual modification behavior, enabling fully data-driven and content-aware optimization without relying on handcrafted filters or quantization-dependent heuristics. In addition, we propose a random masking strategy applied during later training stages to prevent discriminator dominance and maintain informative adversarial feedback. Extensive experiments on multiple benchmark datasets and under various steganalytic detectors demonstrate that the proposed method consistently improves steganographic security over existing JPEG-domain approaches. Ablation studies further validate the effectiveness of the proposed loss design and training strategy.

With the emergence of adversarial steganography, existing specialized steganalysis models suffer a significant performance decline in detecting non-homologous adversarial steganographic methods (i.e., trained on traditional-based method and tested on adversarial-based method), resulting in insufficient robustness in complex network environments. To address this issue, we propose a two-stage robust steganalysis framework with multi-feature enhancement against adversarial steganography. The framework integrates edge-aware attention with multi-dimensional statistical features to enhance robustness against adversarial steganography. In the first stage, we design a covariance pooling based convolutional neural network and integrate an edge-aware attention mechanism to improve the feature representation of subtle steganographic traces, enabling fast detection for most samples. In the second stage, samples with uncertain confidence scores from the first stage are further analyzed by extracting block-wise entropy features, global entropy features, and SRM co-occurrence features, followed by dimensionality reduction via principal component analysis (PCA) and classification using a random forest. The final decision is made through the collaborative fusion of the two stages. Experimental results demonstrate that the proposed method achieves excellent detection performance (2.57% average improvement over the existing best method) with strong robustness for adversarial steganography, and its generalization capability is further validated in cross-dataset scenarios. Furthermore, comprehensive ablation studies validate the efficacy of the network architecture.

The increasing use of Voice over Internet Protocol (VoIP) technology in telecom fraud has become a serious global concern. Its ability to spoof caller IDs and IP addresses, and the use of overseas or anonymized servers make VoIP-based scams difficult to trace and regulate. As a result, distinguishing VoIP calls from conventional mobile phone calls based on voice signal characteristics is crucial for enhancing anti-fraud measures. However, existing forensic techniques often struggle to accurately identify speech transmitted via VoIP. To address this challenge, we propose a dual-level 1D-CNN that leverages both frame and utterance features for effective VoIP detection. After evaluating a range of acoustic features, we primarily focus on short-frame Mel-Frequency Cepstral Coefficients (MFCCs) due to their effectiveness in capturing VoIP characteristics. Given the frame-based processing and transmission nature of VoIP, we employ a 1D-CNN, rather than the more commonly used 2D-CNN that treats spectrograms as image, to extract frame-level codec features. Finally, we propose a dual-level classification strategy: the frame-level classifier captures encoding discrepancies within individual frames, while the utterance-level classifier aggregates these frame-level features to learn global encoding patterns through global covariance pooling. Experimental results on the VoIP Phone Call Identification Database (VPCID) demonstrate that the proposed method consistently outperforms existing approaches, delivering superior accuracy and robustness across a wide range of challenging scenarios. Moreover, comprehensive ablation studies validate the effectiveness and rationale behind the design of the proposed model architecture.

A key challenge in current steganography research is automatically learning image embedding costs without relying on existing costs. To date, there has been limited work on JPEG steganography, and the reported methods primarily rely on learning symmetric embedding, which fails to fully exploit the relationships between different modification directions within an embedding unit. This limitation restricts their security, leaving room for improvements in JPEG steganography techniques. To address this issue, we propose a GAN-based framework for JPEG steganography that learns asymmetric embedding costs from scratch. Our approach extends a modern framework of spatial steganography by incorporating a key IDCT module, which facilitates the conversion between JPEG and spatial domains. This enables the integration of effective spatial steganographic and steganalytic techniques into our framework for JPEG steganography. Additionally, we use a dual-branch UNet to generate separate embedding probabilities for +1 and -1 DCT coefficients and introduce a specialized loss function to guide DCT modifications. This loss function is designed by converting the modified DCT coefficients back to the spatial domain and analyzing various spatial residuals. Extensive experiments demonstrate that our method significantly outperforms existing JPEG steganography techniques, achieving state-of-the-art security performance. Furthermore, many ablation experiments validate the rationale of our model.

Cloud computing has become a crucial IT infrastructure in the big data era, with an increasing number of users outsourcing video storage and computing tasks to cloud servers, raising concerns about privacy leakage and illegal video distribution. Although existing robust video watermarking technologies can trace illegal distribution, they do not address privacy protection. This paper proposes a robust watermarking algorithm for protection of privacy in compressed video streams. We apply selective video encryption in the compressed domain to protect privacy, enabling direct analysis and watermark embedding on encrypted videos without decompression. A dual-channel watermarking strategy is designed to model and analyze CAVLC entropy coding. To enhance robustness, we select syntax elements with anti-recompression capabilities from the compressed domain for embedding and use a majority voting mechanism in the embedding algorithm. Additionally, we introduce an entropy coding-based watermark retrieval algorithm to improve detection accuracy through a sliding window approach. Experimental results show that our method reduces the re-compression error rate by over 44% compared to existing methods.

Most existing steganographic techniques are primarily designed for grayscale images.When directly applied to color images without considering inter-channel color interactions, their security can be significantly compromised.In this paper, we propose a novel color image steganography method based on generative adversarial networks (GANs).Our framework features a dual-branch generator and a discriminator equipped with multiple steganalytic networks, each focused on a specific color channel.This architecture enables the progressive learning of asymmetric embedding costs across channels from scratch.We also introduce a phased training strategy that facilitates the learning of inter-channel color interactions and optimizes the security of each color channel in distinct phases, improving overall security.Furthermore, we propose a new adaptive update strategy and introduce the Mean Absolute Deviation (MAD) loss function to maintain a dynamic balance between the generator and discriminator, thereby progressively enhancing the generator's steganographic performance during training.Extensive comparative experiments demonstrate that the proposed method achieves state-of-the-art security against color image steganalysis.Comprehensive ablation studies further validate the effectiveness and rationale behind our approach.

