Search Results (1,773)

Text-to-image (T2I) generation, a core component of generative artificial intelligence(AI), is increasingly important for creative industries and human–computer interaction. Despite impressive progress in realism and diversity, diffusion models still exhibit critical security blind spots particularly in the Transformer key-value mapping mechanism that underpins cross-modal alignment. Existing backdoor attacks often rely on large-scale data poisoning or extensive fine-tuning, leading to low efficiency and limited stealth. To address these challenges, we propose two efficient backdoor attack methods AttnBackdoor and SemBackdoor grounded in the Transformer’s key-value storage principle. AttnBackdoor injects precise mappings between trigger prompts and target instances by fine-tuning the key-value projection matrices in U-Net cross-attention layers (≈5% of parameters). SemBackdoor establishes semantic-level mappings by editing the text encoder’s MLP projection matrix (≈0.3% of parameters). Both approaches achieve high attack success rates (>90%), with SemBackdoor reaching 98.6% and AttnBackdoor 97.2%. They also reduce parameter updates and training time by 1–2 orders of magnitude compared to prior work while preserving benign generation quality. Our findings reveal dual vulnerabilities at visual and semantic levels and provide a foundation for developing next generation defenses for secure generative AI. Full article

Rapid and accurate acquisition of landslide inventories is essential for effective disaster relief. Deep learning-based pixel-wise semantic segmentation of remote sensing imagery has greatly advanced in landslide mapping. However, the heavy dependance on extensive annotated labels and sensitivity to domain shifts severely constrain the model performance in unseen domains, leading to poor generalization. To address these limitations, we propose LandsDANet, an innovative unsupervised domain adaptation framework for cross-domain landslide identification. Firstly, adversarial learning is employed to reduce the data distribution discrepancies between the source and target domains, thereby achieving output space alignment. The improved SegFormer serves as the segmentation network, incorporating hierarchical Transformer blocks and an attention mechanism to enhance feature representation capabilities. Secondly, to alleviate inter-domain radiometric discrepancies and attain image-level alignment, a Wallis filter is utilized to perform image style transformation. Considering the class imbalance present in the landslide dataset, a Rare Class Sampling strategy is introduced to mitigate bias towards common classes and strengthen the learning of the rare landslide class. Finally, a contrastive loss is adopted to further optimize and enhance the model’s ability to delineate fine-grained class boundaries. The proposed model is validated on the Potsdam and Vaihingen benchmark datasets, followed by validation in two landslide scenarios induced by earthquakes and rainfall to evaluate its adaptability across different disaster domains. Compared to the source-only model, LandsDANet achieved improvements in IoU of 27.04% and 35.73% in two cross-domain landslide disaster recognition tasks, respectively. This performance not only showcases its outstanding capabilities but also underscores its robust potential to meet the demands for rapid response. Full article

To maintain reliability, safety, and sustainability in power transmission, insulator defect detection has become a critical task in power line inspection. Due to the complex backgrounds and small defect sizes encountered in insulator defect images, issues such as false detections and missed detections often occur. The existing You Only Look Once (YOLO) object detection algorithm is currently the mainstream method for image-based insulator defect detection in power lines. However, existing models suffer from low detection accuracy. To address this issue, this paper presents an improved YOLOv5-based MC-YOLO insulator detection algorithm. To effectively extract multi-scale information and enhance the model’s ability to represent feature information, a multi-scale attention convolutional fusion (MACF) module incorporating an attention mechanism is proposed. This module utilises parallel convolutions with different kernel sizes to effectively extract features at various scales and highlights the feature representation of key targets through the attention mechanism, thereby improving the detection accuracy. Additionally, a cross-context feature fusion module (CCFM) is designed, where shallow features gain partial deep semantic supplementation and deep features absorb shallow spatial information, achieving bidirectional information flow. Furthermore, the Spatial-Channel Dual Attention Module (SCDAM) is introduced into CCFM. By incorporating a dynamic attention-guided bidirectional cross-fusion mechanism, it effectively resolves the feature deviation between shallow details and deep semantics during multi-scale feature fusion. The experimental results show that the MC-YOLO algorithm achieves an mAP@0.5 of 67.4% on the dataset used in this study, which is a 4.1% improvement over the original YOLOv5. Although the FPS is slightly reduced compared to the original model, it remains practical and capable of rapidly and accurately detecting insulator defects. Full article

