Search Results (221)

Aircraft cargo compartment fires constitute a significant type of aviation fire, posing a grave threat to aviation safety. To guard against and respond to such fires, existing aircraft cargo compartments are equipped with smoke detection fire detectors, which rely on perceiving changes in smoke transmittance to determine the onset of a fire. However, these detectors offer relatively low recognition accuracy and cannot provide a direct visual representation of the fire. In this work, we introduce a fire recognition method built on image sensors and a deep learning model. In light of the irregular shapes of flames and smoke, an improved interactive triplet attention mechanism (ITAM) is integrated into the You Only Look Once version 5 (YOLOv5) model, enhancing the model’s recognition accuracy. Furthermore, the original Neck structure is replaced with an Asymptotic Feature Pyramid Network (AFPN), improving the model’s ability to recognize small targets, which is particularly useful for detecting flames and smoke early in a fire. This paper further improves the model’s recognition accuracy by introducing the Focaler-IoU loss function, which balances the feature learning of hard and easy samples. Therefore, the network model in this paper is named IATF-YOLO. Ablation experiments demonstrate that our algorithm improves accuracy by 2%, while comparative experiments with several mainstream baseline models show that our algorithm achieves a 0.7% accuracy improvement, with a final peak accuracy of 93.6%. Full article

Machine-learned molecular force fields require many-body geometry, but obtaining it through Clebsch–Gordan tensor products is computationally expensive. For a strong no-Clebsch–Gordan backbone such as GotenNet, we ask whether the limitation in handling three-body geometry is one of representational capacity or one of training supervision, and separate the two factors with three controlled probes on a single-seed, paper-aligned rMD17 aspirin split. (i) While frame projection of tensor features is comparable to scalar cos-angle triplet cross-attention (SCTA) at pilot scale, algebraically its diagonal scalar collapses to a frame-independent inner product and the remaining channel is parity-odd, making SCTA’s cos-angle input the principled $\mathrm{O}\left(3\right)$ scalar choice. (ii) SCTA matches GotenNet’s converged force accuracy within ∼0.4% without independent gain, indicating that three-body representational capacity is not the binding constraint. (iii) A graph-level auxiliary loss on bond-angle and dihedral statistics gives the best force mean absolute error (MAE; $0.1280$ vs. $0.1303$ kcal/mol/Å) and reduces epochs-to-validation-target by 26–55%. Cross-molecule probes do not extend this finding; a paired salicylic acid comparison shows a directional degradation that, under a configuration-level paired block bootstrap, is significant and opposite in sign to the aspirin effect. Across three random seeds, the auxiliary force-MAE gain is small and seed-dependent but consistently reduces seed-to-seed variance and accelerates convergence, indicating that low-cost three-body supervision can be a more effective lever than added three-body capacity. Full article

Molecular property prediction is a fundamental task in drug discovery and materials design. While graph neural networks (GNNs) and SMILES-based Transformers have made significant strides, the former are often limited by local message-passing bottlenecks such as over-squashing, while the latter frequently lack explicit topological constraints and suffer from severe vocabulary imbalance. In this work, we revisit the granularity of molecular modeling and propose a representation learning framework built upon bond-level sequences. Our framework models molecules as sequences of directed bond tokens and introduces a structure-aware hybrid attention mechanism. By imposing hard topological constraints on a subset of attention heads to reinforce local connectivity while preserving global receptive fields in the remaining heads, the design is intended to separate short-range chemical bonding from long-range contextual dependencies. For pre-training, we implemented a multi-scale consistency learning paradigm, which utilizes an atom-centric group masking strategy to induce a hierarchical loss of local structural information and employs contrastive and triplet losses to ensure identity consistency across varying scales of structural degradation. Furthermore, by incorporating macro-scale physicochemical descriptors (e.g., LogP, TPSA) as global anchors, we examined how the inclusion of global attribute bias can provide weak physicochemical priors during pre-training, while its effect during downstream fine-tuning remains task-dependent. Experimental results demonstrate that our lightweight model, with approximately 3.5 million parameters, exhibits a dataset-dependent performance profile across MoleculeNet benchmarks and shows promising behavior on selected topology-sensitive tasks, particularly MUV. Ablation studies further analyze the contribution of bond-level connectivity, the stage-dependent dynamics of global attribute bias, structured masking, and pre-training configurations. Ultimately, this work provides an alternative representation design for molecular modeling, offering a parameter-efficient option for future molecular learning systems alongside traditional SMILES-based and graph-based formulations. Full article