Deep learning-based forensic techniques have emerged as the leading approach for image forgery localization. However, many existing methods struggle with overfitting to the training data, which limits their generalization performance and real-world applicability. To overcome this challenge, we propose a novel forensic framework that incorporates an advanced data augmentation technique. The framework consists of two key components: a generator and a detector. The generator challenges the detector’s learned distribution under constraints of diversity and consistency, ensuring that the generated data diverges from the source domain while maintaining statistical differences related to tampering. The detector, in turn, captures tampering traces from three critical aspects of the tampered image: long-range dependency information, RGB-noise fusion information, and boundary artifacts, resulting in a more comprehensive detection process. By alternating the optimization of the generator and detector, the framework fosters mutual reinforcement, promoting diverse data generation and expanding the distributional coverage, ultimately improving performance. Extensive experiments demonstrate that the proposed method significantly surpasses state-of-the-art approaches in both generalization and robustness, with numerous ablation studies further validating the soundness of the model design.

Facial attribute editing has garnered significant attention, yet prevailing methods struggle with achieving precise attribute manipulation while preserving irrelevant details and controlling attribute styles. This challenge primarily arises from the strong correlations between different attributes and the interplay between attributes and identity. In this paper, we propose Semantic Disentangled GAN (SDGAN), a novel method addressing this challenge. SDGAN introduces two key concepts: a semantic disentanglement generator that assigns facial representations to distinct attribute-specific editing modules, enabling the decoupling of the facial attribute editing process, and a semantic mask alignment strategy that confines attribute editing to appropriate regions, thereby avoiding undesired modifications. Leveraging these concepts, SDGAN demonstrates accurate attribute editing and achieves high-quality attribute style manipulation through both latent-guided and reference-guided manners. We extensively evaluate our method on the CelebA-HQ database, providing both qualitative and quantitative analyses. Our results establish that SDGAN significantly outperforms state-of-the-art techniques, showcasing the effectiveness of our approach. To foster reproducibility and further research, we will provide the code for our method.

Traditional steganographic techniques have often relied on manually crafted attributes related to image residuals. These methods demand a significant level of expertise and face challenges in integrating diverse image residual characteristics. In this paper, we introduce an innovative deep learning-based methodology that seamlessly integrates image residuals, residual distances, and image local variance to autonomously learn embedding probabilities. Our framework includes an embedding probability generator and three pivotal guiding components: Residual guidance strives to facilitate embedding in complex-textured areas. Residual distance guidance aims to minimize the residual differences between cover and stego images. Local variance guidance effectively safeguards against modifications in regions characterized by uncomplicated or uniform textures. The three components collectively guide the learning process, enhancing the security performance. Comprehensive experimental findings underscore the superiority of our approach when compared to traditional steganographic methods and randomly initialized ReLOAD in the spatial domain.

Deep learning solutions for privacy anonymization in video action recognition face significant computational challenges and difficulties in collecting privacy labels. To address these issues, we propose a fast privacy-preserving action recognition framework based on compressed video bitstreams. Our framework introduces two novel anonymization methods for video bitstreams, transforming video data directly within the compressed domain into anonymized compressed data to protect sensitive video information. The anonymized and compressed video data exhibits a compact data representation, enhancing transmission efficiency. We propose a technique to reconstruct anonymized compressed data in the spatial domain and utilize deep learning technology for high-accuracy action recognition. The experiment results demonstrate that our framework excels in balancing privacy protection performance with action recognition, showing a significant trade-off advantage. Our approach also achieves an operational efficiency more than fivefold greater than existing methods.

Designing embedding costs is pivotal in modern image steganography. Many studies have shown adjusting symmetric embedding costs to asymmetric ones can enhance steganographic security. However, most existing methods heavily depend on manually defined parameters or rules, limiting security performance improvements. To overcome this limitation, we introduce an advanced GAN-based framework that transitions symmetric costs to asymmetric ones without the need for the manual intervention seen in existing approaches, such as the detailed specification of cost modulation directions and magnitudes. In our framework, we firstly achieve symmetric costs for a cover image, which is randomly split into two sub-images, with part of the secret information embedded into one. Subsequently, we design a GAN model to adjust the embedding costs of the second sub-image to asymmetric, facilitating the secure embedding of the remaining secret information. To support our phased embedding approach, our GAN's discriminator incorporates two steganalyers with different tasks: distinguishing the generator's final output, i.e., the stego image, from both the input cover image and the partially embedded stego image, providing diverse guidance to the generator. In addition, we introduce a simple yet effective update strategy to ensure a stable training process. Comprehensive experiments demonstrate that our method significantly enhances security over existing symmetric steganography techniques, achieving state-of-the-art levels compared to other methods focused on embedding costs adjustments. Additionally, detailed ablation studies validate our approach's effectiveness.

Recently, facial attribute editing has drawn increasing attention and has achieved significant progress due to Generative Adversarial Network (GAN). Since paired images before and after editing are not available, existing methods typically perform the editing and reconstruction tasks simultaneously, and transfer facial details learned from the reconstruction to the editing via sharing the latent representation space and weights. In this way, they can not preserve those non-targeted regions well during editing. In addition, they usually introduce skip connections between the encoder and decoder to improve image quality at the cost of attribute editing ability. In this paper, we propose a novel method called InterGAN with high-frequency compensation to alleviate above problems. Specifically, we first propose the cross-task interaction (CTI) to fully explore the relationships between editing and reconstruction tasks. The CTI includes two translations: style translation adjusts the mean and variance of feature maps according to style features, and conditional translation utilizes attribute vector as condition to guide feature map transformation. They provide effective information interaction to preserve the irrelevant regions unchanged. Without using skip connections between the encoder and decoder, furthermore, we propose the high-frequency compensation module (HFCM) to improve image quality. The HFCM tries to collect potentially loss information from input images and each down-sampling layers of the encoder, and then re-inject them into subsequent layers to alleviate the information loss. Ablation analysis show the effectiveness of proposed CTI and HFCM. Extensive qualitative and quantitative experiments on CelebA-HQ demonstrate that the proposed method outperforms state-of-the-art methods both in attribute editing accuracy and image quality.

