Search Results (3,449)

Rolling bearings, as essential parts of rotating machinery, face significant challenges in fault diagnosis due to limited fault samples and high noise interference, both of which reduce the effectiveness of traditional methods. To tackle this, this study proposes a fault diagnosis approach that combines Digital Twin (DT) and deep learning. First, actual bearing vibration data were collected using an experimental platform. After denoising the data, a high-fidelity digital twin system was built by integrating the bearing dynamics model with a Generative Adversarial Network (GAN), thereby effectively increasing the fault data. Next, the Wavelet Synchro-Extracting Transform (WSET) is used for high-resolution time-frequency analysis, and convolutional neural networks (CNNs) are employed to extract deep fault features adaptively. The fully connected layer of the CNN is then combined with a Least Squares Support Vector Machine (LSSVM), with key parameters optimized through an Improved Pelican Optimization Algorithm (IPOA) to improve classification accuracy significantly. Experimental results based on both simulated and publicly available datasets show that the proposed model has excellent generalizability and operational flexibility, surpassing existing deep learning-based diagnostic methods in complex industrial settings. Full article

Nanoparticles have found widespread application across diverse fields, including agriculture and animal husbandry. However, a persistent challenge in laboratory-based studies involving nanoparticle exposure is the limited availability of experimental data, which constrains the robustness and generalizability of findings. This study presents a comprehensive analysis of the impact of zinc oxide nanoparticles (ZnO NPs) in feed on elemental homeostasis in male Wistar rats. Using correlation-based network analysis, a correlation graph weight value of 15.44 and a newly proposed weighted importance score of 1.319 were calculated, indicating that a dose of 3.1 mg/kg represents an optimal balance between efficacy and physiological stability. To address the issue of limited sample size, synthetic data generation was performed using generative adversarial networks, enabling data augmentation while preserving the statistical characteristics of the original dataset. Machine learning models based on fully connected neural networks and kernel ridge regression, enhanced with a custom loss function, were developed and evaluated. These models demonstrated strong predictive performance across a ZnO NP concentration range of 1–150 mg/kg, accurately capturing the dependencies of essential element, protein, and enzyme levels in blood on nanoparticle dosage. Notably, the presence of toxic elements and some other elements at ultra-low concentrations exhibited non-random patterns, suggesting potential systemic responses or early indicators of nanoparticle-induced perturbations and probable inability of synthetic data to capture the true dynamics. The integration of machine learning with synthetic data expansion provides a promising approach for analyzing complex biological responses in data-scarce experimental settings, contributing to the safer and more effective application of nanoparticles in animal nutrition. Full article

The proliferation of fake profiles on social media presents critical cybersecurity and misinformation challenges, necessitating robust and scalable detection mechanisms. Such profiles weaken consumer trust, reduce user engagement, and ultimately harm brand reputation and platform credibility. As adversarial tactics and synthetic identity generation evolve, traditional rule-based and machine learning approaches struggle to detect evolving and deceptive behavioral patterns embedded in dynamic user-generated content. This study aims to develop an AI-driven, multi-modal deep learning-based detection system for identifying fake profiles that fuses textual, visual, and social network features to enhance detection accuracy. It also seeks to ensure scalability, adversarial robustness, and real-time threat detection capabilities suitable for practical deployment in industrial cybersecurity environments. To achieve these objectives, the current study proposes an integrated AI system that combines the Robustly Optimized BERT Pretraining Approach (RoBERTa) for deep semantic textual analysis, ConvNeXt for high-resolution profile image verification, and Heterogeneous Graph Attention Networks (Hetero-GAT) for modeling complex social interactions. The extracted features from all three modalities are fused through an attention-based late fusion strategy, enhancing interpretability, robustness, and cross-modal learning. Experimental evaluations on large-scale social media datasets demonstrate that the proposed RoBERTa-ConvNeXt-HeteroGAT model significantly outperforms baseline models, including Support Vector Machine (SVM), Random Forest, and Long Short-Term Memory (LSTM). Performance achieves 98.9% accuracy, 98.4% precision, and a 98.6% F1-score, with a per-profile speed of 15.7 milliseconds, enabling real-time applicability. Moreover, the model proves to be resilient against various types of attacks on text, images, and network activity. This study advances the application of AI in cybersecurity by introducing a highly interpretable, multi-modal detection system that strengthens digital trust, supports identity verification, and enhances the security of social media platforms. This alignment of technical robustness with brand trust highlights the system’s value not only in cybersecurity but also in sustaining platform credibility and consumer confidence. This system provides practical value to a wide range of stakeholders, including platform providers, AI researchers, cybersecurity professionals, and public sector regulators, by enabling real-time detection, improving operational efficiency, and safeguarding online ecosystems. Full article