Autonomous flight of unmanned aerial vehicles (UAVs) in Global Navigation Satellite System (GNSS)-denied environments critically depends on accurate and robust visual localization. To tackle the challenges of cross-view domain discrepancies and real-time high-precision matching, we propose GRiM-Net, a two-stage joint optimization visual localization network. First, a global retrieval module aggregates features and selects the most similar satellite map candidate patches from a pre-built index, efficiently narrowing the search from the global map to a local region. Next, a fine matching module performs pixel-level keypoint detection and description on the query image and candidate patches. Bidirectional matching and weighted homography estimation are then used to map the UAV image center to satellite coordinates, yielding precise geographic positions. Both modules share a backbone with domain-adaptive batch normalization, and joint optimization of global retrieval triplet loss with fine matching keypoint, descriptor, and homography reprojection losses enables synergistic enhancement of feature representations. Ablation and comparison experiments conducted on public urban cross-view benchmarks demonstrate that GRiM-Net can achieve efficient and robust geographic coordinate regression for UAVs, providing a practical localization component for broader navigation systems. Full article

Identifying the sources of Ultra-High Energy Cosmic Rays (UHECRs, $E>{10}^{18}$ eV) remains a fundamental challenge in astrophysics due to the significant deflections of charged particles by Galactic and extragalactic magnetic fields. Until now, dozens of events with energies over ${10}^{20}$ eV—Extreme Energy Cosmic Rays (EECRs)—were detected by the Pierre Auger Observatory and Telescope Array, but none of them showed a statistically significant association with potential sources. In this study, we investigate potential sources of EECRs with arrival directions from Local Void region. Since the energy loss lengths of such EECRs are of order of 20–40 Mpc, i.e., smaller than the Local Void extension (∼60 Mpc), potential sources should be predominantly Galactic ones. Since the most promising UHECR accelerators are mildly relativistic shocks, we consider Galactic microquasars, magnetars, and pulsar wind nebulae as potential sources of EECRs in the Local Void sky region. Using event-by-event reconstruction of trajectories of detected EECRs via CRPropa backtracking in the Galactic magnetic field, we find the potential Galactic sources and corresponding charges Z for some of the detected EECRs. The most promising coincidence is found between the EECR event triplet detected by PAO and TA and SGR 1900+14, a Galactic magnetar exhibiting high-energy flaring activity, with the inferred propagation time delay being consistent with the characteristic age of the magnetar. Full article

Antimicrobial resistance (AMR) is emerging as one of the most serious global health problems, necessitating the urgent development of alternative approaches to pathogen control. The present study describes the synthesis and characterization of a novel iodinated BODIPY derivative (BODIPY5), designed as a highly efficient photosensitizer for antimicrobial photodynamic inactivation (aPDI). The molecular design of the compound involves the introduction of two iodine atoms into the BODIPY5 core, which induces a “heavy atom effect”, accelerates the intersystem transition from the singlet to the triplet state, and leads to increased generation of singlet oxygen upon irradiation with visible light. Photophysical measurements show a significant fluorescence quenching of BODIPY5 compared to its unsubstituted counterpart, which is a direct indicator of increased photodynamic activity. The compound’s antimicrobial efficacy was tested in a homogeneous medium and after immobilization on cotton textiles via physical adsorption. In solution, BODIPY5 nearly eliminated the model bacterial strains B. cereus and P. aeruginosa at a low concentration of 10 µg/mL under light, with cell viability below 1%. The functionalized cotton fabric exhibits pronounced self-disinfection properties, retaining high photodynamic activity against the Gram-negative pathogen P. aeruginosa. Scanning electron microscopy results confirm extensive morphological damage and loss of structural integrity in bacterial cells on the treated textile following irradiation. The non-specific mechanism of action, which generates reactive oxygen species (¹O₂) in situ, prevents the development of bacterial resistance and makes the developed material a promising candidate for use in hospital environments, including antibacterial clothing and protective equipment. Full article

