Search Results (3,353)

In industrial inspection and experimental data-acquisition scenarios, the accuracy and efficiency of digital tubes, which are commonly used display components, directly affect the intelligence of the system. However, models trained on data from specific environments may experience a significant drop in recognition accuracy when applied to different environments derived from impacts of various specific scenarios (e.g., temperature changes, changes in light intensity, changes in rate, and color contrast between equipment displays and environments, among others), which may affect model accuracy. To ensure recognition accuracy, we may need to collect data from specific environments to retrain the model for each specific environment, but manual annotation is often inefficient. To address these issues, this article proposes a solution integrating image processing with deep learning within specific scenarios, encompassing the entire workflow from data acquisition to model training. Employing image processing techniques to provide high-quality training data for models, we construct a semi-supervised adversarial learning framework based on an improved self-training algorithm. The framework employs the k-means clustering algorithm for stratified sampling preparation, adds the Squeeze-and-Excitation B Block to the Convolutional Neural Network backbone, and employs the Adversarial Generative Adversarial Network to generate adversarial examples for adversarial training, thus enhancing both classification accuracy and robustness. Full article

Dynamic heterogeneous redundancy (DHR) architectures combine heterogeneity, redundancy, and dynamism to create security-centric frameworks that can be used to mitigate network attacks that exploit unknown vulnerabilities. However, conventional DHR architectures rely on centralized control modules for scheduling and adjudication, leading to significant single-point failure risks and trust bottlenecks that severely limit their deployment in security-critical scenarios. To address these challenges, this paper proposes a decentralized DHR architecture based on the Raft consensus algorithm. It deeply integrates the Raft consensus mechanism with the DHR execution layer to build a consensus-centric control plane and designs a dual-log pipeline to ensure all security-critical decisions are executed only after global consistency via Raft. Furthermore, we define a multi-dimensional attacker model—covering external, internal executor, internal node, and collaborative Byzantine adversaries—to analyze the security properties and explicit defense boundaries of the architecture under Raft’s crash-fault-tolerant assumptions. To assess the effectiveness of the proposed architecture, a prototype consisting of five heterogeneous nodes was developed for thorough evaluation. The experimental results show that, for non-Byzantine external and internal attacks, the architecture achieves high detection and isolation rates, maintains high availability, and ensures state consistency among non-malicious nodes. For stress tests in which a minority of nodes exhibit Byzantine-like behavior, our prototype preserves log consistency and prevents incorrect state commitments; however, we explicitly treat these as empirical observations under a restricted adversary rather than a general Byzantine fault tolerance guarantee. Performance testing revealed that the system exhibits strong security resilience in attack scenarios, with manageable performance overhead. Instead of turning Raft into a Byzantine-fault-tolerant consensus protocol, the proposed architecture preserves Raft’s crash-fault-tolerant guarantees at the consensus layer and achieves Byzantine-resilient behavior at the execution layer through heterogeneous redundant executors and majority-hash validation. To support evaluation during peer review, we provide a runnable prototype package containing Docker-based deployment scripts, pre-built heterogeneous executors, and Raft control-plane images, enabling reviewers to observe and assess the representative architectural behaviors of the system under controlled configurations without exposing the internal source code. The complete implementation will be made available after acceptance in accordance with institutional IP requirements, without affecting the scope or validity of the current evaluation. Full article

To address the challenges of insufficient uncertainty characterization and inadequate flexible resource coordination in active distribution network (ADN) planning under high-penetration distributed renewable energy integration, this paper proposes a WGAN-GP-based coordinated source–network–storage expansion planning method for ADNs. First, an improved Wasserstein Generative Adversarial Network (WGAN-GP) model is employed to learn the statistical patterns of wind and photovoltaic (PV) power outputs, generating representative scenarios that accurately capture the uncertainty and correlation of renewable generation. Then, an ADN expansion planning model considering the E-SOP (Energy Storage-integrated Soft Open Point) is developed with the objective of minimizing the annual comprehensive cost, jointly optimizing the siting and sizing of substations, lines, distributed generators, and flexible resources. By integrating the energy storage system on the DC side of the SOP, E-SOP achieves coordinated spatial power flow regulation and temporal energy balancing, significantly enhancing system flexibility and renewable energy accommodation capability. Finally, a Successive Convex Cone Relaxation (SCCR) algorithm is adopted to solve the resulting non-convex optimization problem, enabling fast convergence to a high-precision feasible solution with few iterations. Simulation results on a 54-bus ADN demonstrate that the proposed method effectively reduces annual comprehensive costs and eliminates renewable curtailment while ensuring high renewable penetration, verifying the feasibility and superiority of the proposed model and algorithm. Full article

