Search Results (114)

Accurately forecasting waypoint-level tropical cyclone (TC) intensity, defined as the local wind speed at specific maritime route waypoints under TC influence, is crucial for navigation safety and voyage planning. Conventional studies mainly focus on the central intensity of TC systems and underutilize the complementary value of multimodal meteorological data with inconsistent sampling intervals. To address these challenges, this study proposes a multimodal-augmented conditional diffusion model (MADiff) for waypoint-level TC intensity prediction. To exploit the potential of multimodal inputs, we first design a temporal-adaptive dynamic convolution module (TDConv) to capture multi-timescale features, mitigating multimodal sampling discrepancies without rigid temporal alignment. Second, we develop a discriminative cross-fusion module (DisCF) to aggregate multi-timescale features across diverse modalities, quantifying multimodal heterogeneity and integrating valuable modality-specific features while suppressing noise interference. Fused features are fed into a diffusion model with physics-informed regularization to generate final intensity forecasts. Extensive experiments on four Western North Pacific datasets show that MADiff achieves average MAE and RMSE values of 2.08 kt and 2.37 kt, respectively, for 12 h intensity forecasting. Compared with the state-of-the-art baseline (TC-Clouds-DP), MADiff yields substantial performance improvements, reducing MAE by 16.3% and RMSE by 10.6% on average. This study provides an effective framework for fine-grained TC intensity forecasting, offering valuable insights for extreme marine weather early warning and intelligent navigation decision-making. Full article

Multi-source information fusion is regarded as a key approach for improving bearing fault diagnosis. However, due to the heterogeneity of multi-source information, asymmetric information contributions, and imbalanced discriminative features, negative transfer may occur during fusion. To address this issue, this paper proposes a negative-transfer-suppression diagnosis framework based on physical-information guidance and adversarially disentangled representation. First, an adaptive preprocessing mechanism guided by acoustic–vibration cross-correlation and mutual information entropy is constructed to extract intrinsic cross-modal correlations, enabling source-end feature reconstruction and commonality enhancement. Second, an attention-based spatial feature extraction operator and an adversarial common-domain representation model are developed to suppress modality-specific interference and disentangle cross-modal shared features. On this basis, sparse coding is employed to fuse common-domain and modality-specific features. Furthermore, a classification effectiveness evaluation index based on fuzzy clustering is introduced into the loss function to dynamically constrain sparse coding weights, thereby reducing interference features and suppressing negative transfer under strong-noise conditions. Experimental results demonstrate that the proposed method effectively achieves the design objective that “fusion outperforms non-fusion,” exhibiting strong noise robustness and high diagnostic accuracy under complex operating conditions. Full article

Deep learning has been widely deployed in critical scenarios such as the Internet of Things (IoT), industrial sensing, network intrusion detection, and cyber-physical system monitoring, where model inference directly affects system security, operational reliability, and service continuity. However, existing adversarial example detection studies remain insufficient for practical IoT deployment, as their validation often overlooks endpoint resource constraints, heterogeneous data modalities, physical environmental interference, communication protocol specifications, adaptive attacks, and adversary capability models. Moreover, detection outcomes are rarely connected with deployment locations, computational overhead, formal security assurance, and subsequent response strategies, which limits their engineering applicability. To address these limitations, this review systematically synthesizes recent representative studies in adversarial example detection and constructs a unified analytical framework integrating detection evidence, IoT deployment feasibility, and adaptive-attack evaluation. Based on the source of detection evidence, existing methods are categorized into input-consistency-based, feature-statistics-based, predictive-uncertainty-based, model-reconstruction-based, runtime-context-aware, and multi-strategy fusion detection, while formal certification is discussed as an independent security-assurance dimension. The review further analyzes the principles, applicable conditions, limitations, compatibility conflicts with IoT deployment constraints, and typical failure modes of these methods. The analysis identifies four key challenges: the lack of IoT-native adaptive evaluation, limited anomaly-boundary identification and cross-modal generalization, insufficient deployment-time security assurance, and weak coordination between detection decisions and security responses. Future research should therefore emphasize feasible attack paradigms, hierarchical lightweight detection, reliable multimodal fusion, certifiable operational boundaries, and auditable end-to-end response mechanisms, thereby supporting the evaluation and deployment of adversarial example detection in IoT scenarios. Full article