Generative adversarial networks (GANs) are increasingly used for image enhancement tasks like denoising, super-resolution, and artifact removal. In this study, we propose a robust GAN-based architecture featuring a U-Net-style generator and DCNN-based discriminator, optimized for real-time enhancement. Key contributions include multi-task image refinement using residual and attention modules. Our method is tested on paired datasets comprising low-light medical and artificially degraded images. Results show significant improvements in visual clarity, reduced noise, and improved resolution compared to traditional methods. Evaluation metrics such as inception score (IS) and Fréchet inception distance (FID) confirm the system’s performance. The optimal learning rate (0.0008) was empirically selected for training stability. These results validate the proposed GAN’s efficiency for practical deployment in domains requiring high-quality imaging. Full article

The harmonized multi-sensor satellite data assists users by providing seamless analysis-ready data with enhanced temporal resolution. The Harmonized Landsat Sentinel (HLS) product has gained popularity due to the seamless integration of Landsat OLI and Sentinel-2 MSI, achieving a temporal resolution of 2.8 to 3.5 days. However, applications that require monitoring intervals of less than three days or cloudy data can limit the usage of HLS data. Gaofen-1 (GF-1) Wide Field of View (WFV) data provides the capacity further to enhance the data availability by harmonization with HLS. In this study, GF-1/WFV data is harmonized with HLS by employing deep learning-based conditional Generative Adversarial Networks (cGANs). The harmonized WFV data with HLS provides an average temporal resolution of 1.5 days (ranging from 1.2 to 1.7 days), whereas the temporal resolution of HLS varies from 2.8 to 3.5 days. This enhanced temporal resolution will benefit applications that require frequent monitoring. Various processes are employed in HLS to achieve seamless products from the Operational Land Imager (OLI) and Multispectral Imager (MSI). This study applies 6S atmospheric correction to obtain GF-1/WFV surface reflectance data, employs MFC cloud masking, resamples the data to 30 m, and performs geographical correction using AROP relative to HLS data, to align preprocessing with HLS workflows. Harmonization is achieved without using BRDF normalization and bandpass adjustment like in the HLS workflows; instead, cGAN learns cross-sensor reflectance mapping by utilizing a U-Net generator and a patchGAN discriminator. The harmonized GF-1/WFV data were compared to the reference HLS data using various quality indices, including SSIM, MBE, and RMSD, across 126 cloud-free validation tiles covering various land covers and seasons. Band-wise scatter plots, histograms, and visual image color quality were compared. All these indices, including the Sobel filter, histograms, and visual comparisons, indicated that the proposed method has effectively reduced the spectral discrepancies between the GF-1/WFV and HLS data. Full article