Emotion recognition from physiological signals is pivotal for advancing human–computer interaction, yet unimodal pipelines frequently underperform due to limited information, constrained data diversity, and suboptimal cross-modal fusion. Addressing these limitations, the Self-Attention Wasserstein Generative Adversarial Network with Bidirectional Cross-Modal Attention (SAWGAN-BDCMA) framework is proposed. This framework reorganizes the learning process around three complementary components: (1) a Self-Attention Wasserstein GAN (SAWGAN) that synthesizes high-quality Electroencephalography (EEG) and Photoplethysmography (PPG) to expand diversity and alleviate distributional imbalance; (2) a dual-branch architecture that distills discriminative spatiotemporal representations within each modality; and (3) a Bidirectional Cross-Modal Attention (BDCMA) mechanism that enables deep two-way interaction and adaptive weighting for robust fusion. Evaluated on the DEAP and ECSMP datasets, SAWGAN-BDCMA significantly outperforms multiple contemporary methods, achieving 94.25% accuracy for binary and 87.93% for quaternary classification on DEAP. Furthermore, it attains 97.49% accuracy for six-class emotion recognition on the ECSMP dataset. Compared with state-of-the-art multimodal approaches, the proposed framework achieves an accuracy improvement ranging from 0.57% to 14.01% across various tasks. These findings offer a robust solution to the long-standing challenges of data scarcity and modal imbalance, providing a profound theoretical and technical foundation for fine-grained emotion recognition and intelligent human–computer collaboration. Full article

The nonholonomic constraints of fixed-wing UAVs, characterized by coupled heading-curvature feasibility and asymmetric costs, fundamentally deviate from classical Euclidean routing assumptions. While standard neural combinatorial optimization (NCO) architectures could theoretically incorporate Dubins costs via reward signals, such naive adaptations lack the capacity to explicitly model the intrinsic SE(2) geometric invariance and directional asymmetry of fixed-wing motion, leading to suboptimal generalization. To bridge this gap, we propose a Dubins-Aware NCO framework. We design a dual-channel embedding to decouple asymmetric physical distances from rotation-stable geometric features. Furthermore, we introduce a Rotary Phase Encoding (RoPhE) mechanism that theoretically guarantees strict SO(2) equivariance within the attention layer. Extensive sensitivity, ablation, and cross-distribution generalization experiments are conducted on tasks spanning varying turning radii and problem variants with instance scales of 10, 20, 36, and 52 nodes. The results consistently validate the superior optimality and stability of our approach compared with state-of-the-art DRL and NCO baselines, while maintaining significant computational efficiency advantages over classical heuristics. Our results highlight the importance of explicitly embedding geometry-physics consistency, rather than relying on scalar reward signals, for real-world fixed-wing autonomous scheduling. Full article

Continuous heart rate monitoring is crucial for early cardiovascular disease detection. To overcome the discomfort and limitations of ECG in home settings, we propose a multimodal temporal fusion network (MM-TFNet) that integrates ballistocardiography (BCG) and photoplethysmography (PPG) signals. The network extracts temporal features from BCG and PPG signals through temporal convolutional networks (TCNs) and bidirectional long short-term memory networks (BiLSTMs), respectively, achieving cross-modal dynamic fusion at the feature level. First, bimodal features are projected into a unified dimensional space through fully connected layers. Subsequently, a cross-modal attention weight matrix is constructed for adaptive learning of the complementary correlation between BCG mechanical vibration and PPG volumetric flow features. Combined with dynamic focusing on key heartbeat waveforms through multi-head self-attention (MHSA), the model’s robustness under dynamic activity states is significantly enhanced. Experimental validation using a publicly available BCG-PPG-ECG simultaneous acquisition dataset comprising 40 subjects demonstrates that the model achieves excellent performance with a mean absolute error (MAE) of 0.88 BPM in heart rate prediction tasks, outperforming current mainstream deep learning methods. This study provides theoretical foundations and engineering guidance for developing contactless, low-power, edge-deployable home health monitoring systems, demonstrating the broad application potential of multimodal fusion methods in complex physiological signal analysis. Full article