Since the generative adversarial network (GAN) was proposed by Ian Goodfellow et al. in 2014, it has been widely used in various fields. However, there are only a few works related to image steganography so far. Existing GAN-based steganographic methods mainly focus on the design of generator, and just assign a relatively poorer steganalyzer in discriminator, which inevitably limits the performances of their models. In this paper, we propose a novel Steganographic method based on Generative Multi-Adversarial Network (Steg-GMAN) to enhance steganography security. Specifically, we first employ multiple steganalyzers rather than a single steganalyzer like existing methods to enhance the performance of discriminator. Furthermore, in order to balance the capabilities of the generator and the discriminator during training stage, we propose an adaptive way to update the parameters of the proposed GAN model according to the discriminant ability of different steganalyzers. In each iteration, we just update the poorest one among all steganalyzers in discriminator, while update the generator with the gradients derived from the strongest one. In this way, the performance of generator and discriminator can be gradually improved, so as to avoid training failure caused by gradient vanishing. Extensive comparative results show that the proposed method can achieve state-of-the-art results compared with the traditional steganography and the modern GAN-based steganographic methods. In addition, a large number of ablation experiments verify the rationality of the proposed model.

Existing deep learning models for spoofing speech detection often struggle to effectively generalize to unseen spoofing attacks that were not present during the training stage. Moreover, the presence of class imbalance further compounds this issue by biasing the learning process towards seen attack samples. To address these challenges, we present an innovative end-to-end model called One-Class Neural Network with Directed Statistics Pooling (OCNet-DSP). Our model incorporates a feature cropping operation to attenuate high-frequency components, mitigating the risk of overfitting. Additionally, leveraging the time-frequency characteristics of speech signals, we introduce a directed statistics pooling layer that extracts more effective features for distinguishing between bonafide and spoofing classes. We also propose the Threshold One-class Softmax loss, which mitigates class imbalance by reducing the optimization weight of spoofing samples during training. Extensive comparative results demonstrate that the proposed model outperforms all existing single models, achieving an equal error rate of 0.44% and a minimum detection cost function of 0.0145 for the ASVspoof 2019 logical access database. Moreover, the proposed ensemble version, which accommodates speech inputs of varying lengths in each submodel, maintains state-of-the-art performance among reproducible ensemble models. Additionally, numerous ablation experiments, along with a cross-dataset experiment, are conducted to validate the rationality and effectiveness of the proposed model.

Current deep learning-based steganalyzers often depend on specific image dimensions, leading to inevitable adjustments in network structure when dealing with varied image sizes. This impedes their effectiveness in managing the wide range of image sizes commonly found on social media. To address this issue, our paper presents a novel steganalytic network that is optimized for fixed-size (notably,$256\times 256$) color images, but is capable of efficiently detecting stego images of arbitrary size without needing retraining or modifications to the network. Our proposed network is comprised of four modules. In the initial stem module, we calculate truncated residuals for each color channel of the input image. Diverging from existing steganalytic networks that rely on vanilla convolution, we have developed a pixel difference convolution module designed to better capture the artifacts introduced by steganography. Following this, we introduce an enhanced Transformer module with selective pooling, aimed at more effectively extracting global steganalytic features. To guarantee our network’s adaptability to different image sizes, we have developed a selective pooling strategy. This involves using global covariance pooling for fixed-size color images and spatial pyramid pooling for color images of various other sizes. This approach effectively standardizes the feature maps into uniform feature vectors. The final module is focused on classification. Extensive testing results on the ALASKA II color image dataset have demonstrated that our approach significantly improves detection performance for both fixed-size and arbitrary-size images, achieving state-of-the-art results. Additionally, we provide numerous ablation studies to confirm the effectiveness and soundness of our proposed network architecture.

In comparison to symmetric embedding, asymmetric methods generally provide better steganography security. However, the performance of existing asymmetric methods is limited by their reliance on symmetric embedding costs. In this paper, we present a novel Generative Adversarial Network (GAN)-based steganography approach that independently learns asymmetric embedding costs from scratch. Our proposed framework features a generator with a dual-branch architecture and a discriminator that integrates multiple steganalytic networks. To address the issues of model instability and non-convergence that often arise in GAN model training, we implement an adaptive strategy that updates the GAN model parameters according to the performance of multiple steganalytic networks in each iteration. Furthermore, we introduce a new adversarial loss function that effectively learns asymmetric embedding costs by utilizing features like image residuals, gradients, asymmetric embedding probability maps, and the sign of the modification map to train the dual-branch network within the generator. Our comprehensive experiments show that our method achieves state-of-the-art steganography security results, significantly outperforming existing top-performing symmetric and asymmetric methods. Additionally, numerous ablation experiments confirm the rationality of our GAN-based model design.

In this paper, we propose an end-to-end keyword spotting scheme that applies deep learning techniques in a homomorphic encryption domain. Leveraging the complex number encryption capability of homomorphic encryption algorithms, we introduce a complex neural network and design every modules of it, to accommodate the properties and characteristics of homomorphic encryption. Considering the relatively slow computation speed of homomorphic encryption, we employ convolution decomposition to reduce the cost of computational of the network, effectively minimizing computational time while preserving network performance. We have tested our proposed method and found that it significantly accelerates computation speed with minimal performance sacrifice when compared to the current state-of-the-art methods, thereby better balancing network performance and efficiency.

Frequent itemset mining is a data mining technique widely used on massive datasets. In cloud computing, the dataset may be encrypted for privacy protection. Therefore, frequent itemset mining over encrypted data is a crucial application in secure cloud computing. In this paper, we propose an effective privacy-preserving framework where the cloud server can directly perform data mining on the encrypted database without interacting with other cloud servers. We first design three security primitives to implement subset determination, accumulation, and comparison in the encrypted domain for frequent itemset mining. Based on the proposed framework, we then propose two secure protocols that allow the cloud server to perform frequent itemset mining on encrypted cloud data with these security primitives. The first protocol leaks no information to the cloud and the second protocol has the advantage of more efficient mining performance. We then present two strategies with parallel algorithms and GPU computing to accelerate the running time. We also analyze the security of our protocols and the computational complexities. Experimental results show that our serial-based protocols achieve shorter running times and higher levels of privacy than previous solutions. Our multi-CPU (or GPU) based parallel protocol can further reduce the practical running time.