Despite the remarkable success of Deep Neural Networks (DNNs) in Remote Sensing Image (RSI) object detection, they remain vulnerable to adversarial attacks. Numerous adversarial attack methods have been proposed for RSI; however, adding a single large-scale adversarial patch to certain high-value targets, which are typically large in physical scale and irregular in shape, is both costly and inflexible. To address this issue, we propose a strategy of using multiple compact patches. This approach introduces two fundamental challenges: (1) how to optimize patch placement for a synergistic attack effect, and (2) how to retain strong adversarial potency within size-constrained mini-patches. To overcome these challenges, we introduce the Spatially Adaptive and Distillation-Enhanced Mini-Patch Attack (SDMPA) framework, which consists of two key modules: (1) an Adaptive Sensitivity-Aware Positioning (ASAP) module, which resolves the placement challenge by fusing the model’s attention maps from both an explainable and an adversarial perspective to identify optimal patch locations, and (2) a Distillation-based Mini-Patch Generation (DMPG) module, which tackles the potency challenge by leveraging knowledge distillation to transfer adversarial information from large teacher patches to small student patches. Extensive experiments on the RSOD and MAR20 datasets demonstrate that SDMPA significantly outperforms existing patch-based attack methods. For example, against YOLOv5n on the RSOD dataset, SDMPA achieves an Attack Success Rate (ASR) of 88.3% using only three small patches, surpassing other patch attack methods. Full article

This paper proposes a Dynamic Flow Spatio-Temporal Generative Adversarial Network (DFST-GAN) model for high-quality precipitation nowcasting. Current spatio-temporal prediction models struggle with two key limitations: the inability to adaptively capture complex motion patterns and the tendency to generate blurry predictions over time. To address these challenges, DFST-GAN integrates a dynamic flow feature extraction mechanism with a novel specialized meteorological discriminator, enabling adaptive modeling of complex precipitation system trajectories and generating sharper, physically consistent predictions. We evaluate our approach on the HKO-7 dataset using metrics including CSI, HSS, POD, FAR and ETS. Experimental results demonstrate that DFST-GAN consistently outperforms existing methods across all evaluation metrics, with particularly notable improvements for moderate to heavy rainfall events (dBZ ≥ 50), showing a 18.8% relative improvement in CSI compared to PredRNN-V2. The ablation studies confirm that each component makes a meaningful contribution to overall performance, validating the potential of our approach for operational precipitation nowcasting applications. Full article

As a critical technology of autonomous mining trucks, object detection directly determines system safety and operational reliability. However, autonomous mining trucks often work in dusty open-pit environments, in which dusty interference significantly degrades the accuracy of object detection. To overcome the problem mentioned above, a multi-branch feature interaction and location detection network (MBFILNet) is proposed in this study, consisting of multi-branch feature interaction with differential operation (MBFI-DO) and depthwise separable convolution-enhanced non-local attention (DSC-NLA). On one hand, MBFI-DO not only strengthens the extraction of channel-wise semantic features but also improves the representation of salient features of images with dusty interference. On the other hand, DSC-NLA is used to capture long-range spatial dependencies to focus on target-object structural information. Furthermore, a custom dataset called Dusty Open-pit Mining (DOM) is constructed, which is augmented using a cycle-consistent generative adversarial network (CycleGAN). Finally, a large number of experiments based on DOM are conducted to evaluate the performance of MBFILNet in dusty open-pit environments. The results show that MBFILNet achieves a mean Average Precision (mAP) of 72.0% based on the DOM dataset, representing a 1.3% increase compared to the Featenhancer model. Moreover, in comparison with YOLOv8, there is an astounding 2% increase in the mAP based on MBFILNet, demonstrating detection accuracy in dusty open-pit environments can be effectively improved with the method proposed in this paper. Full article

Due to the existence of domain shift, the synthetic aperture radar (SAR) automatic target recognition (ATR) model trained on simulated data will have significant performance degradation when applied to real-world measured data. To bridge the domain gap, this paper proposes a single-domain generalization (SDG) method based on multilevel data augmentation (MLDA), enabling SAR-ATR models that have been fully trained on simulated data to be generalized to unseen real SAR data. The proposed method aims to enhance the model’s generalizable capability through three key components: (1) pixel-level augmentation, which enriches data distribution via random Gaussian noise injection in the spatial domain and high-frequency perturbation in the frequency domain to enhance pixel-level diversity; (2) feature-level style augmentation, which probabilistically mixes instance-wise feature statistics, generating hybrid-styled feature maps to enhance feature-level diversity; (3) domain-adversarial training, which constructs an adversarial framework between a feature extractor and discriminator to enforce the learning of domain-invariant representations. Experiments on two simulation-to-reality SAR datasets demonstrate that the proposed method outperforms existing baselines and other SDG algorithms, achieving state-of-the-art (SOTA) performance on both datasets (96.76% accuracy on the public SAMPLE dataset and 93.70% accuracy on the self-built S2M-5 dataset). Full article

