Search Results (3,552)

High-fidelity imaging in extreme low light is challenged by ultra-low signal-to-noise ratios. We propose a hybrid noise modeling framework integrating physical priors with generative adversarial networks (GANs). The method simulates photon shot noise via Poisson distribution and incorporates readout, row, and quantization noise. A multi-layer perceptron (MLP) dynamically maps ISO levels to noise intensities in logarithmic space, followed by a residual U-Net for non-linear refinement. Results on the SID datasets show that our method outperforms state-of-the-art approaches in terms of Average Kullback–Leibler Divergence (AKLD). Denoising networks trained on our synthetic noise achieve performances comparable to those trained on real-world paired datasets. Full article

Sound plays a fundamental role in human perception, conveying information about events, objects, and spatial dynamics that may not be visually accessible. However, current technologies such as Acoustic Event Detection typically reduce complex soundscapes to textual labels, often failing to preserve their semantic richness. This limitation motivates the exploration of sound-to-image (S2I) translation as an alternative connection between audio and visual modalities. Unlike multimodal approaches guided by intermediary constraints during the learning process, we investigate S2I translation without class supervision, cluster-based alignment, or textual mediation, a paradigm we refer to as direct S2I translation. To the best of our knowledge, apart from our previous work, no prior study addresses S2I translation under this fully direct setting. We propose a convolutional–attention generative framework composed of an audio encoder and a densely connected GAN integrating self-attention and cross-attention mechanisms. The attention-based model is systematically compared with a purely convolutional baseline. Results show that introducing attention at early stages of the generator significantly improves translation performance, increasing the likelihood of producing interpretable and semantically coherent visual representations of sound. These findings indicate that attention strengthens semantic correspondence between audio and vision while preserving the fully direct nature of the translation process. Full article

Inertial localization is particularly valuable in GPS-denied environments such as indoors. However, localization using only Inertial Measurement Units (IMUs) suffers from drift caused by motion-process noise and sensor biases. This paper introduces Uncertainty-aware Map-constrained Inertial Localization (UMLoc), an end-to-end framework that jointly models IMU uncertainty and map constraints to achieve drift-resilient positioning. UMLoc integrates two coupled modules: (1) a Long Short-Term Memory (LSTM) quantile regressor, which estimates the specific quantiles needed to define $68\%$, $90\%$ and $95\%$ prediction intervals serving as a measure of localization uncertainty and (2) a Conditioned Generative Adversarial Network (CGAN) with cross-attention that fuses IMU dynamic data with distance-based floor-plan maps to generate geometrically feasible trajectories. The modules are trained jointly, allowing uncertainty estimates to propagate through the CGAN during trajectory generation. UMLoc was evaluated on three datasets, including a newly collected 2-h indoor benchmark with time-aligned IMU data, ground-truth poses and floor-plan maps. Results show that the method achieves a mean drift ratio of $5.9\%$ over a $70\mathrm{m}$ travel distance and an average Absolute Trajectory Error (ATE) of $1.36\mathrm{m}$, while maintaining calibrated prediction bounds. Full article

Accurate detection of cardiac arrhythmias from electrocardiograms remains challenging for rare rhythm classes due to class imbalance and morphological variability. We present a hybrid deep learning framework combining per-lead convolutional encoders with a cross-lead transformer that models relationships across different lead signals through self-attention, accepting variable lead configurations. To address minority-class scarcity, a generative adversarial network synthesizes physiologically plausible beat segments for underrepresented arrhythmias. Attention-based visualizations localize influential waveform regions aligned with clinically meaningful structures. Post-training pruning and INT8 quantization enable efficient deployment with minimal performance loss. Extensive experiments on the MIT-BIH Arrhythmia Database across sixteen heartbeat classes from two-lead recordings yield exceptional results over ten independent runs: accuracy of 99.67%, F1-score of 99.66%, and AUC of 99.8%. External validation on the ECG5000 single-lead dataset and the St Petersburg INCART twelve-lead dataset confirms robust generalizability with F1-scores of 97.6% and 98% respectively. Our framework delivers accurate, interpretable, stable, and deployable arrhythmia detection across diverse clinical settings. Full article

In recent years, video recognition models have witnessed the rapid development of Deep Neural Networks (DNNs). However, these models remain not robust to adversarial examples that are created by adding imperceptible perturbations to clean samples. Recent studies indicate that generating adversarial examples in the hard-label black-box setting is particularly challenging yet highly practical. Compared to image recognition models, there are few hard-label black-box adversarial example generation algorithms for video recognition models. To this end, we propose a hard-label black-box video adversarial example generation algorithm, referred to as Dynamic Black-box Algorithm (DBA). First, DBA uses the binary search algorithm to find the boundary video between two original videos; then, the sampling-based algorithm is used to estimate the gradient on the boundary video; finally, with a dynamic step size adjustment strategy, DBA moves the boundary video towards the direction of the estimated gradient to generate the adversarial video. Additionally, we designed another strategy to skip invalid samples generated during the adversarial example generation process. Experiments demonstrate that DBA attains a superior trade-off between the magnitude of perturbations and query efficiency. Specifically, DBA outperforms state-of-the-art algorithms, achieving an average reduction in Mean Squared Error (MSE) of over 50%. Full article

