Image steganography has emerged as a promising channel for transmitting task assignments, authentication credentials, and scheduling metadata across distributed cloud nodes without raising suspicion. However, existing steganography methods primarily focus on robustness against attacks while overlooking the credibility of the extracted information. In distributed cloud environments, blindly accepting incorrect or manipulated data can propagate faults across nodes, trigger inconsistent system states, or even cause cascading service failures. To address this, we propose C2IS, a Credibility-Aware Coverless Image Steganography framework designed for distributed cloud services. C2IS enables cloud nodes not only to extract hidden information, but also to assess its credibility before execution or propagation. Specifically, we propose a polar harmonic transform-based feature extraction and selection strategy that extracts both coarse-grained and fine-grained features. The coarse-grained features, characterized by their high stability under various attacks, are utilized to carry secret information. Meanwhile, the fine-grained features quantify the degree of feature perturbation to assess information credibility. Furthermore, we develop a double median thresholding-based hash mapping algorithm, which binarizes inter-image feature differences using the median, thereby significantly enhancing the diversity of the generated hash sequences. Extensive experimental results show that our method achieves a complete hash sequence length of up to 13 bits, enabling higher hiding capacity. Meanwhile, under various attacks, it exhibits stronger robustness than state-of-the-art methods and can accurately determine whether the extracted information contains errors, providing a trustworthy means of covert communication for distributed cloud services.

Deep learning-based watermarking technology has made significant success due to its excellent robustness and convenient traceability. However, previous studies rarely considered users’ demands for high-fidelity images and file size growth rates, especially in scenarios involving the watermarking of a large number of high-resolution images. This paper proposes a high fidelity and robust watermarking using deep reinforcement learning. Specifically, we propose a self-optimization module that utilizes a dueling Q-learning network to select the optimal watermark embedding patch, which minimizes the quality impact after adding a watermark to the image. Additionally, we incorporate a convolutional block attention module (CBAM) into encoder and decoder networks to enhance feature learning in both frequency and spatial dimensions. Experimental results demonstrate that our method significantly enhances image fidelity, achieving PSNR values above 54dB across different datasets. The method also exhibits good robustness against common image manipulations, including both non-geometric and geometric attacks. Furthermore, our method achieves an FSVR value below 0.5, reducing file size inflation by a factor of 34 compared to existing advanced methods. This makes it very friendly for storing a large number of high-resolution watermarked images.

The rapid spread of generative content challenges copyright protection and content identification. Visible watermarks, as a conventional mechanism for ownership attribution, are increasingly required in generated content. However, their localized and fragile nature makes them easily removable, especially via diffusion-based inpainting, which can seamlessly erase watermarks and reconstruct background regions. Although prior studies have explored adversarial perturbations to hinder such inpainting, they suffer from several limitations: 1) trade-off between effectiveness and imperceptibility; 2) vulnerability to diffusion-based purification attacks; and 3) irreversible perturbation injection, which prevents subsequent image modification. To address these issues and enhance visible watermark protection, we propose a localized perturbation method in the latent space, alleviating the trade-off between perturbation effectiveness and imperceptibility in pixel space. To implement perturbation injection, we apply reversible transformations to the latent vectors and leverage natural images as keys, making the perturbations more resistant to diffusion-based purification attacks without adversarial training. Moreover, authorized users can achieve near-lossless recovery by removing the perturbation from the latent space with the correct key. We further explore its potential application to generative video watermark protection. Experiments show that our method achieves strong protection effectiveness and imperceptibility on natural images as well as generative content, making it more suitable for practical applications. Code is available at https://github.com/charles335cs/MarkShield.