Existing multi-model fusion methods generally adopt a global, fixed fusion strategy, applying all base models and weighting rules uniformly to all samples to be classified and all categories, thereby lacking adaptability to specific samples and category-specific targeting. Most traditional dynamic selection and dynamic weighting fusion methods only implement global parameter adjustments at the sample level, without considering the significant differences in category-specific capabilities among the base models. When a single base model exhibits superior recognition capabilities for only certain categories, its prediction accuracy and confidence for the remaining categories are low. If all base models are fused directly, poor-quality class predictions can cause negative interference and even dominate the final decision, leading to classification errors. Furthermore, traditional dynamic fusion suffers from computational redundancy, difficulty in suppressing interference from low-confidence samples, and the challenge of balancing dynamic optimization with inference efficiency. To address these issues, this paper proposes an improved multi-model dynamic fusion classification technique that differs from the traditional global dynamic fusion paradigm. By constructing a voting matrix, contribution weights, and a matrix of effective category voting weights, this method establishes a category-level model performance evaluation and differentiated weighting mechanism. This enables the precise selection of superior base models for each sample and category, thereby filtering out interference from low-confidence and suboptimal category predictions. At the same time, in the network architecture design, lightweight models are organized into a branch structure, and effective branches are dynamically activated as needed to participate in decision-making, significantly reducing the computational overhead of inference. To validate the fusion classification technique proposed in this paper, for the experiments, we selected mainstream lightweight models such as MobileNetV2, EfficientNetB0, ShuffleNetv2, MNASNet 0.75, and MobileNetV3_Small for evaluation on the NEU dataset and NASA’s Milling Data Set. The experimental results demonstrate that the fusion method proposed in this paper can fully aggregate the category-specific strengths of different lightweight models, effectively mitigate the risk of misclassification associated with traditional fusion methods, and enhance model robustness while ensuring high classification accuracy. It achieves classification performance comparable to that of large deep models with extremely low computational overhead. This method is not only suitable for application in multiple-criteria decision-making but can also be implemented and extended to multi-source/multi-modal data fusion and deep neural networks, making it of practical value. Full article

For large-scale photovoltaic plant inspection, UAV-based infrared–visible real-time detection can combine thermal abnormality information with appearance and structural cues. This is useful for improving inspection and maintenance efficiency. However, in real UAV inspection, differences in sensor resolution, field of view, and flight attitude can cause weak misalignment between the two modalities. Since complex image registration is difficult to perform before real-time inference, this misalignment can affect cross-modal feature fusion and defect localization. To address this problem, this paper proposes Frequency-Aware Fusion YOLO (FAF-YOLO) for dual-modal photovoltaic defect detection. We also build a real-scene infrared–visible dual-modal photovoltaic defect dataset, named DM-PV, which covers six defect categories related to thermal anomalies and external environmental interference. FAF-YOLO is based on a dual-branch YOLO detection framework. The C3k2-DPRG module is used to enhance defect boundaries, local details, and neighborhood context. The Frequency-aware Selective Fusion (FSF) module models low-frequency structural information and high-frequency detail responses separately, which reduces edge ghosting and background mis-fusion caused by weak misalignment. A Multi-Scale Differentiated Decoupled Head is then used to handle scale-specific prediction and improve small-defect localization and regional-anomaly discrimination. Experimental results show that FAF-YOLO achieves 92.5% Precision, 86.7% Recall, 91.7% ${\mathrm{mAP}}_{50}$, and 61.4% ${\mathrm{mAP}}_{50:95}$ on the DM-PV dataset. It outperforms several mainstream dual-modal detection methods and has lower parameters and computational complexity. Further tests for real-time inspection show that the proposed method keeps more stable performance under weak misalignment perturbations. It also reaches an inference speed of 33 FPS on the Jetson Orin Nano edge platform, which verifies its effectiveness and deployability for UAV-based real-time photovoltaic inspection. Full article