Network Intrusion Detection Systems (NIDS) play a crucial role in protecting network environments against cyberattacks. However, traditional NIDS rely heavily on predefined attack signatures, which limits their ability to detect zero-day attacks. Although machine learning-based intrusion detection techniques have been widely adopted in Network Intrusion Prevention Systems (NIPS), publicly available network traffic datasets often suffer from severe class imbalance, leading to biased learning and degraded detection performance. To address this issue, this study proposes data augmentation framework based on a 3D-GAN (Three-Discriminator Generative Adversarial Network). The proposed architecture integrates an autoencoder, a CNN (Convolutional Neural Network), and an LSTM (Long Short-Term Memory) network as parallel discriminators to capture the statistical, spatial, and temporal characteristics of network traffic. By jointly optimizing multiple discriminator losses, the framework enhances training stability and generates high-quality synthetic samples. Experiments were conducted on the CIC-UNSW-NB15 dataset using Random Forest-, XGBoost (eXtreme Gradient Boosting)-, and BiGRU (Bidirectional Gated Recurrent Unit)-based classifiers. Two augmented datasets were constructed to address class imbalance, containing approximately 100,000 and 350,000 samples, respectively. Among them, Dataset 2, augmented using the proposed 3D-GAN, demonstrated the most significant performance improvement. Compared to the original imbalanced dataset, the XGBoost classifier trained on Dataset 2 achieved approximately a 4% increase in both accuracy and F1-score, while reducing the false positive rate and false negative rate by approximately 3.5%. Furthermore, the optimal configuration attained an F1-score of 0.9816, indicating superior capability in modeling complex network traffic patterns. Overall, this study highlights the potential of GAN-based data augmentation for alleviating class imbalance and improving the robustness and generalization of intrusion detection systems. Full article

Existing intelligent monitoring methods are limited by insufficient training samples and target-feature degradation in complex environments. To address these issues, an industrial visual inspection scheme with dual verification is proposed for material sheds. The scheme integrates sample enhancement preprocessing based on a Dynamic Enhanced Generative Adversarial Network (DEGAN) with an Attention-Enhanced YOLO-SLOWFAST (AE-YOLO-SLOWFAST) model for target and behavior detection, enabling feature enhancement, real-time dust monitoring, and timely dust suppression. A dynamic enhancement module is first introduced into a GAN, creating DEGAN to generate high-quality samples and augment the training dataset. An AE-YOLO model is then developed to improve static feature extraction under low illumination and enhance small-target detection. The objective function is refined to improve recognition of hard-to-distinguish samples during training. AE-YOLO is combined with SLOWFAST to recognize vehicle behaviors. Dual verification is performed using dust and vehicle detection results together with action recognition outputs, enabling precise control of dust suppression equipment for targeted water mist spraying. The improved AE-YOLO model achieves an mAP@50 of 94.4%. The proposed method delivers a vehicle–dust association matching accuracy of up to 97.2%, which enables all-weather, intelligent, traceable dust suppression in material sheds, reduces false recognition interference, and ensures timely suppression in areas where vehicles are operating. Full article

Network intrusion detection systems (NIDS) are critical for maintaining the security and integrity of modern networks. Traditional IDS techniques, while effective, often struggle with the evolving nature of cyber threats and the need for real-time detection. This paper proposes WASAE-NIDS, a deep learning-based NIDS that leverages a generative adversarial network (GAN)-assisted conditional autoencoder combined with reverse-frequency class weighting to enhance detection, particularly under severe class imbalance. In evaluating NIDS benchmark datasets, our method demonstrates superior performance in detecting various types of cyber threats with high accuracy and improved performance on minority classes. The results demonstrate the potential of combining GAN-assisted representation learning and class weighting to improve NIDS robustness and effectiveness. Full article

With the rapid advancement of deep learning, data acquisition remains a persistent challenge, as model effectiveness heavily relies on the quality and quantity of training data. To address the difficulties of time-consuming and labor-intensive data collection, data augmentation techniques are commonly adopted. In this study, the proposed WACGAN-GP, a Generative Adversarial Network (GAN) architecture, serves as an effective data augmentation tool designed to augment training datasets and bolster model performance. This method integrates the advantages of the Auxiliary Classifier GAN and the Wasserstein GAN with gradient penalty to generate diverse and realistic samples. Experiments were conducted on three image datasets—MNIST, CIFAR-10, and a ship classification dataset—under limited training data conditions. By incorporating WACGAN-GP generated synthetic samples into the original training sets, classification performance was evaluated in both balanced and imbalanced scenarios. The results demonstrate that the proposed GAN-based approach significantly improves recognition accuracy and outperforms conventional augmentation methods, such as horizontal and vertical flipping. Full article

