Search Results (801)

Event cameras provide microsecond temporal resolution, high dynamic range, and low latency by asynchronously capturing per-pixel luminance changes, thereby introducing a novel sensing paradigm. These advantages render them well-suited for high-speed applications such as autonomous vehicles and dynamic environments. Nevertheless, the sparsity of event data and the absence of dense annotations are significant obstacles to supervised learning for motion segmentation from event streams. Domain adaptation is also challenging due to the considerable domain shift in intensity images. To address these challenges, we propose a two-phase cross-modality adaptation framework that translates motion segmentation knowledge from labeled RGB-flow data to unlabeled event streams. A dual-branch encoder extracts modality-specific motion and appearance features from RGB and optical flow in the source domain. Using reconstruction networks, event voxel grids are converted into pseudo-image and pseudo-flow modalities in the target domain. These modalities are subsequently re-encoded using frozen RGB-trained encoders. Multi-level consistency losses are implemented on features, predictions, and outputs to enforce domain alignment. Our design enables the model to acquire domain-invariant, semantically rich features through the use of shallow architectures, thereby reducing training costs and facilitating real-time inference with a lightweight prediction path. The proposed architecture, alongside the utilized hybrid loss function, effectively bridges the domain and modality gap. We evaluate our method on two challenging benchmarks: EVIMO2, which incorporates real-world dynamics, high-speed motion, illumination variation, and multiple independently moving objects; and MOD++, which features complex object dynamics, collisions, and dense 1kHz supervision in synthetic scenes. The proposed UDA framework achieves 83.1% and 79.4% accuracy on EVIMO2 and MOD++, respectively, outperforming existing state-of-the-art approaches, such as EV-Transfer and SHOT, by up to 3.6%. Additionally, it is lighter and faster and also delivers enhanced mIoU and F1 Score. Full article

Hand Gesture Recognition (HGR) is a vital technology that enables intuitive human–computer interaction in various domains, including augmented reality, smart environments, and assistive systems. Achieving both high accuracy and real-time performance remains challenging due to the complexity of hand dynamics, individual morphological variations, and computational limitations. This paper presents a lightweight and efficient skeleton-based HGR framework that addresses these challenges through an optimized multi-stream Convolutional Neural Network (CNN) architecture and a trainable ensemble tuner. Dynamic 3D gestures are transformed into structured, noise-minimized 2D spatiotemporal representations via enhanced data-level fusion, supporting robust classification across diverse spatial perspectives. The ensemble tuner strengthens semantic relationships between streams and improves recognition accuracy. Unlike existing solutions that rely on high-end hardware, the proposed framework achieves real-time inference on consumer-grade devices without compromising accuracy. Experimental validation across five benchmark datasets (SHREC2017, DHG1428, FPHA, LMDHG, and CNR) confirms consistent or superior performance with reduced computational overhead. Additional validation on the SBU Kinect Interaction Dataset highlights generalization potential for broader Human Action Recognition (HAR) tasks. This advancement bridges the gap between efficiency and accuracy, supporting scalable deployment in AR/VR, mobile computing, interactive gaming, and resource-constrained environments. Full article

Accurate heavy-duty vehicle load estimation is crucial for transportation and environmental regulation, yet current methods lack precision in data accuracy and practicality for field implementation. We propose a Self-Supervised Reconstruction Heterogeneous Graph Convolutional Network (SSR-HGCN) for load estimation using On-Board Diagnostics (OBD) data. The method integrates physics-constrained heterogeneous graph construction based on vehicle speed, acceleration, and engine parameters, leveraging graph neural networks’ information propagation mechanisms and self-supervised learning’s adaptability to low-quality data. The method comprises three modules: (1) a physics-constrained heterogeneous graph structure that, guided by the symmetry (invariance) of physical laws, introduces a structural asymmetry by treating kinematic and dynamic features as distinct node types to enhance model interpretability; (2) a self-supervised reconstruction module that learns robust representations from noisy OBD streams without extensive labeling, improving adaptability to data quality variations; and (3) a multi-layer feature extraction architecture combining graph convolutional networks (GCNs) and graph attention networks (GATs) for hierarchical feature aggregation. On a test set of 800 heavy-duty vehicle trips, SSR-HGCN demonstrated superior performance over key baseline models. Compared with the classical time-series model LSTM, it achieved average improvements of 20.76% in RMSE and 41.23% in MAPE. It also outperformed the standard graph model GraphSAGE, reducing RMSE by 21.98% and MAPE by 7.15%, ultimately achieving < 15% error for over 90% of test samples. This method provides an effective technical solution for heavy-duty vehicle load monitoring, with immediate applications in fleet supervision, overloading detection, and regulatory enforcement for environmental compliance. Full article