The recent literature has shown that adversarial embedding has promise for enhancing the security of steganography. However, existing methods achieve the final stego mainly based on a pre-trained Convolutional Neural Network (CNN)-based steganalyzer without considering any other steganalytic features. When the steganalyzer is re-trained, its performance usually drops significantly. We propose a novel adversarial embedding method via stego generation and selection. To improve the diversity of the stego images, this method first randomly generates many candidate stegos according to the amplitudes of the gradients and embedding costs of a given cover. Since the image residuals are the commonly used low-level features in many steganalyzers, the proposed method carefully designs different adaptive high-pass filters to calculate the image residuals, and then selects a final stego from among those candidate stegos which can successfully fool the pre-trained steganalyzer, according to the residual distance between stego and the cover. Extensive experimental evaluations on re-trained CNN-based and traditional steganalyzers demonstrate that the proposed method can significantly enhance the security of the modern steganographic methods in both spatial and JPEG domains, and achieve much better performance than related adversarial embedding methods.

Up to now, most existing steganalytic methods were designed for grayscale images, and are not suitable for the color images that are widely used in social networks. In this paper, we design a universal color image steganalysis network (called UCNet) for the spatial and JPEG domains. The proposed method includes preprocessing, convolutional, and classification modules. To preserve the steganalytic features in each color channel, the preprocessing module first separates the input image into three channels based on the corresponding embedding spaces (i.e., RGB in the spatial domain, and YCbCr in the JPEG domain), and then extracts the image residuals with 62 fixed high-pass filters. Finally, all truncated residuals are concatenated for subsequent analysis, rather than adding them together in the first layer as in existing CNN-based steganalyzers. To accelerate network convergence and effectively reduce the number of parameters, the convolutional module contains three carefully designed types of layers with different shortcut connections and group convolution structures, to further learn the high-level steganalytic features. In the classification module, we employ global average pooling and a fully connected layer for classification. We conduct extensive experiments on ALASKA II to demonstrate that the proposed method can achieve state-of-the-art results that are comparable with other modern CNN-based steganalyzers (e.g., SRNet and LC-Net) in both the spatial and JPEG domains, with relatively few memory requirements and short training times. Furthermore, we also provide some necessary descriptions and carry out numerous ablation experiments to verify the rationality of the network design.

Unlike most existing steganography methods which are mainly focused on designing embedding cost, in this paper, we propose a new method to enhance existing steganographic methods via stego generation and selection. The proposed method firstly trains a steganalytic network according to the steganography to be enhanced, and then tries to adjust a tiny part of original embedding costs based on the magnitudes of it and the corresponding gradients obtained from the pre-trained network, and generates many candidate stegos in a random manner. Finally, the method selects a stego according to its image residual distance to cover. Extensive experimental results have shown that the proposed method can siginficantly enhance the security performance of current steganography in spatial domain against four steganalytic classifiers. In addition, comparative analysis between original stegos and the resulting ones with the proposed method are given.

Signal Processing in the Encrypted Domain (SPED) has received considerable attention as it aims at privacy-preserving solutions for various applications. Discrete cosine transform (DCT) is a popular signal transform widely used in signal processing. It has many applications in speech processing, still picture coding, image and video transformation, compression coding, and so on. In this paper, we mainly study how to implement DCT in the encryption domain with high precision. We propose a new scheme to implement encrypted domain DCT. This scheme encodes a complex number as a unit root polynomial in the evaluation, and realizes the high precision representation of complex numbers. With this representation, this scheme can also realize the high precision representation of decimals. To improve the computational efficiency, we also propose a fast implementation of DCT in the encryption domain, which can significantly improve the DCT speed for large-scale matrices. We conducted experiments to verify the effectiveness and efficiency. When the matrix size is small, our original method has the advantages of both high accuracy and fast speed. When the matrix size is large, the fast implementation outperforms the original method.

Line detection is an important research topic in image processing and computer vision. Hough transform is a widely used technique to detect lines. In the scenario of cloud computing, performing the Hough transform and line detection needs to consider privacy protection issues. In this paper, we propose implementing Hough transform in the encrypted domain and its application to line detection on encrypted images. Using the proposed encrypted domain Hough transform, we detect the extreme points in the encrypted parameter space. After transforming the extreme point into the encrypted spatial domain, we obtain the detected lines on the encrypted image. We conducted experiments and demonstrated the viability and effectiveness of our secure Hough transform and line detection on encrypted images.

Image steganography aims to embed secret messages into cover images in an imperceptible manner. While steganalysis tries to identify stegos from covers, which is a special binary classification problem. Recently, some literatures show that the adversarial embedding can mislead the advanced steganalyzers based on convolutional neural network (CNN), and thus enhance the steganography security. Since adding perturbations to stegos may lead to messages extraction failure due to properties of syndrome-trellis codes (STC), the existing adversarial examples are derived from covers or their enhanced versions, while those stegos are not fully utilized. In this paper, we propose a new adversarial embedding scheme for image steganography. Unlike those related works, we first combine multiple gradients of cover and generated stegos to determine the directions of cost modifications. Next, instead of adjusting all or a random part of embedding costs in existing works, we carefully select the candidate costs according to the amplitudes of cover gradients and their costs. Extensive experimental results demonstrate that by adjusting a tiny part of embedding costs (less than 5% in most cases), the proposed method can significantly improve the security of five modern steganographic methods evaluated on both re-trained CNN-based and traditional steganalyzers, and achieve much better security performances compared with related methods. In addition, the security performances evaluated on different image database show that the generalization of the proposed method is good.

Recently, convolutional neural networks (CNNs) have demonstrated superior performance on digital multimedia steganalysis. However, some studies have noted that most CNN-based classifiers can be easily fooled by adversarial examples, which form slightly perturbed inputs to a target network according to the gradients. Inspired by this phenomenon, we first introduce a novel steganography method based on adversarial examples for digital audio in the time domain. Unlike related methods for image steganography, such as [1]-[4], which are highly dependent on some existing embedding costs, the proposed method can start from a flat or even a random embedding cost and then iteratively update the initial costs by exploiting the adversarial attacks until satisfactory security performances are obtained. The extensive experimental results show that our method significantly outperforms the existing nonadaptive and adaptive steganography methods and achieves state-of-the-art results. Moreover, we also provide experimental results to investigate why the proposed embedding modifications seem evenly located at all audio segments despite their different content complexities, which is contrary to the content adaptive principle widely employed in modern steganography methods.