To address the prevalent noise issue in images generated by HumanNeRF, this paper proposes an image denoising method that combines self-supervised contrastive learning and Generative Adversarial Networks (GANs). While HumanNeRF excels in realistic 3D human reconstruction tasks, its generated images often suffer from noise and detail loss due to incomplete training data and sampling noise during the rendering process. To solve this problem, our method first utilizes a self-supervised contrastive learning strategy to construct positive and negative sample pairs, enabling the network to effectively distinguish between noise and human detail features without external labels. Secondly, it introduces a Generative Adversarial Network, where the adversarial training between the generator and discriminator further enhances the detail representation and overall realism of the images. Experimental results demonstrate that the proposed method can effectively remove noise from HumanNeRF images while significantly improving detail fidelity, ultimately yielding higher-quality human images and providing crucial support for subsequent 3D human reconstruction and realistic rendering. Full article

Reliable assessment of building damage is essential for effective disaster management. Synthetic Aperture Radar (SAR) has become a valuable tool for damage detection, as it operates independently of the daylight and weather conditions. However, the limited availability of high-resolution pre-disaster SAR data remains a major obstacle to accurate damage evaluation, constraining the applicability of traditional change-detection approaches. This study proposes a comprehensive framework that leverages generated SAR data alongside optical imagery for building damage detection and further examines the influence of elevation data quality on SAR synthesis and model performance. The method integrates SAR image synthesis from a Digital Surface Model (DSM) and land cover inputs with a multimodal deep learning architecture capable of jointly localizing buildings and classifying damage levels. Two data modality scenarios are evaluated: a change-detection setting using pre-disaster authentic SAR and another using GAN-generated SAR, both combined with post-disaster SAR imagery for building damage assessment. Experimental results demonstrate that GAN-generated SAR can effectively substitute for authentic SAR in multimodal damage mapping. Models using generated pre-disaster SAR achieved comparable or superior performance to those using authentic SAR, with $F_1$ scores of 0.730, 0.442, and 0.790 for the survived, moderate, and destroyed classes, respectively. Ablation studies further reveal that the model relies more heavily on land cover segmentation than on fine elevation details, suggesting that coarse-resolution DSMs (30 m) are sufficient as auxiliary input. Incorporating additional training regions further improved generalization and inter-class balance, confirming that high-quality generated SAR can serve as a viable alternative especially in the absence of authentic SAR, for scalable, post-disaster building damage assessment. Full article

Panoramic radiography remains a cornerstone diagnostic tool in dentistry; however, its two-dimensional nature limits the visualisation of complex maxillofacial anatomy. Three-dimensional reconstruction from single panoramic images addresses this limitation by computationally generating spatial representations without additional radiation exposure or expensive cone-beam computed tomography (CBCT) scans. This systematic review and conceptual study traces the evolution of 3D reconstruction approaches, from classical geometric and statistical shape models to modern artificial intelligence-based methods, including convolutional neural networks, generative adversarial networks, and neural implicit fields such as Occudent and NeBLa. Deep learning frameworks demonstrate superior accuracy in reconstructing dental and jaw structures compared to traditional techniques. Building on these advancements, this paper proposes HoloDent3D, a theoretical framework that combines AI-driven panoramic reconstruction with real-time holographic visualisation. The system enables interactive, radiation-free volumetric inspection for diagnosis, treatment planning, and patient education. Despite significant progress, persistent challenges include limited paired 2D–3D datasets, generalisation across anatomical variability, and clinical validation. Continued integration of multimodal data fusion, temporal modelling, and holographic visualisation is expected to accelerate the clinical translation of AI-based 3D reconstruction systems in digital dentistry. Full article