In complex UAV aerial scenes, small object detection tasks face challenges such as extremely low pixel occupancy, strong background interference, and sparse effective features, which are further compounded by environmental factors like low illumination. Consequently, single-modality detection algorithms are prone to severe target feature loss and missed detections. Multi-modal image fusion, which complements the texture details of visible light with the thermal radiation characteristics of infrared, is considered an effective approach to overcome the limitations of single physical imaging. However, conventional fusion mechanisms often suffer from semantic gaps when processing heterogeneous data, easily introducing redundant noise and background false alarms. To further improve the accuracy and robustness of small object detection in UAV aerial scenes, this paper proposes a multi-modal detection network that integrates direction-aware multi-granularity and asymmetric context guidance, termed DMAC-Net. Specifically, a Direction-Aware Granularity Enhancement (DAGE) module is first constructed for unified backbone feature extraction, which captures local directions and contour edges of small objects in UAV aerial images with high sensitivity, and expands the receptive field through a multi-granularity mechanism, effectively suppressing false positives induced by complex backgrounds while enhancing the recall of occluded and weakly featured targets. Additionally, the Asymmetric Context Guided Fusion (ACGF) module builds a spatial mechanism via asymmetric receptive fields and performs semantic soft alignment of cross-modal features with dynamic weight assignment, effectively filtering out artifacts and clutter from cross-modal interaction. Experimental results on multiple aerial datasets, including RGBTDronePerson, AVMS and LLVIP demonstrate that the proposed method outperforms existing mainstream models in terms of overall detection accuracy and missed-detection suppression, while exhibiting strong generalization capability and stability under complex lighting transitions and multi-scale variations in UAV monitoring environments. Full article

While generative models have become a standard approach for addressing the semantic-to-visual gap in Generalized Zero-Shot Learning (GZSL), existing architectures often struggle with two persistent limitations: cross-modal interference during condition fusion and severe overfitting to the visual distributions of seen classes. To address these bottlenecks, this paper introduces SemanticFlowNet, a framework based on Decoupled Semantic Flow Matching. Specifically, we propose a Decoupled Multi-modal Conditioning mechanism that relies on channel-wise concatenation of temporal encodings, semantic attributes, and visual contexts, which preserves the orthogonal subspaces of each modality and reduces interference. Additionally, we integrate a Dropout-enhanced Adaptive Layer Normalization (AdaLN) module to perturb the rigid memorization of seen classes, utilizing stochastic dropout within the state evolution to simulate the distributional variance of unseen domains. Finally, a Time-Aware Dynamic Reconstruction Penalty is introduced to enforce progressively stricter semantic alignment as the generative ordinary differential equation (ODE) trajectory converges to the target manifold. Evaluations on the CUB, SUN, and AWA2 benchmarks demonstrate the effectiveness of the proposed framework. Notably, SemanticFlowNet achieves a harmonic mean of 77.90% on the CUB dataset in the single-seed full-model setting, providing a competitive baseline for generative GZSL applications. Full article

Object detection in real-world scenarios is often challenged by adverse visual conditions, such as low illumination, strong glare, and dense fog, which severely degrade visible-spectrum features and lead to missed detections, inaccurate localization, and reduced detection accuracy. To address these issues, this paper proposes AQFS-Net, a dual-modal fusion detection network for visible-infrared object detection. Built upon YOLOv13, AQFS-Net adopts a symmetric dual-branch backbone by incorporating infrared images, thereby exploiting the complementary information between the visible and infrared modalities. To alleviate the negative transfer caused by conventional static fusion strategies, an Adaptive Quality-Aware Fusion Module (AQFM) is designed to dynamically enhance informative features and suppress degraded information according to the modality-specific reliability of different regions. In addition, a Foreground-Aware Saliency Guidance (FASG) branch is introduced to guide the network to focus on target regions through foreground supervision, reducing interference from complex backgrounds. Experimental results on the public LLVIP and M3FD datasets show that the proposed method improves mAP@0.5 by 6.8 and 3.1 percentage points, respectively, compared with the baseline using only visible images. These results demonstrate the effectiveness of AQFS-Net in improving dual-modal fusion quality and detection performance under challenging visual conditions, providing a practical reference for visible-infrared object detection in complex illumination scenarios. Full article