Most modern steganography methods focus on designing an effective cost function. To our best knowledge, there is no related works concerned about modifying stego to enhance steganography security. In this paper, therefore, we propose a novel post-processing for stego image in the spatial domain. To ensure the correct extraction of hidden message, our method restricts the modification amplitude of each pixel according to the characteristics of STCs (Syndrome-Trellis Codes). To enhance steganography security, our method traverses the stego image pixel by pixel, and modifies those pixels that can reduce the image residual difference between cover and stego under some criterion. Experimental results show that the proposed method can improve the security of current steganography especially for large payloads, e.g. larger than 0.3 bpp. In addition, the post-modification rate is rather low, for instance less than 8 \textpertenthousand \ pixels have been changed in the enhanced stego image for the five existing steganography methods for payload as large as 0.5 bpp.

Recently, deep learning based methods have achieved superior performance compared to conventional methods based on hand-crafted features in image steganalysis. However, most modern methods are usually quite time consuming. For instance, it takes over 3 days to train a state-of-the-art neural network, i.e. SRNet [3] in our experiments. In this paper, therefore, we propose a fast yet very effective convolutional neural network (CNN) for image steganalysis in spatial domain. To make a good tradeoff between training time and performance, we carefully design the architecture of the proposed network according to our extensive experiments. In addition, we first introduce the global covariance pooling into steganalysis to exploit the second-order statistic of high-level features for further improving the performance. Experimental results show that the proposed network can outperform the current best one, while its training time is significantly reduced.

As cloud storage becomes more common, concerns about the invasion of privacy are increasing. When images are stored in an encrypted form in the public cloud, reversible data hiding in the encrypted domain can be applied to embed additional data within the encrypted images for ease of management. Most existing works focus on uncompressed images and are not applicable to JPEG images, which are widely used throughout the Internet. Therefore, in this paper, a novel reversible data hiding scheme for encrypted JPEG bitstreams is proposed. First, an effective method of bitstream-based JPEG image encryption is employed to encrypt plaintext JPEG images. Then, we present a reversible data hiding technique for encrypted JPEG images based on invariant zero-run length in the zero-run value pairs. In the cloud, additional data, such as labels, timestamps, origins, and authentication messages, are directly embedded into the encrypted JPEG images with our proposed data hiding technique. From the marked encrypted JPEG images, the extraction of the hidden data and the recovery of the original images can be done independently. Extensive experiments performed on several typical images and three well-known image databases show that the proposed scheme can achieve much higher embedding capacity than that of most recent schemes, and the file sizes of the marked encrypted JPEG images are well preserved compared to those of the related methods. In addition, we provide further analysis to show that the proposed scheme has good format compatibility and low computational complexity.

To reduce the tampering artifacts and/or enhance audio quality, some audio processing operations are often applied in the resulting tampered audio. Like image forensics, the detection of various post processing operations has become very important for audio authentication. In this paper, we propose a convolutional neural network (CNN) to detect audio processing operations. In the proposed method, we carefully design the network architecture, with particular attention to the frequency representation for the audio input, the activation function and the depth of the network. In our experiments, we evaluate the proposed method on audio clips with 12 commonly used audio processing operations and of three different small sizes. The experimental results show that our method can significantly outperform related methods based on hand-crafted features and other CNN architectures, and can achieve state-of-the-art results for both binary and multiple classification.

Generative Adversarial Network (GAN) is a prominent generative model that are widely used in various applications. Recent studies have indicated that it is possible to obtain fake face images with a high visual quality based on this novel model. If those fake faces are abused in image tampering, it would cause some potential moral, ethical and legal problems. In this paper, therefore, we first propose a Convolutional Neural Network (CNN) based method to identify fake face images generated by the current best method [20], and provide experimental evidences to show that the proposed method can achieve satisfactory results with an average accuracy over 99.4%. In addition, we provide comparative results evaluated on some variants of the proposed CNN architecture, including the high pass filter, the number of the layer groups and the activation function, to further verify the rationality of our method.

Image forensics has attracted wide attention during the past decade. However, most existing works aim at detecting a certain operation, which means that their proposed features usually depend on the investigated image operation and they consider only binary classification. This usually leads to misleading results if irrelevant features and/or classifiers are used. For instance, a JPEG decompressed image would be classified as an original or median filtered image if it was fed into a median filtering detector. Hence, it is important to develop forensic methods and universal features that can simultaneously identify multiple image operations. Based on extensive experiments and analysis, we find that any image operation, including existing anti-forensics operations, will inevitably modify a large number of pixel values in the original images. Thus, some common inherent statistics such as the correlations among adjacent pixels cannot be preserved well. To detect such modifications, we try to analyze the properties of local pixels within the image in the residual domain rather than the spatial domain considering the complexity of the image contents. Inspired by image steganalytic methods, we propose a very compact universal feature set and then design a multiclass classification scheme for identifying many common image operations. In our experiments, we tested the proposed features as well as several existing features on 11 typical image processing operations and four kinds of anti-forensic methods. The experimental results show that the proposed strategy significantly outperforms the existing forensic methods in terms of both effectiveness and universality.

Digital multimedia steganalysis has attracted wide attention over the past decade. Currently, there are many algorithms for detecting image steganography. However, little research has been devoted to audio steganalysis. Since the statistical properties of image and audio files are quite different, features that are effective in image steganalysis may not be effective for audio. In this article, we design an improved audio steganalytic feature set derived from both the time and Mel-frequency domains for detecting some typical steganography in the time domain, including LSB matching, Hide4PGP, and Steghide. The experiment results, evaluated on different audio sources, including various music and speech clips of different complexity, have shown that the proposed features significantly outperform the existing ones. Moreover, we use the proposed features to detect and further identify some typical audio operations that would probably be used in audio tampering. The extensive experiment results have shown that the proposed features also outperform the related forensic methods, especially when the length of the audio clip is small, such as audio clips with 800 samples. This is very important in real forensic situations.