In reversible data hiding (RDH) in the plaintext domain, the reversibility of the data and the image is the greatest strength but also comes with limitations, such as low embedding capacity and weak generalization ability. These limitations make it challenging for RDH to be applied in scenarios that require the concealment of high-capacity data. To address these issues, we propose a compatible reversible data hiding with high capacity and generalization (CRDH), which can perform a second embedding based on all existing RDH methods and the two extractions are independent of each other. The nearest-neighbor interpolation (NNI) algorithm and integer wavelet transform are initially designed to create additional redundancy room, diverging from existing RDH methods that typically exploit the inherent redundancy within the image itself. Following this, we derive a novel method to prevent pixel value overflow or underflow, which is employed to guide the data embedding process. In the experimental results on standard test images, the average maximum embedding capacity of the CRDH method reaches 4.41 bits per pixel (BPP), which is 1.98 times that of other methods. As the embedded data increases, the peak signal-to-noise ratio (PSNR) of CRDH’s stego-images becomes higher compared to other methods. Furthermore, CRDH exhibits a significantly superior generalization ability in terms of both capacity and quality compared to state-of-the-art RDH methods.

To avoid the pixel modification traces left by steganography from being detected by passive steganalysis, and to prevent the hidden data from being destroyed by active steganalysis attacks, Coverless Image Steganography (CIS) that does not modify pixels has attracted widespread attention. However, most existing CIS methods are limited in their maximum capacity due to insufficient diversity in their hash sequences. In addition, these methods struggle to maintain high robustness against both geometric and non-geometric attacks simultaneously. To address these two issues, a new coverless image steganography method is proposed to enhance CIS methods’ applicability, security, and robustness in highly insecure networks. During the hiding process, hash sequences are generated by a SHA-256 algorithm that integrates inter-block and inter-channel fusion, providing higher diversity than other CIS methods. Consequently, the proposed CIS method achieves higher capacity on publicly available datasets. During the extraction process, an evaluation metric that combines visual and histogram similarity is designed to improve the accuracy of inverse image retrieval. The experimental results demonstrate that the proposed CIS method achieves capacity increases of 26.73% and 38.34% over other CIS methods on the VOC and COCO datasets, respectively. Moreover, this method exhibits nearly 100% robustness against common active steganalysis.

Orthogonal Moment-based Robust Reversible Watermarking (OM-RRW) is crucial for intellectual property protection, providing the dual benefits of robustness and reversibility. However, OM-RRW embeds watermarks into visually sensitive global low-frequency features, which easily leads to ring-like distortions that expose watermark locations, making them vulnerable to removal through image inpainting. To address this issue, this paper makes the first attempt to introduce an innovative strategy to eliminate these visible distortions, thereby overcoming OM-RRW's inherent limitations. The strategy innovates on two fronts: first, it customizes varying embedding step sizes based on the stability differences of moment values to minimize distortion; second, it designs a texture-aware adaptive basis function fine-tuning strategy. This strategy adjusts the representation capability of the basis functions in different regions based on the human eye's sensitivity to various texture areas, helping to avoid visible ring-like distortions. The performance of the proposed method is evaluated using Polar Harmonic Transform (PHT) moments, comprising three moments that exhibit remarkable performance in existing OM-RRW methods. Extensive experiments show that the proposed method can embed 128-bit watermarks with no visible distortions while minimizing the loss of robustness. In addition, this paper finds that OM-RRW demonstrates satisfactory robustness against VAE watermark removal attacks.

The construction of deep neural networks depends on a significant number of parameters and computational complexity, which poses a challenge in the field of image processing. To address the issue of the Transformer network model's large size and inability to effectively capture local features of the image, this paper proposes a lightweight composite Transformer structure that combines a spectral feature refinement module (SFRM) and a parameterless attention augmentation module (PAAM). The SFRM and PAAM work together to improve the quality of the spectral features used in the transformer. The proposed structure aims to enhance the performance of the transformer without adding unnecessary complexity. The SFRM utilises the two-dimensional discrete cosine transform to convert the image from the spatial domain to the frequency domain. This process extracts both the overall image structure and detailed feature information from the high-frequency and low-frequency regions, respectively. The aim is to purify the spatially-insignificant features in the original image. The PAAM introduces a parameter-free channel, spatial, and 3D attention enhancement mechanism to extract correlation features of local information in the spatial domain without increasing the number of parameters. This improves the expression of local features in the image. Additionally, Depth Separable (DConv MLP) is introduced to further reduce the network model's weight. The experimental results show that the proposed algorithm achieves an accuracy of 79.6% on the ImageNet-1K dataset, 91.6% on the Oxford 102 Flower Dataset, and 94.1% on the CIFAR-10 dataset. Compared to ViT-B, Swin-T, and CSwin-T, respectively, the number of covariates decreases by 86.11%, 58.62%, and 47.83%. The number of parameters is also lower than VGG-16 and ResNet-110 by 91.07% and 77.70%, respectively.