Damage localization based on ultrasonic guided waves (UGWs) ensures the reliability and safety of composites. Efficient and accurate damage localization requires full integration of different modality features. However, existing deep learning-based damage localization methods usually focus on single-modal features and cannot deeply mine and fuse different features. In this paper, we propose a novel Multimodal Feature Sensing and Fusion Neural Network (MSFN) for damage localization by UGWs in composites. This method uses an innovative multimodal input mode, in which three different modal signals, namely, the damage signal, scattered wave signal, and energy density signal, are fed into the network as inputs. We use Convolutional Neural Networks (CNNs), Gated Recurrent Units (GRUs) and Bidirectional Gated Recurrent Units (BiGRUs) to construct specific encoders for the characteristics of the three signals to extract the features of different modalities efficiently and quickly. Then we employ an attention mechanism-guided feature fusion strategy to aggregate the various features, map out the correlation between the damage zones and the signal features, and finally decode them through successive linear layers to output the final damage localization results. Subsequent experimental results show that the damage localization accuracy of the MSFN can reach 98.13% even under noise interference. It is shown that its robustness and accuracy are much better than those of other existing networks and it has better localization speed and generalization. The proposed MSFN architecture comprises a CNN-based DS-encoder, a GRU-based SW-encoder, and a BiGRU-based ES-encoder, followed by an attention-guided fusion module, demonstrating its feasibility for near-real-time SHM applications. Full article

Robust monitoring of weld pool dynamics is critical for automated arc welding; however, single-modality sensors are frequently constrained by severe optical interference and high-frequency environmental noise. To address these limitations, we propose M2WPR-Net, a novel multimodal framework that synergizes visual and acoustic signals for simultaneous weld width regression and physical quality classification. The architecture employs a dual-stream ResNet50 backbone to process heterogeneous sensory data. Specifically, the visual stream utilizes a Convolutional Block Attention Module (CBAM) to suppress intense arc glare and localize the weld pool. Concurrently, the acoustic stream transforms 1D audio sequences into 2D Gramian Angular Summation Field (GASF) textures, which are subsequently refined by Squeeze-and-Excitation (SE) networks to isolate target frequency channels. A central contribution of this study is a bidirectional cross-modal attention mechanism based on Query–Key–Value (Q-K-V) matrix operations. Overcoming the shortcomings of static feature concatenation, this module dynamically aligns the modalities, enabling acoustic cues to guide visual feature extraction and vice versa, thereby mitigating information bottlenecks. Optimized via a joint multi-task loss function, the proposed M2WPR-Net significantly outperforms existing single-modal and conventional fusion baselines. Experimental results demonstrate that the network achieves a Mean Absolute Error (MAE) of 0.18 mm for width prediction and a 93.5% accuracy in penetration state classification, confirming its resilience and practical applicability in complex industrial welding environments. Full article