Streaming media data have become pervasive in modern commercial systems. To address large-scale data processing in intelligent transportation systems (ITSs), recent research has focused on deep neural network–based (DNN-based) approaches to improve the performance of cross-modal hashing retrieval (CMHR) systems. However, due to their high dimensionality and network depth, DNN-based CMHR systems inherently suffer from vulnerabilities to malicious adversarial examples (AEs). This paper investigates the robustness of CMHR-based ITS systems against AEs. Prior work typically formulates AE generation as an optimization-driven, iterative process, whose high computational cost and slow generation speed limit research efficiency. To overcome these limitations, we propose a parallel cross-modal relevance-guided generative network (ReGeNet) that captures the semantic characteristics of the target deep hashing model. During training, we design a relevance-guided adversarial generative framework to efficiently learn AE generation. During inference, the well-trained parallel adversarial generator produces adversarial cross-modal data with effectiveness comparable to that of iterative methods. Experimental results demonstrate that ReGeNet can generate AEs significantly faster while achieving competitive attack performance relative to iterative-based approaches. Full article

Accurate prediction of the stress–strain fields in fiber-reinforced composites is crucial for performance analysis and structural design. However, due to their complex microstructures, traditional finite element analysis (FEA) entails a very high computational cost. Therefore, this study proposes a conditional generative adversarial network (cGAN) framework, named FiberGAN, to enable rapid prediction of the microscopic stress–strain fields in fiber-reinforced composites. The method employs an adaptive representative volume element (RVE) generation algorithm to construct random fiber arrangements with fiber volume fractions ranging from 30% to 50% and uses FEA to obtain the corresponding stress and strain fields as training data. A U-Net generator, combined with a PatchGAN discriminator, captures both global distribution patterns and fine local details. Under tensile and shear loading conditions, the R² values of FiberGAN predictions range from 0.96 to 0.99, while the structural similarity index (SSIM) values range from 0.95 to 0.99. The error maps show that prediction residuals are mainly concentrated in high-gradient regions with small magnitudes. These results demonstrate that the proposed deep learning model can successfully predict stress–strain field distributions for different fiber volume fractions under various loading conditions. Full article

The increasing complexity of multi-scale energy systems makes robust anomaly detection essential to ensure system resilience and operational continuity. Recent advances in DL enable effective modeling of high-dimensional, non-linear energy data by capturing latent spatio-temporal patterns. In this paper, we proposed an optimized deep reinforcement learning–generative adversarial network (ODRL-GAN) framework for reliable anomaly detection in multi-scale energy systems. The integration of DRL and GAN brings a key innovation: while DRL enables adaptive decision-making under dynamic operating conditions, GAN enhances detection by reconstructing normal patterns and exposing subtle deviations. To further strengthen the model, a novel multi-objective chimp optimization algorithm (NMOChOA) is employed for hyper-parameter tuning, improving accuracy, and convergence. This design allows the ODRL–GAN to effectively capture high-dimensional spatio-temporal dependencies while maintaining robustness against diverse anomaly patterns. The framework is validated on two benchmark datasets, PSML and LEAD1.0, and compared against state-of-the-art baselines including transformer, deep belief network (DBN), convolutional neural network (CNN), gated recurrent unit (GRU), and support vector machines (SVM). Experimental results demonstrate that the proposed method achieves a maximum detection accuracy of 99.58% and recall of 99.75%, significantly surpassing all baselines. Furthermore, the model exhibits superior runtime efficiency, faster convergence, and lower variance across trials, highlighting both robustness and scalability. The optimized DRL–GAN framework provides a powerful and generalizable solution for anomaly detection in complex energy systems, offering a pathway toward secure and resilient next-generation energy infrastructures. Full article

Anomalous IP packet fragmentation, whether caused by evasion attacks, misconfigurations, or network policy interference, presents a measurable threat to network integrity and intrusion detection systems. This paper introduces a lightweight, rule-based framework for detecting and classifying fragmented IP traffic. Unlike complex machine learning models that operate as “black boxes,” our model leverages the deterministic semantics of RFC 791 to inspect structural packet characteristics—such as offset alignment, Time-to-Live (TTL) consistency, and payload regularity—and classifies traffic into three transparent categories: normal (NONE), misconfigured (MISCONFIG), and adversarial (ATTACK). We generate an open-source and synthetic dataset of 10,000 packets, meticulously engineered to simulate a wide spectrum of benign and malicious fragmentation scenarios. Evaluation demonstrates high accuracy (99.23% overall) on this controlled dataset. Crucially, validation on the CIC-IDS-2017 real-world dataset confirms the model’s practical utility, showing a low false-positive rate (0.8%) on normal traffic and a significant increase in detectable anomalies during attack periods. This work provides a reproducible, interpretable, and efficient tool for forensic analysis and intrusion detection, enabling the precise diagnostics of packet-level fragmentation anomalies in operational networks. Full article