Accurate 3D reconstruction of guidewires is crucial in minimally invasive surgery and interventional procedures. Traditional biplanar X-ray–based reconstruction methods can achieve reasonable accuracy but involve high radiation doses, limiting their clinical applicability; meanwhile, single-view images inherently lack reliable depth cues. To address these issues, this paper proposes a multimodal guidewire 3D reconstruction approach that integrates magnetic field information. The method first employs the MiDaS v3 network to estimate an initial depth map from a single image and then incorporates tri-axial magnetic field measurements to enrich and refine the spatial information. To effectively fuse the two modalities, we design a multi-stage strategy combining nearest-neighbor matching (KNN) with a cross-modal attention mechanism (Cross-Attention), enabling accurate alignment and fusion of image and magnetic features. The fused representation is subsequently fed into a PointNet-based regressor to generate the final 3D coordinates of the guidewire. Experimental results demonstrate that our method achieves a root-mean-square error of 2.045 mm, a mean absolute error of 1.738 mm, and a z-axis MAE of 0.285 mm on the test set. These findings indicate that the proposed multimodal framework improves 3D reconstruction accuracy under single-view imaging and offers enhanced visualization support for interventional procedures. Full article

Background/Objectives: Cardiovascular diseases (CVDs) remain the leading cause of morbidity and mortality worldwide and increasingly are recognized as a continuum of interconnected conditions rather than isolated entities. Methods: A structured narrative literature search was performed in PubMed, Scopus, and Google Scholar for publications from 2015 to 2025 using combinations of different keywords: “cardiovascular disease spectrum”, “multi-omics”, “precision cardiology”, “machine learning”, and “artificial intelligence in cardiology”. Results: Evidence was synthesized across seven major clusters of cardiovascular conditions, and across these domains, common biological pathways were mapped onto heterogeneous clinical phenotypes, and we summarize how multi-omics integration, AI-enabled imaging and digital tools contribute to improved risk prediction and more informed clinical decision-making within this spectrum. Conclusions: Interpreting cardiovascular conditions as components of a shared disease spectrum clarifies cross-disease interactions and supports a shift from organ- and syndrome-based classifications toward mechanism- and data-driven precision cardiology. The convergence of multi-omics, and AI offers substantial opportunities for earlier detection, individualized prevention, and tailored therapy, but requires careful attention to data quality, equity, interpretability, and practical implementation in routine care. Full article

Wild ginkgo, as an endangered species, holds significant value for genetic resource conservation, yet its practical applications face numerous challenges. Traditional field surveys are inefficient in mountainous mixed forests, while satellite remote sensing is limited by spatial resolution. Current deep learning approaches relying on single-source data or merely simple multi-source fusion fail to fully exploit information, leading to suboptimal recognition performance. This study presents a multimodal ginkgo crown dataset, comprising RGB and multispectral images acquired by an UAV platform. To achieve precise crown segmentation with this data, we propose a novel dual-branch dynamic weighting fusion network, termed dual-branch cross-modal attention-enhanced UNet (DCA-UNet). We design a dual-branch encoder (DBE) with a two-stream architecture for independent feature extraction from each modality. We further develop a cross-modal interaction fusion module (CIF), employing cross-modal attention and learnable dynamic weights to boost multi-source information fusion. Additionally, we introduce an attention-enhanced decoder (AED) that combines progressive upsampling with a hybrid channel-spatial attention mechanism, thereby effectively utilizing multi-scale features and enhancing boundary semantic consistency. Evaluation on the ginkgo dataset demonstrates that DCA-UNet achieves a segmentation performance of 93.42% IoU (Intersection over Union), 96.82% PA (Pixel Accuracy), 96.38% Precision, and 96.60% F1-score. These results outperform differential feature attention fusion network (DFAFNet) by 12.19%, 6.37%, 4.62%, and 6.95%, respectively, and surpasses the single-modality baselines (RGB or multispectral) in all metrics. Superior performance on cross-flight-altitude data further validates the model’s strong generalization capability and robustness in complex scenarios. These results demonstrate the superiority of DCA-UNet in UAV-based multimodal ginkgo crown recognition, offering a reliable and efficient solution for monitoring wild endangered tree species. Full article