Background: Prolonged sitting during lecture-based academic activities may be accompanied by sustained attentional engagement and cognitive fatigue, with potential consequences for cognitive efficiency and broader psychological functioning. Physical activity breaks (PABs) represent a feasible strategy to interrupt sitting time within university timetables, yet evidence in higher-education settings remains limited, particularly regarding modality-specific effects. This randomized crossover study examined the acute effects of a 10 min PAB on attentional and executive functioning in undergraduate students and compared an outdoor physical activity break (OPAB) with an exergame-based PAB (PABEx) versus a no-break control (NPAB). Methods: Forty-two undergraduate students (26 males, 16 females; mean age = 22.78 ± 5.84 years) completed three weekly conditions in randomized order following two consecutive hours of seated lectures: NPAB (seated rest), OPAB (2 min warm-up, 6 min light-to-moderate walking, 2 min cool-down), and PABEx (2 min warm-up, 6 min Fruit Ninja Kinect, 2 min cool-down). Cognitive performance was assessed immediately after each condition using the Trail Making Test A-B (TMT A-B) and the Stroop Color-Word Test (SCWT). Results: Significant condition effects were found for TMT-A (χ² = 53.976, p < 0.001), TMT-B (χ² = 44.635, p < 0.001), TMT B-A (χ² = 10.841, p = 0.004), SCWT interference time (χ² = 44.714, p < 0.001), and SCWT interference errors (χ² = 23.211, p < 0.001). Post-hoc tests showed that both OPAB and PABEx were associated with better performance on TMT-A, TMT-B, and SCWT interference time versus NPAB (all Benjamini–Hochberg-adjusted p < 0.001); PABEx was associated with better TMT-A performance than OPAB (Benjamini–Hochberg-adjusted p = 0.047). TMT B-A decreased only for OPAB versus NPAB (Benjamini–Hochberg-adjusted p = 0.009). SCWT interference errors were lower for OPAB versus NPAB (Benjamini–Hochberg-adjusted p < 0.001) and for PABEx versus NPAB (Benjamini–Hochberg-adjusted p = 0.012). Conclusions: A 10 min PAB implemented immediately after a lecture-based academic activity was associated with more favorable post-condition attentional and executive performance in undergraduate students compared with a passive no-break condition. OPAB and PABEx yielded broadly comparable benefits across executive outcomes, whereas PABEx showed an additional advantage for TMT-A, suggesting a possible modality-specific effect on processing speed and visuoperceptual tracking. These findings support the integration of brief active breaks into university schedules as a pragmatic strategy to promote post-lecture cognitive efficiency during academically demanding periods. Trial registration: ClinicalTrials.gov, NCT07624084; retrospectively registered on 28 May 2026. Full article

The scientific foundations of sport-training methodology are commonly attributed to physiological principles; however, the extent to which these principles directly inform practical training models remains unclear. This narrative review examines the historical development of training theory—from early adaptology and integrative physiology to contemporary molecular discoveries in muscle biology—and evaluates their relevance to strength development. Strength expression is shown to be highly variable, influenced by neural, mechanical, technical, and psychological factors, challenging the traditional reliance on fixed percentages of maximal strength for training prescription. Additional complexities arise from individual response variability, performance plateaus, and the interference between molecular pathways activated by strength and endurance training. Emerging artificial intelligence systems offer new opportunities for individualized training optimization, injury prediction, and motor-learning analysis, while advances in brain decoding technologies highlight the potential role of willpower and cognitive processes in strength expression. Overall, current training methodologies remain heterogeneous and incomplete, although recent evidence increasingly supports modality-specific prescription principles based on load, contraction velocity, movement intent, and athlete training status. So, substantial research is required to more clearly connect physiological mechanisms with practical training applications. Full article