Few-shot learning, which aims to identify new classes with few samples, is an increasingly popular and crucial research topic in the machine learning . Recently, the development of deep learning has deepened the network structure of a few-shot model, thereby obtaining deeper features from the samples. This trend led to an increasing number of few-shot learning models pursuing more complex structures and deeper features. However, discarding shallow features and blindly pursuing the depth of sample feature levels is not reasonable. The features at different levels of the sample have different information and characteristics. In this paper, we propose a few-shot image classification model based on deep and shallow feature fusion and a coarse-grained relationship score network (HFFCR). First, we utilize networks with different depth structures as feature extractors and then fuse the two kinds of sample features. The fused sample features collect sample information at different levels. Second, we condense the fused features into a coarse-grained prototype point. Prototype points can better represent the information in this class and improve classification efficiency. Finally, we construct a relationship score network, concatenating the prototype points and query samples into a feature map and sending it into the network to calculate the relationship score. The classification criteria for learnable relationship scores reflect the information difference between the two samples. Experiments on three datasets show that HFFCR has advanced performance.

Medical images play a crucial role in doctors' clinical diagnosis and treatment. However, the transmission and sharing of such private information raises security concerns. To address this issue, image hiding is used as an effective technique to protect images. To achieve large hiding capacity, lossless recovery and anti-steganalysis, we propose a novel two-stage medical image-hiding method in this paper. In the first stage, a QR code for the patient diagnosis information (PDI) is generated and embedded into a secret medical image using reversible data hiding. In the second stage, the secret medical image containing PDI is hidden in a natural target image. A kind of lossless compression technique named soft compression is innovatively introduced in two hiding stages, to ensure that the reconstructed secret medical image and PDI are exactly identical to the original ones. Moreover, an adaptiven-LSB model is proposed to improve the stego image quality. Extensive experimental results show that our method achieves a PSNR of over 40dB for the stego image at 2 BPP while recovering the PDI and secret medical image with 100% accuracy on the DIV2K, COCO and ImageNet datasets. It outperforms other state-of-the-art methods in terms of hiding invisibility, recovery accuracy and security.

The goal of multiple-image hiding is to hide several secret images within another carrier image without significantly changing its appearance, and then perfectly reconstruct all of the secret images. The challenge is to ensure that the stego-image has great visual quality and can resist various steganalysis under the premise of hiding as much information as possible in one image. To address this issue, the majority of known image-hiding methods focus on hiding images using compression techniques. In this article, we present a novel multiple-image hiding method based on up-sampling and reversible color transformation. First, the interpolation algorithm up-samples the carrier image, so that the attribute of similar neighboring pixel values in the up-sampled image can significantly improve the effect of image hiding. The embedding procedure is then performed using the proposed Euclidean Distance (ED)-based block matching and reversible color transformation, which decreases the chance of local blurring in the stego-image. Experimental results show that the proposed method surpasses existing advanced methods by achieving an average of 33 dB and 28 dB of PSNR for the stego-image with a hiding capacity 2 BPP and 8 BPP, and obtaining 100% reconstructing accuracy for all secret images. It also has a high level of resistance to steganalysis and a strong robustness against various image-processing attacks.

In the field of image steganography, quality, security, and capacity emerge as three crucial aspects for ensuring the security of images stored on cloud servers. However, many existing methods fail to strike a balance on these three aspects according to the various requirements of image users. To solve this issue, we propose an Adaptable Image Steganography Model (AISM) capable of customizing suitable steganography strategies for different user requirements. Initially, AISM customizes appropriate down-sampling methods, ratios, and up-sampling ratios for secret/cover images based on requirements, followed by the image sampling process. Following that, an embedding algorithm based on pixel-value coding is proposed, which maps pixel values from [0, 255] to [-9, 9] and then replaces the high-frequency sub-band coefficients of the up-sampled image. During the embedding process, no auxiliary information is generated, which is a key aspect for user-friendliness. Extensive experimental results demonstrate that our method is capable of customizing satisfactory steganography strategies for various user requirements. Moreover, our method outperforms many state-of-the-art methods in terms of quality and security, i.e., lossless recovery of secret images, PSNR of 40 dB for stego-images, evasion from detection by six classic steganalysis tools.