With rapid advancements in sub-meter satellite and aerial imaging technologies, high-resolution remote sensing imagery has become a pivotal source for geospatial information acquisition. However, current semantic segmentation models encounter two primary challenges: (1) the inherent trade-off between capturing long-range global context and preserving precise local structural details—where excessive reliance on downsampled deep semantics often results in blurred boundaries and the loss of small objects and (2) the difficulty in modeling complex scenes with extreme scale variations, where objects of the same category exhibit drastically different morphological features. To address these issues, this paper introduces MAFMamba, a multi-scale adaptive fusion visual Mamba network tailored for high-resolution remote sensing images. To mitigate scale variation, we design a lightweight hybrid encoder incorporating an Adaptive Multi-scale Mamba Block (AMMB) in each stage. Driven by a Multi-scale Adaptive Fusion (MSAF) mechanism, the AMMB dynamically generates pixel-level weights to recalibrate cross-level features, establishing a robust multi-scale representation. Simultaneously, to strictly balance local details and global semantics, we introduce a Global–Local Feature Enhancement Mamba (GLMamba) in the decoder. This module synergistically integrates local fine-grained features extracted by convolutions with global long-range dependencies modeled by the Visual State Space (VSS) layer. Furthermore, we propose a Multi-Scale Cross-Attention Fusion (MSCAF) module to bridge the semantic gap between the encoder’s shallow details and the decoder’s high-level semantics via an efficient cross-attention mechanism. Extensive experiments on the ISPRS Potsdam and Vaihingen datasets demonstrate that MAFMamba surpasses state-of-the-art Convolutional Neural Network (CNN), Transformer, and Mamba-based methods in terms of mIoU and mF1 scores. Notably, it achieves superior accuracy while maintaining linear computational complexity and low memory usage, underscoring its efficiency in complex remote sensing scenarios. Full article

The rapid expansion of wind power as a key component of global renewable energy systems has led to the widespread deployment of wind turbines in environments exposed to diverse natural hazards. While hazard effects are often investigated individually, real wind turbine systems frequently experience concurrent or sequential hazards over their operational lifetime, giving rise to interaction effects that are not adequately captured by conventional design approaches. This paper presents Part 2 of a comprehensive review on natural hazards affecting wind turbine performance, combining bibliometric keyword co-occurrence analysis with a critical synthesis of recent technical studies. The review focuses on earthquakes, sea waves, and extreme wind events, while also highlighting other hazard types that have received comparatively limited attention in the literature, examining their effects on wind turbine systems and the mitigation strategies reported to address associated risks. Rather than treating hazards in isolation, their impacts are synthesised through cross-hazard interaction pathways and component-level failure modes. The findings indicate that wind turbine vulnerability under multi-hazard conditions is governed not only by load magnitude but also by hazard-induced changes in system properties and operational state. Key research gaps are identified, emphasising the need for state-aware, mechanism-consistent multi-hazard assessment frameworks to support the resilient design and operation of future wind energy systems. Full article

Infrared images are often degraded by complex noise due to hardware and environmental factors, posing challenges for subsequent processing and target detection. To overcome the shortcomings of existing denoising methods in balancing noise removal and detail preservation, this paper proposes a Wavelet Transform Enhanced Infrared Denoising Model (WTEIDM). Firstly, a Wavelet Transform Self-Attention (WTSA) is designed, which combines the frequency-domain decomposition ability of the discrete wavelet transform (DWT) with the dynamic weighting mechanism of self-attention to achieve effective separation of noise and detail. Secondly, a Multi-Scale Gated Linear Unit (MSGLU) is devised to improve the ability to capture detail information and dynamically control features through dual-branch multi-scale depth-wise convolution and gating strategy. Finally, a Parallel Hybrid Attention Module (PHAM) is proposed to enhance cross-dimensional feature fusion effect through the parallel cross-interaction of spatial and channel attention. Extensive experiments are conducted on five infrared datasets under different noise levels (σ = 15, 25, and 50). The results demonstrate that the proposed WTEIDM outperforms several state-of-the-art denoising algorithms on both PSNR and SSIM metrics, confirming its superior generalization capability and robustness. Full article