Underwater laser range-gated imaging (ULRGI) effectively suppresses water backscattering via time-slicing mechanisms, making it a primary modality for underwater vision. However, factors such as the inherent optical properties of water, intra-slice residual scattering, gating timing errors, and sensor noise make it difficult to separate target signals from the background. Consequently, the resulting images are generally affected by texture degradation and low contrast, severely limiting the accuracy of downstream tasks like object detection and environmental perception. To this end, we propose the use of a Wavelet-guided Multi-stage Gated Enhancement Network (WMGNet). Operating progressively across three stages, WMGNet’s first two stages employ an encoder–decoder architecture that leverages multi-scale frequency decomposition in the wavelet domain to pinpoint intra-slice scattering and decouple target signals from noise. To precisely extract fine details, we design a TextureBlock integrating feature gating (ConvGLU) and high-frequency attention (HFAttention). Additionally, a pixel-wise ground-truth guided attention module (GGAM) is introduced to optimize the precision and target-specificity of multi-stage feature fusion. Extensive comparative and ablation experiments demonstrate that the proposed WMGNet effectively eliminates scattering interference and restores texture details in underwater imaging. On our custom ULRGI dataset, it achieves state-of-the-art performance with a PSNR of 36.31 dB, an SSIM of 0.921, an MAE of 2.672, and an LPIPS of 0.060. Notably, it outperforms the second-best method by a margin of 3.06 dB in PSNR and reduces the MAE by 50.69%. Furthermore, evaluations on three public datasets confirm its robust cross-scenario generalization, yielding competitive PSNR values of 33.22 dB, 31.59 dB, and 32.06 dB, respectively. Overall, WMGNet provides a highly effective and robust solution for high-resolution underwater imaging. Full article

Ion channels are validated drug targets, but they remain difficult to study as their pharmacology is influenced by rapid gating, conformational state transitions, cell-type-specific expression, and narrow safety margins. Recent advances in cryo-electron microscopy, structure-based in silico screening, machine-learning-guided prioritization, optical high-throughput screening, automated patch-clamp electrophysiology, and human-relevant organoid or microphysiological system (MPS) models are transforming this field. In this expanded review, we examine how these modalities can be integrated into a hybrid discovery pipeline that begins with computational triage, proceeds through scalable functional screening and state-aware electrophysiological validation, and concludes with multi-channel safety de-risking and translational analysis in complex human models. We also discuss disease-associated channel remodeling in cancer and inflammatory disorders, with an emphasis on transient receptor potential channels, voltage-gated potassium channel 1.3 (Kv1.3), Piezo channels, transmembrane protein 16A/anoctamin-1 (TMEM16A/ANO1), chloride channels, and proarrhythmic safety risks. Additionally, we highlight unresolved challenges, including bias in artificial intelligence models, incomplete conformational sampling, assay interference, organoid heterogeneity, and regulatory acceptance of MPS platforms. This review proposes a staged decision framework in which computational prioritization, scalable functional screening, direct electrophysiological confirmation, safety pharmacology, DMPK assessment, and disease-relevant human models serve as complementary filters rather than competing platforms for the identification of selective and translatable ion-channel therapeutics. Full article

The pathological mechanism of autism spectrum disorder (ASD) exhibits dual heterogeneity: abnormal local energy metabolism and brain-wide high-order topological failure. To synergistically characterize these complex signals captured by advanced neuroimaging sensors, we propose the Quantum-Enhanced Graph Convolutional Wide Residual Network (QG-WRN), a modality-specific, decoupled parallel dual-stream architecture. In the classical branch, to accurately capture the spatial distribution of local metabolic abnormalities, we employ a wide residual network (WRN) to extract amplitude of low-frequency fluctuation (ALFF) features, leveraging its expanded feature channels to effectively mine regional neurodynamic properties. Furthermore, to overcome the representational bottlenecks of classical linear operators in parsing hidden, long-range network connections, we introduce quantum computing, exploiting its exponentially expansive state space and intrinsic low-parameter regularization mechanism. Guided by these properties, the quantum branch utilizes a variational quantum graph convolutional (QGCN) module—featuring a trainable circular encoding strategy and a hardware-efficient 4-qubit configuration—with a 2-layer nested message passing structure to process the functional connectivity (FC) matrix, harnessing quantum interference in Hilbert space to parse complex topology while effectively mitigating overfitting on small-sample medical data. A unified training scheme achieves full-dimensional fusion of node activity and topology. Achieving 68.49% accuracy, our method outperforms 10 classic and recent new baselines, providing a powerful computational intelligence tool for sensor-based ASD clinical diagnosis. Furthermore, interpretability analysis successfully maps core disease hubs to standard AAL116 atlas coordinates, providing a powerful tool for computationally aided ASD diagnosis. Full article