In federated learning with differential privacy, an obvious phenomenon of local gradient sparsification emerges in some training rounds. When training with low privacy budgets, there is a risk of excessive noise being introduced into the uploaded gradients, leading to a significant decrease in the accuracy of the global model. To tackle the trade-off between privacy protection and model accuracy with low privacy budgets, we propose a differential privacy federated aggregation method based on gradient sparsification (DP-FedAGS), which not only prevents excessive noise addition by protecting only significant gradients, but also accelerates global model convergence by dynamically calculating the weight of the gradient. Experimental results indicate that DP-FedAGS achieves comparable privacy protection to DP-FedAvg and cpSGD, while outperforming DPFedSNLC. Moreover, our approach respectively attains an approximate average test accuracy improvement of $2 .45 \%, 4 . 79$% and $0 . 29$% over the above three methods, rendering DP-FedAGS a promising approach for exploring a balance between privacy protection and model accuracy.

In edge networks, distributed computing resources have been widely utilized to collaboratively perform a machine learning task by multiple nodes. However, the model training time in heterogeneous edge networks is becoming longer because of excessive computation and delay caused by slow nodes, namely, stragglers. The parameter server even abandons stragglers which fail to return the outcome within a reasonable deadline, called straggler dropout, decreasing the model accuracy. To optimize the computation cost and maintain the model accuracy, we focus on mitigating the heavy computation of stragglers and preventing straggler dropout. Therefore, we propose a novel scheme named dynamic grouping and heterogeneity-aware gradient coding (DGH-GC) to tolerate stragglers by employing dynamic grouping and gradient coding. DGH-GC evenly distributes stragglers in each group and encodes gradients based on their computation capacity to prevent them drop out. However, DGH-GC exacerbates the communication burden by making data duplication to tolerate stragglers. Relying on the scheme, we further propose an algorithm called DGH-(GC)2 to compress transferred gradients in both upstream communication and downstream communication. Experimental evaluations prove that DGH-(GC) outperforms all state-of-the-art methods and DGH-(GC)2 further speeds up the convergence time of the trained model and saves about 26% average iteration time compared to the DGH-(GC).

Cleanliness of city streets has an important impact on city environment and public health. Conventional street cleaning methods involve street sweepers going to many spots and manually confirming if the street needs to be clean. However, this method takes a substantial amount of manual operations for detection and assessment of street’s cleanliness which leads to a high cost for cities. Using pervasive mobile devices and AI technology, it is now possible to develop smart edge-based service system for monitoring and detecting the cleanliness of streets at scale. This paper explores an important aspect of cities — how to automatically analyze street imagery to understand the level of street litter. A vehicle (i.e. trash truck) equipped with smart edge station and cameras is used to collect and process street images in real time. A deep learning model is developed to detect, classify and analyze the diverse types of street litters such as tree branches, leaves, bottles and so on. In addition, two case studies are reported to show its strong potential and effectiveness in smart city systems.

Due to the low quality of the existing online graph partition algorithm, the graph partition problem is solved through the Cat Swarm Optimization (CSO) algorithm to improve the partition quality. To avoid falling into the local optimum with CSO, an improved graph partition approach based on Cat Swarm Optimization with the Cauchy mutation and the Inertia weight (CICSO) was proposed. CICSO adopts the Cauchy mutation to update the optimal position, which can increase the accuracy of graph partition. Meanwhile, the self-adaptive inertia weight with the dynamic change is introduced in the tracking mode to increase the convergence speed and stability. Experimental results show that the improved cat algorithm CICSO has better performance than the standard cat algorithm in terms of partition quality and convergence time, compared with the LDG, FENNEL, and the standard CSO.