Cross-age face analysis has emerged as an important topic in biometric recognition due to substantial facial appearance variations caused by aging. Nevertheless, most existing approaches primarily focus on face verification (1:1 matching) and frequently rely on explicit age annotations, which limit their applicability in large-scale retrieval scenarios. In this study, large-scale cross-age face retrieval (1:N matching) is investigated, and a Hybrid Metric Learning Framework (HMLF) is proposed to learn age-invariant and retrieval-oriented facial representations without requiring age labels. The proposed framework integrates Additive Angular Margin Loss (ArcFace) with supervised contrastive learning to enhance feature discriminability. Furthermore, a mixed triplet mining strategy is introduced to improve the effectiveness of hard sample selection. A memory bank-based InfoNCE formulation is incorporated to provide a large number of negative samples, and an uncertainty-based adaptive weighting scheme is designed to automatically balance multiple loss components during optimization. To better simulate realistic retrieval scenarios, an extended cross-age retrieval evaluation protocol is established. Extensive experimental results demonstrate that the proposed framework achieves superior retrieval performance across different backbone architectures. The results further provide systematic insights into the influence of backbone design, loss formulation, and optimization strategies on cross-age retrieval accuracy. Full article

With the increasing demand for intelligent fault monitoring, acoustic-based diagnosis has emerged as a promising solution for industrial applications such as pipeline leakage and electrical equipment fault detection. However, complex working conditions and domain shifts significantly degrade model performance, especially when unseen target domain data is unavailable. To address this, we propose an amplitude-phase collaborative augmentation network named AP-CANet tailored for acoustic fault diagnosis. Specifically, the network adaptively aligns amplitude and phase features across multiple source domains and performs label-consistent sample augmentation to enrich data diversity while preserving semantic consistency. A frequency–spatial interaction module further integrates global spectral information with local temporal details to improve feature discriminability. Moreover, we introduce a manifold triplet loss that scales shortest path distances in the feature manifold, encouraging the model to better capture subtle distinctions among hard samples and improving intra-class compactness and inter-class separability. We evaluate the proposed method on two publicly available datasets: the Pipeline Leak Acoustic Dataset (GPLA-12) and the Electrical Sound Dataset (MIMII-DG). Experimental results demonstrate superior performance under domain-shift scenarios, highlighting the method’s potential for scalable and low-cost acoustic fault diagnosis in real-world industrial environments. Full article

To address the challenges of scarce unmanned aerial vehicle (UAV) track samples, severe class imbalance, and high motion similarity between UAVs and birds in low-altitude radar recognition, this paper proposes a trajectory classification method integrating a TCN-Transformer model with a physics-constrained TimeGAN (PC-TimeGAN) data augmentation framework. Specifically, the PC-TimeGAN generates high-quality, kinematically compliant UAV trajectories to alleviate data scarcity and class imbalance. A multi-scale TCN-Transformer is then constructed to comprehensively extract features, utilizing multi-kernel dilated convolutions for local temporal correlations and self-attention mechanisms for global temporal dependencies, thereby improving the discrimination between UAV and bird trajectories with similar motion patterns. Furthermore, a joint loss function combining Focal Loss and Triplet Loss is employed to optimize the decision boundaries and feature space, enhancing model robustness and generalization. Experiments on a measured dataset demonstrate that, under the 15-dimensional input setting, the proposed method achieves a UAV recall of 80.00%, an FAR of 3.15%, a precision of 64.00%, and an F1-score of 0.7111. Compared to baseline methods (e.g., SVM, LSTM, GRU, Transformer, and 1D-CNN), the proposed approach significantly improves UAV recall under limited trajectory information while keeping the false-alarm rate of misclassifying birds as UAVs low. Ultimately, this method markedly enhances the comprehensive performance of rapid track-level target classification for low-altitude surveillance radars. Full article