This work investigates the extension of StyleGAN2-ADA to three-dimensional prostate T2-weighted (T2W) MRI generation. The architecture is adapted to operate on 3D anisotropic volumes, enabling stable training at a clinically relevant resolution of $256\times 256\times 24$, where a baseline 3D-StyleGAN fails to converge. Quantitative evaluation using Fréchet Inception Distance (FID), Kernel Inception Distance (KID), and generative Precision–Recall metrics demonstrates substantial improvements over a 3D-StyleGAN baseline. Specifically, FID decreased from 114.2 to 27.3, while generative Precision increased from 0.22 to 0.82, indicating markedly improved fidelity and alignment with the real data distribution. Beyond generative metrics, the synthetic volumes were evaluated through radiomic feature analysis and downstream prostate segmentation. Synthetic data augmentation resulted in segmentation performance comparable to real-data training, supporting that volumetric generation preserves anatomically relevant structures, while multivariate radiomic analyses showed strong global feature alignment between real and synthetic volumes. These findings indicate that a 3D extension of StyleGAN2-ADA enables stable high-resolution volumetric prostate MRI synthesis while preserving anatomically coherent structure and global radiomic characteristics. Full article

Background/Objectives: Contrast-enhanced CT (CECT) is essential for mediastinal and lymph node assessment but is often limited in patients with renal dysfunction, prior severe contrast reactions, or pediatric populations. Deep learning approaches, such as generative adversarial networks (GANs), allow the generation of synthetic CECT (sCECT) from non-contrast CT (NCCT) without contrast injection. Materials and Methods: A GAN-based model was trained using 400 CECT scans acquired between March and July 2024. The model was tested in 20 patients with lymphoma or metastatic lymphadenopathy diagnosed between January and July 2025, using only NCCT scans. Quantitative evaluation compared sCECT with CECT using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), and Pearson Correlation Coefficient (PCC). Two radiologists performed qualitative assessment, and Signal-to-Noise Ratio (SNR)/Contrast-to-Noise Ratio (CNR) values were measured for thoracic structures. Results: Compared with NCCT, sCECT demonstrated slightly lower MAE (20.87 ± 8.84 vs. 21.26 ± 9.26) and RMSE (45.22 ± 14.22 vs. 45.94 ± 15.07), and marginally higher PSNR (15.44 ± 2.70 vs. 15.38 ± 3.02), indicating modest improvements in pixel-wise similarity. SSIM values were comparable (0.610 ± 0.09 vs. 0.63 ± 0.10), while PCC decreased (0.61 ± 0.09 vs. 0.77 ± 0.15). All differences were statistically significant (p < 0.001). Despite these mixed quantitative results, sCECT was qualitatively rated significantly higher by radiologists, with improved visualization of mediastinal structures. SNR and CNR analyses further supported enhanced contrast depiction in sCECT compared with NCCT. Conclusions: The GAN-based model successfully generated sCECT from NCCT with modest quantitative similarity gains but clear qualitative improvement, particularly for mediastinal lymph node evaluation. Although synthetic enhancement represents a learned intensity transformation rather than true iodine-based attenuation, sCECT may serve as a valuable adjunct in patients with contraindications to iodinated contrast. Full article

Industrial-Internet security faces a core challenge: improving detection accuracy for critical minority-class network attacks. The existing intrusion detection methods based on Conditional Generative Adversarial Nets (CGANs) aim to achieve data balance by reconstructing minority-class attack samples. However, they encounter problems such as generating deceptive samples, poor sample quality, vanishing gradients and difficulties in training. This paper proposes an intrusion detection method based on the Multi-Discriminator Conditional Classification Generative Adversarial Network (MDCCGAN), an improved variant of CGAN, which integrates multiple discriminators and an independent classifier into the traditional CGAN framework. The multiple discriminators reduce the probability of generating deceptive samples, the independent classifier decouples the classification loss to clarify the direction of gradient updates, and the introduction of the Wasserstein distance fundamentally addresses the gradient-vanishing problem. Experiments conducted on the NSL-KDD and UNSW-NB15 datasets demonstrate that the proposed method significantly improves the recall, F1-score and accuracy for minority-class attacks. Specifically, on the NSL-KDD dataset, the overall accuracy increases from 74% to 94%, and the F1-score for the extremely rare U2R attack surges from 0% to 77%. Similarly, on the UNSW-NB15 dataset, the accuracy reaches 88%, a 10% improvement over the baseline DNN, and the F1-scores for extreme minority attacks such as Analysis, Backdoor, and Worms improved to 97%, 62%, and 84%, respectively. These results confirm that our method effectively outperforms traditional generation models and common class-balancing methods. It provides reliable technical support for industrial-Internet security. Full article