In recent years, deep learning has achieved breakthrough results in various areas, such as computer vision, audio recognition, and natural language processing. However, just several related works have been investigated for digital multimedia forensics and steganalysis. In this paper, we design a novel CNN (convolutional neural networks) to detect audio steganography in the time domain. Unlike most existing CNN based methods which try to capture media contents, we carefully design the network layers to suppress audio content and adaptively capture the minor modifications introduced by ±1 LSB based steganography. Besides, we use a mix of convolutional layer and max pooling to perform subsampling to achieve good abstraction and prevent over-fitting. In our experiments, we compared our network with six similar network architectures and two traditional methods using handcrafted features. Extensive experimental results evaluated on 40,000 speech audio clips have shown the effectiveness of the proposed convolutional network.

Image inpainting, an image processing technique for restoring missing or damaged image regions, can be utilized by forgers for removing objects in digital images. Since no obviously perceptible artifacts are left after inpainting, it is necessary to develop methods for detecting the presence of inpainting. In general, there are two main categories of image inpainting techniques: exemplar-based and diffusion-based techniques. Although several methods have been proposed for detecting exemplar-based inpainting, there is still no effective method for detecting diffusion-based inpainting. Usually, the tampered regions manipulated by diffusion-based inpainting techniques are much smaller than those manipulated by exemplar-based ones, presenting more challenges in detecting these regions. As a pioneering attempt, this paper proposes a method for the localization of diffusion-based inpainted regions in digital images. We first analyze the diffusion process in inpainting, and observe that the changes in the image Laplacian along the direction perpendicular to the gradient are different in the inpainted and untouched regions. Following this observation, we construct a feature set based on the intra-channel and inter-channel local variances of the changes to identify the inpainted regions. Finally, two effective post-processing operations are designed for further refining of the localization result. The extensive experimental results evaluated on both synthetic and realistic inpainted images show the effectiveness of the proposed method.

Over the past decade, many efforts have been made in passive image forensics. Although it is able to detect tampered images at high accuracies based on some carefully designed mechanisms, localization of the tampered regions in a fake image still presents many challenges, especially when the type of tampering operation is unknown. Some researchers have realized that it is necessary to integrate different forensic approaches in order to obtain better localization performance. However, several important issues have not been comprehensively studied, for example, how to select and improve/readjust proper forensic approaches, and how to fuse the detection results of different forensic approaches to obtain good localization results. In this paper, we propose a framework to improve the performance of forgery localization via integrating tampering possibility maps. In the proposed framework, we first select and improve two existing forensic approaches, i.e., statistical feature-based detector and copy-move forgery detector, and then adjust their results to obtain tampering possibility maps. After investigating the properties of possibility maps and comparing various fusion schemes, we finally propose a simple yet very effective strategy to integrate the tampering possibility maps to obtain the final localization results. The extensive experiments show that the two improved approaches used in our framework significantly outperform the state-of-the-art techniques, and the proposed fusion results achieve the best F1-score in the IEEE IFS-TC Image Forensics Challenge.

It is conventionally assumed that steganographic schemes for gray-scale images can be directly applied to color images by embedding messages independently in different color channels. However, the correlation among color channels may be disturbed and it is unclear how to preserve the channel correlation so as to increase empirical security. In this paper, we propose a strategy called CMD-C (clustering modification directions for color components). The basic idea of the strategy is to change different color components from the same pixel location towards a positive or negative direction consistently. To implement the strategy, we decompose an image into several sub-images in which segmented hidden message bits are successively embedded. The embedding costs of a sub-image are computed by considering the correlation both within and among color channels. Experimental results show that the proposed CMD-C strategy has made great improvement over conventional methods in resisting state-of-the-art steganalytic methods.

In modern steganography, embedding modifications are highly concentrated on the textural regions within an image, as such regions are difficult to model for steganalysis. Previous studies have shown that compared with non-adaptive strategies, this content adaptive strategy achieves stronger security against existing steganalysis. Based on the experiments and analyses, however, we found that this embedding property would inevitably lead to a large limitation in existing adaptive steganography. That is, it is possible for steganalyzers to estimate the regions that have probably been modified after data hiding. In this paper, we propose an adaptive steganalytic scheme based on embedding probabilities of pixels. The main idea of our scheme is that we assign different weights to different pixels in feature extraction. For those pixels with high embedding probabilities, their corresponding weights are larger, since they should contribute more to steganalysis and vice versa. By doing so, we can concentrate our attention on the regions that have probably been modified and significantly reduce the impact of other unchanged smooth regions. It is expected that our proposed method is an improvement on the existing steganalytic methods, which usually assume every pixel has the same contribution to steganalysis. The extensive experiments evaluated on four typical adaptive steganographic methods have shown the effectiveness of the proposed scheme, especially for low embedding rates, for example, lower than 0.20 bpp.

Due to the popularity of JPEG compression standard, JPEG images have been widely used in various applications. Nowadays, detection of JPEG forgeries becomes an important issue in digital image forensics, and lots of related works have been reported. However, most existing works mainly rely on a pre-trained classifier according to the quantization table shown in the file header of the suspicious JPEG image, and they assume that such a table is authentic. This assumption leaves a potential flaw for those wise forgers to confuse or even invalidate the current JPEG forensic detectors. Based on our analysis and experiments, we found that the generalization ability of most current JPEG forensic detectors is not very good. If the quantization table changes, their performances would decrease significantly. Based on this observation, we propose a universal anti-forensic scheme via replacing the quantization table adaptively. The extensive experimental results evaluated on 10,000 natural images have shown the effectiveness of the proposed scheme for confusing four typical JPEG forensic works.