High-voltage bushings, the fault-prone key electrical components of transformers, are critical for real-time and high-accuracy fault monitoring and management. Intelligent fault detection via infrared images is plagued by low classification accuracy due to massive interference from similar tubular objects and small target characteristics. This study proposes a lightweight deep learning model, MTrip–YOLO, an improved YOLO11n integrated with Mamba-2 (Structured State Space Duality, SSD) and Triplet Attention, to achieve efficient fault monitoring in complex backgrounds. The training and validation dataset comprises open-source images, on-site data from a substation, and field-collected infrared images, categorized into four types: normal bushings, poor contact, oil shortage, and high dielectric loss faults. Mamba-2 captures the long-range global context of infrared features with its linear-complexity long-range modeling capability to enhance feature extraction, while Triplet Attention suppresses complex background radiation noise through cross-dimensional interaction without dimensionality reduction, enabling the model to focus on small targets and accurately classify bushings from morphologically similar strip-shaped objects. Experimental results show that MTrip–YOLO achieves a top mAP50 of 91.6% and a minimal parameter count of 1.9 M, outperforming Faster R-CNN, RT-DETR, and YOLO26n across all evaluated metrics and being potentially suitable for edge deployment on UAV-mounted or handheld infrared platforms, pending hardware validation on embedded computing devices. Ablation experiments verify the independent contributions of Mamba-2 (0.8027% mAP50 improvement) and Triplet Attention (0.89327% mAP50 improvement), with a synergistic effect from their combination. MTrip–YOLO provides a potential edge-deployable solution for high-voltage bushing fault monitoring, offering important application value for the intelligent operation and maintenance of substations. Full article

The rapid proliferation of low-altitude drones has led to increasingly congested and heterogeneous electromagnetic environments, posing significant challenges to fine-grained spectrum awareness and reliable drone management. Specific emitter identification (SEI), which exploits inherent hardware-dependent radio frequency fingerprints, provides an effective physical-layer solution for emitter-level discrimination. However, practical SEI systems often suffer from two critical issues: extremely limited labeled samples for newly emerging emitters and heterogeneous data distributions collected by geographically distributed receivers with mismatched label spaces. To address these challenges, this paper proposes a heterogeneous federated learning (HFL)-based framework for few-shot specific emitter identification (FS-SEI). The proposed framework decouples feature embedding learning from task-specific classification and enables collaborative representation learning across distributed receivers without sharing raw signal data. A metric learning-based training strategy is adopted, where only the feature embedding models are aggregated in the federated process, effectively alleviating the impact of label space mismatch by utilizing center loss and an improved triplet loss. Moreover, two federated optimization schemes, namely gradient averaging (GA) and model averaging (MA), are systematically investigated to analyze their effectiveness under fully heterogeneous settings. Extensive experiments conducted on a real-world dataset demonstrate that the proposed HFL framework significantly outperforms isolated local training. In particular, the GA-based scheme achieves a few-shot identification performance that closely approaches centralized learning while preserving data privacy and robustness against data heterogeneity. The results validate the effectiveness of the proposed approach for practical FS-SEI in low-altitude drone management scenarios. Full article

Cross-domain few-shot hyperspectral image (HSI) classification is challenging due to limited labeled samples and distribution shifts across sensors and acquisition scenes, which often degrade feature representation and classification performance. This study proposes a lightweight hierarchical CNN–Transformer framework, termed LGDAF-Net (Lightweight Global and Local Dual Attention Fusion Network), for effective cross-domain few-shot HSI classification. The framework progressively enhances spectral–spatial representation through three stages: spectral–spatial feature recalibration, local spatial structure perception, and global contextual modeling. Specifically, a spectral–spatial dual-attention enhancement module (SESA) is introduced to emphasize informative spectral responses and suppress redundancy. A Local Attention Spatial Perception Module (LASPM) is designed to capture fine-grained spatial structures, while a lightweight Transformer-based Global Attention Context Modeling Module (GACM) models long-range spatial dependencies. In addition, kernel triplet loss and domain adversarial learning are incorporated to improve feature discrimination and promote cross-domain feature alignment. Experimental results on three benchmark datasets demonstrate that the proposed method achieves competitive performance compared with existing methods. Full article