Acoustic Side-Channel Attacks (ASCAs) exploit the sound produced by keyboards and other devices to infer sensitive information without breaching software or network defenses. Recent advances in deep learning, large language models, and signal processing have greatly expanded the feasibility and accuracy of these attacks. To clarify the evolving threat landscape, this survey systematically reviews ASCA research published between January 2020 and February 2025. We categorize modern ASCA methods into three levels of text reconstruction—individual keystrokes, short text (words/phrases), and long-text regeneration— and analyze the signal processing, machine learning, and language-model decoding techniques that enable them. We also evaluate how environmental factors such as microphone placement, ambient noise, and keyboard design influence attack performance, and we examine the challenges of generalizing laboratory-trained models to real-world settings. This survey makes three primary contributions: (1) it provides the first structured taxonomy of ASCAs based on text generation granularity and decoding methodology; (2) it synthesizes cross-study evidence on environmental and hardware factors that fundamentally shape ASCA performance; and (3) it consolidates emerging countermeasures, including Generative Adversarial Network-based noise masking, cryptographic defenses, and environmental mitigation, while identifying open research gaps and future threats posed by voice-enabled IoT and prospective quantum side-channels. Together, these insights underscore the need for interdisciplinary, multi-layered defenses against rapidly advancing ASCA techniques. Full article

In recent years, the convergence of advanced technologies has enabled real-time data access and sharing across diverse devices and networks, significantly amplifying cybersecurity risks. For organizations with digital infrastructures, network security is crucial for mitigating potential cyber-attacks. They establish security policies to protect systems and data, but employees may intentionally or unintentionally bypass these policies, rendering the network vulnerable to internal and external threats. Detecting these policy violations is challenging, requiring frequent manual system checks for compliance. This paper addresses key challenges in safeguarding digital assets against evolving threats, including rogue access points, man-in-the-middle attacks, denial-of-service (DoS) incidents, unpatched vulnerabilities, and AI-driven automated exploits. We propose a Blockchain-Enhanced Network Scanning and Monitoring (BENSAM) Framework, a multi-layered system that integrates advanced network scanning with a structured database for asset management, policy-driven vulnerability detection, and remediation planning. Key enhancements include device profiling, user activity monitoring, network forensics, intrusion detection capabilities, and multi-format report generation. By incorporating blockchain technology, and leveraging immutable ledgers and smart contracts, the framework ensures tamper-proof audit trails, decentralized verification of policy compliance, and automated real-time responses to violations such as alerts; actual device isolation is performed by external controllers like SDN or NAC systems. The research provides a detailed literature review on blockchain applications in domains like IoT, healthcare, and vehicular networks. A working prototype of the proposed BENSAM framework was developed that demonstrates end-to-end network scanning, device profiling, traffic monitoring, policy enforcement, and blockchain-based immutable logging. This implementation is publicly released and is available on GitHub. It analyzes common network vulnerabilities (e.g., open ports, remote access, and disabled firewalls), attacks (including spoofing, flooding, and DDoS), and outlines policy enforcement methods. Moreover, the framework anticipates emerging challenges from AI-driven attacks such as adversarial evasion, data poisoning, and transformer-based threats, positioning the system for the future integration of adaptive mechanisms to counter these advanced intrusions. This blockchain-enhanced approach streamlines security analysis, extends the framework for AI threat detection with improved accuracy, and reduces administrative overhead by integrating multiple security tools into a cohesive, trustworthy, reliable solution. Full article