To better understand the relationship between meniscus instabilities and the high levels of turbulence in the fluid dynamics of a continuous medium slab mold, this study investigates the magnitudes of meniscus dynamics distortions and their fluid dynamic origin using a full-scale water modeling experiment and mathematical simulations. The three-dimensional mathematical model is composed of the continuity and momentum equations, together with the standard k-ε turbulence model and the volume of fluid model, to track the dynamics of the steel interface. The results show that the medium slab mold shares flow patterns common to both conventional slab molds and funnel thin slab molds, making its fluid dynamics more complex. Despite this, the fluid dynamics within the mold do not develop a dynamic distortion phenomenon but induce upward stream flows with instability and high velocities, which generate an unstable meniscus behavior characterized by significant surface oscillations, variations in velocity, and high deformations. These latter flow characteristics are the origin of meniscus dynamic distortion (MDD), which shows a constant frequency with non-constant periodicity and different median lifecycle ranges. Full article

Unmanned aerial vehicle (UAV) remote sensing is critical in assessing post-earthquake building damage. However, intelligent disaster assessment via remote sensing faces formidable challenges from complex backgrounds, substantial scale variations in targets, and diverse spatial disaster dynamics. To address these issues, we propose UEBNet, a high-precision post-earthquake building instance segmentation model that systematically enhances damage recognition by integrating three key modules. Firstly, the Depthwise Separable Convolutional Block Attention Module suppresses background noise that visually resembles damaged structures. This is achieved by expanding the receptive field using multi-scale pooling and dilated convolutions. Secondly, the Multi-feature Fusion Module generates scale-robust feature representations for damaged buildings with significant size differences by processing feature streams from different receptive fields in parallel. Finally, the Adaptive Multi-Scale Interaction Module accurately reconstructs the irregular contours of damaged buildings through an advanced feature alignment mechanism. Extensive experiments were conducted using UAV imagery collected after the Ms 6.8 earthquake in Tingri County, Tibet Autonomous Region, China, on 7 January 2025, and the Ms 6.2 earthquake in Jishishan County, Gansu Province, China, on 18 December 2023. Results indicate that UEBNet enhances segmentation mean Average Precision (${mAP}^{seg}$) and bounding box mean Average Precision (${mAP}^{box}$) by 3.09% and 2.20%, respectively, with equivalent improvements of 2.65% in F1-score and 1.54% in overall accuracy, outperforming state-of-the-art instance segmentation models. These results demonstrate the effectiveness and reliability of UEBNet in accurately segmenting earthquake-damaged buildings in complex post-disaster scenarios, offering valuable support for emergency response and disaster relief. Full article

Thise article examines the reliable online detection and IDS/IPS intrusion prevention in dynamic corporate networks problems, where traditional signature-based methods fail to keep pace with new and evolving attacks, and streaming data is susceptible to drift and targeted “poisoning” of the training dataset. As a solution, we propose a hybrid neural network system with adaptive online training, a formal minimax false-positive control framework, and a robustness mechanism set (a Huber model, pruned learning rate, DRO, a gradient-norm regularizer, and a prioritized replay). In practice, the system combines modal encoders for traffic, logs, and metrics; a temporal GNN for entity correlation; a variational module for uncertainty assessment; a differentiable symbolic unit for logical rules; an RL agent for incident prioritization; and an NLG module for explanations and the preparation of forensically relevant artifacts. In this case, the applied components are connected via a cognitive layer (cross-modal fusion memory), a Bayesian-neural network fuser, and a single multi-task loss function. The practical implementation includes the pipeline “novelty detection → active labelling → incremental supervised update” and chain-of-custody mechanisms for evidential fitness. A significant improvement in quality has been experimentally demonstrated, since the developed system achieves an ROC AUC of 0.96, an F1-score of 0.95, and a significantly lower FPR compared to basic architectures (MLP, CNN, and LSTM). In applied validation tasks, detection rates of ≈92–94% and resistance to distribution drift are noted. Full article