Image forensics have made great progress during the past decade. However, almost all existing forensic methods can be regarded as the specific way, since they mainly focus on detecting one type of image processing operations. When the type of operations changes, the performances of the forensic methods usually degrade significantly. In this paper, we propose a universal forensics strategy based on steganalytic model. By analyzing the similarity between steganography and image processing operation, we find that almost all image operations have to modify many image pixels without considering some inherent properties within the original image, which is similar to what in steganography. Therefore, it is reasonable to model various image processing operations as steganography and it is promising to detect them with the help of some effective universal steganalytic features. In our experiments, we evaluate several advanced steganalytic features on six kinds of typical image processing operations. The experimental results show that all evaluated steganalyzers perform well while some steganalytic methods such as the spatial rich model (SRM) [4] and LBP [19] based methods even outperform the specific forensic methods significantly. What is more, they can further identify the type of various image processing operations, which is impossible to achieve using the existing forensic methods.

WOW (Wavelet Obtained Weights) [5] is one of the advanced steganographic methods in spatial domain, which can adaptively embed secret message into cover image according to textural complexity. Usually, the more complex of an image region, the more pixel values within it would be modified. In such a way, it can achieve good visual quality of the resulting stegos and high security against typical steganalytic detectors. Based on our analysis, however, we point out one of the limitations in the WOW embedding algorithm, namely, it is easy to narrow down those possible modified regions for a given stego image based on the embedding costs used in WOW. If we just extract features from such regions and perform analysis on them, it is expected that the detection performance would be improved compared with that of extracting steganalytic features from the whole image. In this paper, we first proposed an adaptive steganalytic scheme for the WOW method, and use the spatial rich model (SRM) based features [4] to model those possible modified regions in our experiments. The experimental results evaluated on 10,000 images have shown the effectiveness of our scheme. It is also noted that our steganalytic strategy can be combined with other steganalytic features to detect the WOW and/or other adaptive steganographic methods both in the spatial and JPEG domains.

Bit rate is one of the important criterions for digital video quality. With some video tools, however, video bit rate can be easily increased without improving the video quality at all. In such a case, a claimed high bit rate video would actually have poor visual quality if it is up-converted from an original lower bit rate version. Therefore, exposing fake bit rate videos becomes an important issue for digital video forensics. To the best of our knowledge, although some methods have been proposed for exposing fake bit rate MPEG-2 videos, no relative work has been reported to further estimate their original bit rates. In this paper, we first analyze the statistical artifacts of these fake bit rate videos, including the requantization artifacts based on the first-digit law in the DCT frequency domain (12-D) and the changes of the structural similarity indexes between the query video and its sequential bit rate down-converted versions in the spatial domain (4-D), and then we propose a compact yet very effective 16-D feature vector for exposing fake bit rate videos and further estimating their original bit rates. The extensive experiments evaluated on hundreds of video sequences with four different resolutions and two typical compression schemes (i.e., MPEG-2 and H.264/AVC) have shown the effectiveness of the proposed method compared with the existing relative ones.

Audio signal is sometimes stored and/or processed in WAV (waveform) format without any knowledge of its previous compression operations. To perform some subsequent processing, such as digital audio forensics, audio enhancement and blind audio quality assessment, it is necessary to identify its compression history. In this article, we will investigate how to identify a decompressed wave audio that went through one of three popular compression schemes, including MP3, WMA (windows media audio) and AAC (advanced audio coding). By analyzing the corresponding frequency coefficients, including modified discrete cosine transform (MDCT) and Mel-frequency cepstral coefficients (MFCCs), of those original audio clips and their decompressed versions with different compression schemes and bit rates, we propose several statistics to identify the compression scheme as well as the corresponding bit rate previously used for a given WAV signal. The experimental results evaluated on 8,800 audio clips with various contents have shown the effectiveness of the proposed method. In addition, some potential applications of the proposed method are discussed.

Video bitrate, as one of the important factors that reflect the video quality, can be easily manipulated via some video editing softwares. In some forensic scenarios, for example, video uploaders of video-sharing websites may increase video bitrate for seeking more commercial profits. In this paper, we try to detect those fake high bitrate videos, and then further to estimate their original bitrates. The proposed method is mainly based on the fact that if the video bitrate has been increased with the help of video editing software, its essential video quality will not increase at all. By analyzing the quality of the questionable video and a series of its re-encoded versions with different lower bitrates, we can obtain a feature curve to measure the change of the video quality, and then we propose a compact feature vector (3-D) to expose fake bitrate videos and their original bitrates. The experimental results evaluated on both CIF and QCIF raw sequences have shown the effectiveness of the proposed method.

In this paper, we present a new framework for designing distortion functions of joint photographic experts group (JPEG) steganography with uncompressed side-image. In our framework, the discrete cosine transform (DCT) coefficients, including all direct current (DC) coefficients and alternating current (AC) coefficients, are divided into two groups: first-priority group (FPG) and second-priority group (SPG). Different strategies are established to associate the distortion values to the coefficients in FPG and SPG, respectively. In this paper, three scenarios for dividing the coefficients into FPG and SPG are exemplified, which can be utilized to form a series of new distortion functions. Experimental results demonstrate that while applying these generated distortion functions to JPEG steganography, the intrinsic statistical characteristics of the carrier image will be preserved better than the prior-art, and consequently the security performance of the corresponding JPEG steganography can be improved significantly.

Compression history identification plays a very important role in digital multimedia forensics. However, most existing literatures mainly focus on digital image forensics, and just a few works consider digital audio. In this paper, we investigate two popular compression schemes in digital audio, that is, MP3 and WMA, and try to reveal the compression history for a questionable audio signal in the original uncompressed WAV format via analyzing some statistical characteristics of the modified discrete cosine transform coefficients of the audio. The extensive experimental results have shown that the proposed method can effectively identify whether the given audio has been previously compressed with MP3 and/or WMA, and can further estimate the hidden compression rates, even the compression rate is as high as 128 K bps (bits per second).

The quantization artifacts and blocking artifacts are the two significant properties in the JPEG compressed images. Most relative forensic techniques usually use such inherent properties to provide some evidences on how image data is acquired and/or processed. A wise attacker, however, may perform some post-operations to confuse the two artifacts to fool current forensic techniques. Recently, Stamm et al. in [1] propose a novel anti-JPEG compression method via adding anti-forensic dither to the DCT coefficients and further reducing the blocking artifacts. In this paper, we found that the dithering operation will inevitably destroy the statistical correlations among the 8 × 8 intrablock and interblock within an image. In the view of JPEG steganalysis, we employ the transition probability matrix of the DCT coefficients to measure such modifications for identifying the forged images from those original JPEG decompressed images and uncompressed ones. On average, we can obtain a detection accuracy as high as 99% on the image database of UCID [2].