Background/Objectives: Fragile X Syndrome (FXS) is a neurodevelopmental disorder caused by a triplet repeat expansion in the Fmr1 gene leading to the loss of Fragile X Messenger Ribonucleoprotein (Fmr1 protein). The loss of Fmr1 protein modulates many cell biological processes and leads to the emergence of intellectual disability and autism. FXS is modeled in Fmr1-KO mice that display features consistent with human FXS, including hypersensitivity, cognitive and learning deficits, hyperactivity and audiogenic seizures. Here, we investigated the effect of auditory stimulation during a range of developmental stages on recognition memory and sociability deficits in Fmr1-KO mice. Methods: Fmr1-KO mice were subjected to auditory stimulation for 2 min three times a day at one-hour intervals for 5 days at the nursing, juvenile and adult stages. The animals were tested for social interaction and novel object recognition at 2 to 3 months old. Results: During auditory stimulation, the wild running phenotype was observed in the Fmr1-KO juvenile animals and two animals at the nursing stage experienced status epilepticus and died. Fmr1-KO animals showed social deficits compared to both the control and animals exposed to auditory stimulation at the juvenile stage. In the novel object recognition task, auditory stimulation was more effective at the nursing and juvenile stages. Conclusions: These data show that auditory stimulation may be an effective way to restore cognitive and social deficits in FXS. Full article

This paper proposes a novel method for measuring music similarity. Existing music similarity measurements have often been used for music appreciation, but this paper proposes a method for measuring the similarity between music samples which are used for music production. Conventional music recommendation approaches often rely on either metadata-based similarity or audio-based feature similarity in isolation, which limits their effectiveness in sample-based recommendation scenarios where both compositional context and acoustic characteristics are important. To address this limitation, the proposed framework combines a hypergraph-based information similarity module with a feature-based similarity module learned using Siamese networks and triplet loss. In the information-based module, metadata attributes such as beats per minute (BPM), genre, chord, key, and instrument are modeled as vertices in a hypergraph, and Random Walk–Word2Vec embeddings are learned to capture structural relationships between music samples and their attributes. In parallel, the feature-based module employs vertex-specific Siamese networks trained on instrument and key classification tasks to learn perceptual similarity directly from audio signals. The two modules are trained independently and jointly utilized at the recommendation stage to provide attribute-specific similarity results for a given query sample. Results show that the proposed system achieves high Precision@k across multiple attributes and forms stable similarity structures in the embedding space, even without relying on user interaction data. These results reflect embedding consistency evaluated over the entire dataset where training and retrieval are performed on the same sample pool, rather than generalization to unseen samples. These results demonstrate that the proposed hybrid framework effectively captures both structural and perceptual similarity among music samples and is well suited for sample-based music recommendation in music production environments. Full article

As one of the world’s leading cotton-producing countries, China frequently experiences severe yield reductions due to crop diseases and pest infestations, with losses often exceeding 20%. Although computer vision models can identify diseased plants, they currently fail to connect visual symptoms to the diagnostic reasoning process used by agronomists. This leads to text descriptions that ignore the biological causes of the damage. To fix this, we built Multimodal Knowledge Graph-Enhanced Cross Vision Transformer-18-Dagger-408 and Text-to-Text Transfer Transformer for Cotton Disease and Pest Image Captioning (MKG-CottonCapT6), a model that uses a local knowledge database to generate professional diagnostic reports from field images. The technical core consists of a Multimodal Knowledge Graph (MMKG) containing 14 types of entities (such as Pathogens and Control Agents) and 12 types of relations. We use a Cross Vision-Transformer-18-Dagger-408 (CrossViT) encoder to capture both the overall leaf shape and microscopic details of pests. Through a Visual Entity Grounding (VEG) module, the model maps visual features directly to specific triplets in the graph. These triplets are then turned into text sequences and fused with image data in a Text-to-Text-Transfer-Transformer (T5) decoder. To train the model, we collected a dataset of cotton images paired with expert descriptions of lesions, colors, and affected plant parts. Tests show that MKG-CottonCapT6 performs better than standard models, reaching an Information-based Metric for Image Captioning (InfoMetIC) score of 72.6%. Results prove that by using a specific alignment loss (${\mathcal{L}}_{align}$), the model generates reports that correctly name the disease stage and recommend specific chemicals, such as Carbendazim or Triadimefon. This framework provides a practical tool for farmers to record and treat cotton diseases with high precision. Full article