Heterogeneous remote sensing image change detection (Hete-CD) holds significant research value in military and civilian fields. The existing methods often rely on expert experience to design fixed deep network architectures for cross-modal feature alignment and fusion purposes. However, when faced with diverse land cover types, these methods often lead to blurred change boundaries and structural distortions, resulting in significant performance degradations. To address this, we propose an adaptive adversarial learning-based heterogeneous remote sensing image change detection method based on the differentiable filter combination search (DFCS) strategy to provide enhanced generalizability and dynamic learning capabilities for diverse scenarios. First, a fully reconfigurable self-learning discriminator is designed to dynamically synthesize the optimal convolutional architecture from a library of atomic filters containing basic operators. This provides highly adaptive adversarial supervision to the generator, enabling joint dynamic learning between the generator and discriminator. To further mitigate modality differences in the input stage, we integrate a feature fusion module based on the Gabor and local normalized cross-correlation (G-LNCC) to extract modality-invariant texture and structure features. Finally, a geometric structure-based collaborative supervision (GSCS) loss function is constructed to impose fine-grained constraints on the change map from the perspectives of regions, boundaries, and structures, thereby enforcing physical properties. Comparative experimental results obtained on five public Hete-CD datasets show that our method achieves the best F1 values and overall accuracy levels, especially on the Gloucester I and Gloucester II datasets, achieving F1 scores of 93.7% and 95.0%, respectively, demonstrating the strong generalizability of our method in complex scenarios. Full article

Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR) systems face significant challenges due to limited labeled data availability and persistent domain gaps between synthetic and measured imagery. This paper presents a comprehensive self-supervised learning (SSL) framework that eliminates dependency on synthetic data while achieving state-of-the-art performance through multi-task pretext training and extensive downstream classifier evaluation. We systematically evaluate our SSL framework across diverse downstream classifiers spanning different computational paradigms and architectural families. Our study encompasses traditional machine learning approaches (SVM, Random Forest, XGBoost, Gradient Boosting), deep convolutional neural networks (ResNet, U-Net, MobileNet, EfficientNet), and a generative adversarial network. We conduct extensive experiments using the SAMPLE dataset with rigorous evaluation protocols. Results demonstrate that SSL significantly improves SAR ATR performance, with SVM achieving 99.63% accuracy, ResNet18 reaching 97.40% accuracy, and Random Forest demonstrating 99.26% accuracy. Our multi-task SSL framework employs nine carefully designed pretext tasks, including geometric invariance, signal robustness, and multi-scale analysis. Cross-validation experiments validate the generalizability and robustness of our findings. Rigorous comparison with SimCLR baseline validates that task-based SSL outperforms contrastive learning for SAR ATR. This work establishes a new paradigm for SAR ATR that leverages inherent radar data structure without synthetic augmentation, providing practical guidelines for deploying SSL-based SAR ATR systems and a foundation for future domain-specific self-supervised learning research in remote sensing applications. Full article

Synthetic aperture radar (SAR), with its all-weather and all-day observation capabilities, plays a significant role in the field of remote sensing. However, due to the unique imaging mechanism of SAR, its interpretation is challenging. Translating SAR images into optical remote sensing images has become a research hotspot in recent years to enhance the interpretability of SAR images. This paper proposes a deep learning-based method for SAR-to-optical remote sensing image translation. The network comprises three parts: a global representor, a generator with cascaded multi-head attention, and a multi-scale discriminator. The global representor, built upon InternImage with deformable convolution v3 (DCNv3) as its core operator, leverages its global receptive field and adaptive spatial aggregation capabilities to extract global semantic features from SAR images. The generator follows the classic “encoder-bottleneck-decoder” structure, where the encoder focuses on extracting local detail features from SAR images. The cascaded multi-head attention module within the bottleneck layer optimizes local detail features and facilitates feature interaction between global semantics and local details. The discriminator adopts a multi-scale structure based on the local receptive field PatchGAN, enabling joint global and local discrimination. Furthermore, for the first time in SAR image translation tasks, structural similarity index metric (SSIM) loss is combined with adversarial loss, perceptual loss, and feature matching loss as the loss function. A series of experiments demonstrate the effectiveness and reliability of the proposed method. Compared to mainstream image translation methods, our method ultimately generates higher-quality optical remote sensing images that are semantically consistent, texturally authentic, clearly detailed, and visually reasonable appearances. Full article