To overcome the limitations of low texture accuracy in traditional sculpture color restoration methods, this study proposes an improved Deep Convolutional Generative Adversarial Network (DCGAN) model incorporating a dual attention mechanism (spatial and channel attention) and a channel converter to enhance restoration quality. First, the theoretical foundations of the DCGAN algorithm and its key components (generator, discriminator, etc.) are systematically introduced. Subsequently, a DCGAN-based application model for sculpture color restoration is developed. The generator employs a U-Net architecture integrated with a dual attention module and a channel converter, enhancing both local feature representation and global information capture. Meanwhile, the discriminator utilizes an image region segmentation approach to optimize the assessment of consistency between restored and original regions. The loss function follows a joint optimization strategy, combining perceptual loss, adversarial loss, and structural similarity index (SSIM) loss, ensuring superior restoration performance. In the experiments, mean square error (MSE), peak signal-to-noise ratio (PSNR), and SSIM were used as evaluation metrics, and sculpture color restoration tests were conducted on an Intel Xeon workstation. The performance of the proposed model was compared against the traditional DCGAN and other restoration models. The experimental results demonstrate that the improved DCGAN outperforms traditional methods across all evaluation metrics, and compared to traditional DCGAN, the proposed model achieves significantly higher SSIM and PSNR, while reducing MSE. Compared to other restoration models, PSNR and SSIM are further enhanced, MSE is reduced, and the visual consistency between the restored and undamaged areas is significantly improved, with richer texture details. Full article

Autonomous ship control faces significant challenges due to the diversity of ship types, the complexity of task scenarios, and the uncertainty of dynamic marine environments. These factors limit the effectiveness of traditional control approaches that rely on explicit dynamics modeling and handcrafted control laws. With the rapid advancement of computing and artificial intelligence, imitation learning offers a promising alternative by directly learning expert behaviors from data. This paper proposes a Generative Adversarial Imitation Learning (GAIL) method for heading and path-following control of autonomous surface ships. It employs an adversarial learning structure, in which a generator learns control policies that reproduce expert behavior while a discriminator distinguishes between expert and learned trajectories. In this way, the control strategies can be learned from expert demonstrations without requiring explicit reward design. The proposed method is validated through simulations on a model-scale tug. Compared with a behavioral cloning (BC) baseline controller, the GAIL-based controller achieves superior performance in terms of path-following accuracy, heading stability, and control smoothness, confirming its effectiveness and potential for real-world deployment. Full article

We address the task of point cloud denoising by leveraging a diffusion-based generative framework augmented with adversarial training. While recent diffusion models have demonstrated strong capabilities in learning complex data distributions, their effectiveness in recovering fine geometric details remains limited, especially under severe noise conditions. To mitigate this, we propose the Adversarial Diffusion Bridge Model (ADBM), a novel approach for denoising 3D point cloud data by integrating a diffusion bridge model with adversarial learning. ADBM incorporates a lightweight discriminator that guides the denoising process through adversarial supervision, encouraging sharper and more faithful reconstructions. The denoiser is trained using a denoising diffusion objective based on a Schrödinger Bridge, while the discriminator distinguishes between real, clean point clouds and generated outputs, promoting perceptual realism. Experiments are conducted on the PU-Net and PC-Net datasets, with performance evaluation employing the Chamfer distance and Point-to-Mesh metrics. The qualitative and quantitative results both highlight the effectiveness of adversarial supervision in enhancing local detail reconstruction, making our approach a promising direction for robust point cloud restoration. Full article