As a core component of the gearbox, bearings are crucial to the stability and reliability of the transmission system. However, dynamic variations in operating conditions and complex noise interference present limitations for existing fault diagnosis methods in processing non-stationary signals and capturing complex features. To address the aforementioned challenges, this paper proposes a bearing fault diagnosis method based on a multi-feature fusion dual-channel CNN-Transformer-CAM framework. The model cross-fuses the two-dimensional feature images from Gramian Angular Difference Field (GADF) and Generalized S Transform (GST), preserving complete time–frequency domain information. First, a dual-channel parallel convolutional structure is employed to separately sample the generalized S-transform (GST) maps and the Gramian Angular Difference Field (GADF) maps, enriching fault information from different dimensions and effectively enhancing the model’s feature extraction capability. Subsequently, a Transformer structure is introduced at the backend of the convolutional neural network to strengthen the representation and analysis of complex time–frequency features. Finally, a cross-attention mechanism is applied to dynamically adjust features from the two channels, achieving adaptive weighted fusion. Test results demonstrate that under conditions of noise interference, limited samples, and multiple operating states, the proposed method can effectively achieve the accurate assessment of bearing fault conditions. Full article

Exercise-induced myokines have emerged as crucial mediators of the beneficial effects of physical activity on neurodegenerative diseases through complex molecular mechanisms involving oxidative stress reduction, neuroinflammation suppression, and synaptic plasticity enhancement. Among these myokines, irisin, encoded by the FNDC5 gene, has gained significant attention as a potential therapeutic target in neurodegenerative conditions due to its ability to cross the blood–brain barrier and exert pleiotropic neuroprotective effects. This review synthesizes current evidence from both preclinical and clinical studies examining the role of exercise-induced irisin in neurodegeneration, with particular emphasis on translational potential and therapeutic applications. A comprehensive search was conducted across PubMed, Web of Science, Scopus, and EMBASE databases (spanning January 2015 to December 2024) to identify peer-reviewed articles investigating irisin’s neuroprotective mechanisms in neurodegenerative diseases. Ten studies met the inclusion criteria (five rodent/primate model studies and five human clinical investigations), which were analyzed for methodological rigor, intervention protocols, biomarker quantification methods, and reported outcomes. Reviewed studies consistently demonstrated that exercise-induced endogenous irisin elevation correlates with improved cognitive function, reduced neuroinflammatory markers, enhanced synaptic plasticity, and modulation of neurodegenerative pathways, with exogenous irisin administration reproducing several neuroprotective benefits observed with exercise training in animal models. However, substantial heterogeneity exists regarding exercise prescription parameters (intensity, duration, frequency, modality), training-induced irisin quantification methodologies (ELISA versus mass spectrometry), and study designs (ranging from uncontrolled human observations to randomized controlled trials in animal models). Critical appraisal reveals that human studies lack adequate control for confounding variables including baseline physical fitness, comorbidities, concurrent medications, and potential sources of bias, while biochemical studies indicate distinct pharmacokinetics between endogenous training-induced irisin and exogenous bolus dosing, necessitating careful interpretation of therapeutic applicability. The translational potential of irisin as a therapeutic agent or drug target depends on resolving methodological standardization in biomarker measurement, conducting well-designed clinical trials with rigorous control for confounders, and integrating findings from molecular/biochemical studies to elucidate mechanisms linking irisin to disease modification. Future research should prioritize establishing clinical trial frameworks that harmonize exercise prescriptions, employ robust biomarker quantification (mass spectrometry), and stratify participants based on disease stage, comorbidities, and genetic predisposition to clarify irisin’s role as a potential therapeutic intervention in neurodegenerative disease management. Full article

To address the dual demands of accuracy and real-time performance in power quality disturbance (PQD) recognition for new power system, this paper proposes a lightweight model named the Cross-Channel Attention Three-Layer Convolutional Model (1D-CCANet-3), specifically designed for edge deployment. Based on the one-dimensional convolutional neural network (1D-CNN), the model features an ultra-compact architecture with only three convolutional layers and one fully connected layer. By incorporating a set of cross-channel attention (CCA) mechanisms in the final convolutional layer, the model further enhances disturbance recognition accuracy. Compared to other deep learning models, 1D-CCANet-3 significantly reduces computational and storage requirements for edge devices while achieving accurate and efficient PQD recognition. The model demonstrates robust performance in recognizing 10 types of PQD under varying signal-to-noise ratio (SNR) conditions. Furthermore, the model has been successfully deployed on the FPGA platform and exhibits high recognition accuracy and efficiency in real-world data validation. This work provides a feasible and effective solution for accurate and real-time PQD monitoring on edge devices in new power systems. Full article