Connected autonomous vehicles (CAVs) rely on deep neural network-based perception systems to operate safely in complex driving environments. However, these systems remain vulnerable to adversarial perturbations that can induce misclassification without perceptible changes to human observers. Explainable artificial intelligence (XAI) has been proposed as a potential adversarial detection mechanism by exposing inconsistencies in model attention. This study evaluated the effectiveness of NoiseCAM-based explanation-space detection on the German Traffic Sign Recognition Benchmark (GTSRB) using a single 32 × 32 CNN architecture. Adversarial examples were generated using FGSM under perturbation budgets $ϵ$ = 0.01–0.10, and detection performance was evaluated using accuracy, precision, recall, F1-score, and ROC–AUC. Results show that NoiseCAM achieves detection accuracies between 51.8% and 52.9% with ROC–AUC values of 0.52–0.53, only marginally above random discrimination (0.5). Class-wise analysis further reveals substantial variability in detection reliability across traffic sign categories, with visually structured regulatory signs exhibiting higher separability than complex warning signs. These findings suggest that explanation-space inconsistencies alone provide limited adversarial detection capability in low-resolution, safety-critical perception pipelines. The study contributes to the understanding of the operational limits of explanation-based adversarial detection and highlights the need to integrate XAI signals with complementary robustness or uncertainty-aware mechanisms for reliable deployment in autonomous driving systems. Full article

Background/Objectives: Biopsy examination remains the gold standard for cancer diagnosis, relying on histopathological assessment of tissue samples to identify malignant changes. However, manual interpretation of histopathological slides is time-consuming, subjective, and susceptible to inter-observer variability. The digitization of histopathological images enables automated analysis and offers opportunities to support clinicians with more consistent and objective diagnostic tools. This study aims to enhance cancer diagnosis by proposing a hybrid framework that integrates deep-learning-based histopathological image analysis with Whispering Gallery Mode (WGM) optical sensing for complementary tissue characterization. Methods: The proposed framework combines automated tumor classification from histopathological images with biochemical signal analysis obtained from WGM optical sensors. Deep learning models, including EfficientNet-B0, InceptionV3, and Vision Transformer (ViT), were employed for binary and multi-class tumor classification using the BreakHis dataset. To address class imbalance, a Deep Convolutional Generative Adversarial Network (DCGAN) was utilized to generate synthetic histopathological images alongside conventional data augmentation techniques. In parallel, WGM optical sensors were incorporated to capture subtle tissue-specific signatures, with machine learning algorithms enabling automated feature extraction and classification of the acquired signals. Results: In multi-class classification, InceptionV3 combined with DCGAN-based augmentation achieved an accuracy of 94.45%, while binary classification reached 96.49%. Fine-tuned Vision Transformer models achieved a higher classification accuracy of 98% on the BreakHis dataset. The integration of WGM optical sensing provided additional biochemical information, offering complementary insights to image-based analysis and supporting more robust diagnostic decision-making. Conclusions: The proposed hybrid framework demonstrates the potential of combining deep-learning-based histopathological image analysis with WGM optical sensing to improve the accuracy and reliability of cancer classification. By integrating morphological and biochemical information, the framework offers a promising approach for enhanced, objective, and supportive cancer diagnostic systems. Full article

Given the growing concern over the operational safety and long-term reliability of lithium-ion batteries, the accurate assessment of battery state of health (SOH) is of paramount importance. With the aim of elevating the SOH estimation exactitude and remedying the model degradation induced by data paucity, this paper proposes an SOH estimation method that integrates a data-augmentation strategy with a Long Short-Term Memory (LSTM)-iTransformer model. Specifically, multiple health characteristic factors characterizing the aging behavior are first extracted from the battery charge–discharge curves and incremental capacity (IC) curves, and the features that are highly correlated with the SOH are screened by a Pearson correlation coefficient analysis. Subsequently, the data augmentation technique is used to extend the degradation sample set. The LSTM-iTransformer model is trained based on the extended samples and evaluated on multiple performance metrics. A comparative analysis reveals a marked enhancement in predictive accuracy achieved by this method over the baseline model trained with the initial data, which validates the effectiveness of the data augmentation strategy in improving the performance of SOH estimation models. Additionally, in scenarios characterized by abundant data availability, the direct application of this model facilitates enhanced predictive precision. Full article