Industrial Wireless Sensor Networks (IWSNs) play a critical role in Industry 4.0 environments, enabling real-time monitoring and control of industrial processes. However, existing lightweight authentication protocols for IWSNs remain vulnerable to sophisticated security attacks because of inadequate initial authentication phases. This study presents a security analysis of Gope et al.’s PUF-based authentication protocol for IWSNs and identifies critical vulnerabilities that enable man-in-the-middle (MITM) and denial-of-service (DoS) attacks. We demonstrate that Gope et al.’s protocol is susceptible to MITM attacks during both authentication and Secure Periodical Data Collection (SPDC), allowing adversaries to derive session keys and compromise communication confidentiality. Our analysis reveals that the sensor registration phase of the protocol lacks proper authentication mechanisms, enabling attackers to perform unauthorized PUF queries and subsequently mount successful attacks. To address these vulnerabilities, we propose an enhanced authentication scheme that introduces a sensor-initiated registration process. In our improved protocol, sensor nodes generate and control PUF challenges rather than passively responding to gateway requests. This modification prevents unauthorized PUF queries while preserving the lightweight characteristics essential for resource-constrained IWSN deployments. Security analysis demonstrates that our enhanced scheme effectively mitigates the identified MITM and DoS attacks without introducing significant computational or communication overhead. The proposed modifications maintain compatibility with the existing IWSN infrastructure while strengthening the overall security posture. Comparative analysis shows that our solution addresses the security weaknesses of the original protocol while preserving its practical advantages for industrial use. The enhanced protocol provides a practical and secure solution for real-time data access in IWSNs, making it suitable for deployment in mission-critical industrial environments where both security and efficiency are paramount. Full article

Super-resolution (SR) is a long-standing task in the field of computer vision that aims to improve the quality and resolution of an image. ESRGAN is a representative generative adversarial network specialized to produce perceptually convincing SR images. However, it often fails to recover local details and still produces blurry or unnatural visual artifacts, resulting in producing SR images that people do not prefer. To address this problem, we propose to adopt Direct Preference Optimization (DPO), which was originally devised to fine-tune large language models based on human preferences. To this end, we develop a method for applying DPO to ESRGAN, and add a DPO loss for training the ESRGAN generator. Through ×4 SR experiments utilizing benchmark datasets, it is demonstrated that the proposed method can produce SR images with a significantly higher perceptual quality and higher human preference than ESRGAN and other ESRGAN variants that have modified the loss or network structure of ESRGAN. Specifically, when compared to ESRGAN, the proposed method achieved, on average, 0.32 lower PieAPP values, 0.79 lower NIQE values, and 0.05 higher PSNR values on the BSD100 dataset, as well as 0.32 lower PieAPP values, 0.32 lower NIQE values, and 0.17 higher PSNR values on the Set14 dataset. Full article

The rapid advancement of Artificial Intelligence of Things (AIoT) has driven an urgent demand for intelligent video anomaly detection (VAD) to ensure industrial safety. However, traditional approaches struggle to detect unknown anomalies in complex and dynamic environments due to the scarcity of abnormal samples and limited generalization capabilities. To address these challenges, this paper presents an adaptive VAD framework powered by edge intelligence tailored for resource-constrained industrial settings. Specifically, a lightweight feature extractor is developed by integrating residual networks with channel attention mechanisms, achieving a 58% reduction in model parameters through dense connectivity and output pruning. A multidimensional evaluation strategy is introduced to dynamically select optimal models for deployment on heterogeneous edge devices. To enhance cross-scene adaptability, we propose a multilayer adversarial domain adaptation mechanism that effectively aligns feature distributions across diverse industrial environments. Extensive experiments on a real-world coal mine surveillance dataset demonstrate that the proposed framework achieves an accuracy of 86.7% with an inference latency of 23 ms per frame on edge hardware, improving both detection efficiency and transferability. Full article