Video sequence matching aims to locate a query video clip in a video database. It plays an important role in reducing storage redundancy and detecting video copies for copyright protection. In this paper, we propose an effective method for video sequence matching based on the invariance of color correlation. The proposed method first splits each key-frame into nonoverlapping blocks. For each block, we sort the red, green, and blue color components according to their average intensities, and use the percentage of the color correlation to generate a frame feature with a small size. Finally, the resulting video feature is made up of the consecutive frame features, which is demonstrated to be robust against most typical video content-preserving operations, including geometric distortion, blurring, noise contamination, contrast enhancement, and strong re-encoding. The experimental results show that the proposed method outperforms the existing methods in the literature, as well as the method based on the traditional color histogram. Furthermore, the time and space complexity of our algorithm are both satisfactory, which are very important for many real-time applications.

Recently, a new high-performance JPEG steganography with a complementary embedding strategy (JPEG-CES) was presented. It can disable many specific steganalysers such as the Chi-square family and S family detectors, which have been used to attack J-Steg, JPHide, F5 and OutGuess successfully. In this work, a study on the security performance of JPEG-CES is reported. Our theoretical analysis demonstrates that in this algorithm, the number of the different quantised discrete cosine transform (qDCT) coefficients and the symmetry of the qDCT coefficient histogram both will be disturbed when the secret message is embedded. Moreover, the intrinsic sign and magnitude dependencies that existed in intra-block and inter-block qDCT coefficients will be disturbed too. Thus it may be detected by some modern universal steganalysers which can catch these disturbances. In this work, the authors have proposed two new steganalytic approaches. Through exploring the distortions that have been introduced into the qDCT coefficient histogram and the dependencies existed in the intra-block and inter-block sense, respectively, these two alternative steganalysers can detect JPEG-CES effectively. In addition, via merging the features of these two steganalysers, a more reliable classifier can be obtained.

Although many existing steganalysis works have shown that the spatial ±1 steganography on JPEG pre-compressed images is relatively easier to be detected compared with that on the never-compressed images, most experimental results seem not very convincing since these methods usually assume that the quantization table of the JPEG stegos previously used is known before detection and/or the length of embedded message is fixed. Furthermore, there are just few effective quantitative algorithms for further estimating the spatial modifications. In this letter, we firstly propose an effective method to detect the quantization table from the contaminated digital images which are originally stored as JPEG format based on our recently developed work about JPEG compression error analysis , and then we present a quantitative method to reliably estimate the length of spatial modifications in those gray-scale JPEG stegos by using data fitting technology. The extensive experimental results show that our estimators are very effective, and the order of magnitude of prediction error can remain around measured by the mean absolute difference.

The least-significant-bit (LSB)-based approach is a popular type of steganographic algorithms in the spatial domain. However, we find that in most existing approaches, the choice of embedding positions within a cover image mainly depends on a pseudorandom number generator without considering the relationship between the image content itself and the size of the secret message. Thus the smooth/flat regions in the cover images will inevitably be contaminated after data hiding even at a low embedding rate, and this will lead to poor visual quality and low security based on our analysis and extensive experiments, especially for those images with many smooth regions. In this paper, we expand the LSB matching revisited image steganography and propose an edge adaptive scheme which can select the embedding regions according to the size of secret message and the difference between two consecutive pixels in the cover image. For lower embedding rates, only sharper edge regions are used while keeping the other smoother regions as they are. When the embedding rate increases, more edge regions can be released adaptively for data hiding by adjusting just a few parameters. The experimental results evaluated on 6000 natural images with three specific and four universal steganalytic algorithms show that the new scheme can enhance the security significantly compared with typical LSB-based approaches as well as their edge adaptive ones, such as pixel-value-differencing-based approaches, while preserving higher visual quality of stego images at the same time.

JPEG is one of the most extensively used image formats. Understanding the inherent characteristics of JPEG may play a useful role in digital image forensics. In this paper, we introduce JPEG error analysis to the study of image forensics. The main errors of JPEG include quantization, rounding, and truncation errors. Through theoretically analyzing the effects of these errors on single and double JPEG compression, we have developed three novel schemes for image forensics including identifying whether a bitmap image has previously been JPEG compressed, estimating the quantization steps of a JPEG image, and detecting the quantization table of a JPEG image. Extensive experimental results show that our new methods significantly outperform existing techniques especially for the images of small sizes. We also show that the new method can reliably detect JPEG image blocks which are as small as 8 × 8 pixels and compressed with quality factors as high as 98. This performance is important for analyzing and locating small tampered regions within a composite image.

Quantization is one of the commonly used techniques in most lossy image source encoders. It is observed that the quantization operation usually introduces some obvious artifacts into the histogram of the corresponding transform coefficients under various compression schemes. By investigating such inherent artifacts over all candidate transform coefficients, it is possible to identify the transform, as well as some parameters previously employed in the transform-based encoder from a decompressed image. In this paper, we first analyze the properties of the quantized coefficients and present a simple yet effective way to detect the quantization artifacts, and then we propose an approach to identify the transform-based encoder based on the quantization artifacts detection. The simulation results evaluated on thousands of natural images with some popular compression schemes demonstrate the effectiveness of our method.

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

One of the most common practices in image tampering involves cropping a patch from a source and pasting it onto a target. In this paper, we present a novel method for the detection of such tampering operations in JPEG images. The lossy JPEG compression introduces inherent blocking artifacts into the image and our method exploits such artifacts to serve as a 'watermark' for the detection of image tampering. We develop the blocking artifact characteristics matrix (BACM) and show that, for the original JPEG images, the BACM exhibits regular symmetrical shape; for images that are cropped from another JPEG image and re-saved as JPEG images, the regular symmetrical property of the BACM is destroyed. We fully exploit this property of the BACM and derive representation features from the BACM to train a support vector machine (SVM) classifier for recognizing whether an image is an original JPEG image or it has been cropped from another JPEG image and re-saved as a JPEG image. We present experiment results to show the efficacy of our method.