Space-based infrared (IR) detection, with wide coverage, all-time operation, and stealth, is crucial for aerial target surveillance. Under low signal-to-noise ratio (SNR) conditions, however, its small target size, limited features, and strong clutters often lead to missed detections and false alarms, reducing stability and real-time performance. To overcome these issues of energy-integration imaging in perceiving dim targets, this paper proposes a biomimetic vision-inspired Infrared Temporal Differential Detection (ITDD) method. The ITDD method generates sparse event streams by triggering pixel-level radiation variations and establishes an irradiance-based sensitivity model with optimized threshold voltage, spectral bands, and optical aperture parameters. IR sequences are converted into differential event streams with inherent noise, upon which a lightweight multi-modal fusion detection network is developed. Simulation experiments demonstrate that ITDD reduces data volume by three orders of magnitude and improves the SNR by 4.21 times. On the SITP-QLEF dataset, the network achieves a detection rate of 99.31%, and a false alarm rate of $1.97\times {10}^{-5}$, confirming its effectiveness and application potential under complex backgrounds. As the current findings are based on simulated data, future work will focus on building an ITDD demonstration system to validate the approach with real-world IR measurements. Full article

The new type of gold cone filter cartridge has dual functions of increasing filter area and enhancing pulsed-jet cleaning, but the issue of patchy cleaning remains to be addressed. This study further enhances the pulsed-jet cleaning performance of cone filter cartridges by employing a porous diffusion nozzle. The temporal and spatial distributions of pulse jet velocity and pressure under the condition of porous nozzles were investigated through numerical modeling. The variation law of pressure on the side wall of the filter cartridge was analyzed. The influence of jet distance of porous nozzles on pulsed-jet pressure and pulsed-jet uniformity was experimentally investigated. Dust filtration and cleaning experiments were conducted, and the filtration pressure drop, dust emission concentration, and comprehensive filtration performance were compared. It was found that the airflow jetted by the porous diffusion nozzle is more divergent than that of the common round nozzle. This results in a larger entrainment of the jet stream, a milder collision of the jet stream with the cartridge cone, and a slower overall velocity reduction. More airflow is generated into the filter cartridge and accumulated; the accumulated static pressure covers a larger range of the upper section of the filter cartridge, with a longer duration of static pressure. In the online dust filtration and cleaning experiment, compared with the condition of the common round nozzle, the porous nozzle can reduce the residual pressure drop by 27.0%, increase the filtration cleaning interval by a factor of 3.80, reduce the average dust emission concentration by 45.2%, and increase the comprehensive performance index Q_F by 5.2%. The research conclusions can provide references for the design and optimization of industrial filter cartridge dust collectors. Full article

This study investigates the species composition and distribution of blackflies (Diptera: Simuliidae) in Kazakhstan, with a focus on two species newly recorded for the country: Simulium murmanum (Enderlein, 1935) and Simulium reptans (Linnaeus, 1758). The presence of S. murmanum in Kazakhstan is reported for the first time, supported by morphological and molecular genetic analyses. Diagnostic features of the larva, pupa, and adult stages are described in detail, including the structure and coloration of the larval head capsule, pupal cocoon, and genitalia of both sexes. Habitat preferences and pupal substrate attachment patterns are illustrated, with observations on variations in cocoon branching across different flow regimes. Species identification was conducted using the morphological keys of Rubtsov and Yankovsky, and taxonomic classification was confirmed using the framework proposed by Adler. Molecular confirmation of S. murmanum was performed via DNA analysis. The species was found to be restricted to the foothill regions of East Kazakhstan, suggesting a distribution closely associated with the Altai mountain systems and adjacent regions in Mongolia and China. Unlike its status as a dominant hematophagous species in parts of Russia, S. murmanum has not demonstrated biting activity in Kazakhstan, Mongolia, or China. Additionally, the study provides the first records of S. reptans within the fauna of Kazakhstan, initially identified in the Irtysh River (Pavlodar Region). Subsequent sampling conducted in June 2024 revealed a continuous distribution of S. reptans along the Irtysh River through to the mountain streams of East Kazakhstan. The species was found in mountainous, foothill, and lowland environments, highlighting its wide ecological plasticity. Full article

With the proliferation of the Internet of Things (IoT), gesture recognition has attracted attention as a core technology in human–computer interaction (HCI). In particular, mmWave frequency-modulated continuous-wave (FMCW) radar has emerged as an alternative to vision-based approaches due to its robustness to illumination changes and advantages in privacy. However, in real-world human–machine interface (HMI) environments, hand gestures are inevitably accompanied by torso- and arm-related reflections, which can also contain gesture-relevant variations. To effectively capture these variations without discarding them, we propose a preprocessing method called Adaptive Top-K Selection, which leverages vector entropy to summarize and preserve informative signals from both hand and body reflections. In addition, we present a Multi-Stream EfficientNetV2 architecture that jointly exploits temporal range and Doppler trajectories, together with radar-specific data augmentation and a training optimization strategy. In experiments on the publicly available FMCW gesture dataset released by the Karlsruhe Institute of Technology, the proposed method achieved an average accuracy of 99.5%. These results show that the proposed approach enables accurate and reliable gesture recognition even in realistic HMI environments with co-existing body reflections. Full article

The European eel (Anguilla anguilla), a catadromous species currently listed as Critically Endangered by the IUCN, is undergoing a severe continental decline. Among the multiple contributing factors, chemical contamination of aquatic environments—particularly by heavy metals—plays a major role. This study analyzed the concentrations of 16 trace elements in the muscle tissue of A. anguilla specimens collected from three ecologically distinct sites in Southern Italy: an estuary (Foce del Crati), a lagoon (Laghi di Gizzeria) and a stream (torrente Raganello). Correlations between trace element accumulation and the onset of oxidative stress were also examined. To assess eel health status, oxidative biomarkers were also analyzed in heart, liver, and gill tissues. Statistical analysis among populations revealed significant differences in the bioaccumulation of 10 of the 16 elements, with Cd and As being of particular concern. No significant correlations were found between these two elements and oxidative biomarkers, but Spearman analysis identified both positive and negative correlations with other elements varying by the site of collection. Oxidative biomarkers also showed site- and tissue-specific variation. In particular, SOD activity was highest in the liver and varied across sites; LPO and protein carbonyl levels were generally lower in eels from the Crati River, although heart values deviated from this trend, highlighting tissue-specific response patterns. These results underscore the complex interplay between chemical contamination and the physiology of the European eel, emphasizing the influence of environmental context in modulating tissue-specific oxidative responses. Full article

To address the challenges of low detection accuracy and weak generalization in UAV aerial imagery caused by complex ground environments, significant scale variations among targets, dense small objects, and background interference, this paper proposes an improved lightweight multi-scale small-object detection model, MBD-YOLO (MBFF module, BiMS-FPN, and Dual-Stream Head). Specifically, to enhance multi-scale feature extraction capabilities, we introduce the Multi-Branch Feature Fusion (MBFF) module, which dynamically adjusts receptive fields through parallel branches and adaptive depthwise convolutions, expanding the receptive field while preserving detail perception. We further design a lightweight Bidirectional Multi-Scale Feature Aggregation Pyramid Network (BiMS-FPN), integrating bidirectional propagation paths and a Multi-Scale Feature Aggregation (MSFA) module to mitigate feature spatial misalignment and improve small-target detection. Additionally, the Dual-Stream Head with NMS-free architecture leverages a task-aligned architecture and dynamic matching strategies to boost inference speed without compromising accuracy. Experiments on the VisDrone2019 dataset demonstrate that MBD-YOLO-n surpasses YOLOv8n by 6.3% in mAP50 and 8.2% in mAP50–95, with accuracy gains of 17.96–55.56% for several small-target categories, while increasing parameters by merely 3.1%. Moreover, MBD-YOLO-s achieves superior detection accuracy, efficiency, and generalization with only 12.1 million parameters, outperforming state-of-the-art models and proving suitable for resource-constrained embedded deployment scenarios. The superior performance of MBD-YOLO, which harmonizes high precision with low computational demand, fulfills the critical requirements for real-time deployment on resource-limited UAVs, showing great promise for applications in traffic monitoring, urban security, and agricultural surveying. Full article

Urbanization, characterized by rapid construction land expansion, has transformed natural landscapes and significantly altered river networks in emerging metropolitan areas. Understanding the historical and current conditions of river networks is crucial for policy-making in sustainable urban development planning. Based on the topographic maps and remote sensing images, this study employs a multi-metric framework to investigate river network variations in the Suzhou-Wuxi-Changzhou metropolitan area, a rapidly urbanized plain with high-density river networks in the Yangtze River Delta, China. The results indicate a significant decline in the quantity of rivers, with the average river density in built-up areas falling from 2.70 km·km⁻² in the 1960s to 1.95 km·km⁻² in the 2010s, along with notable variations in the river network’s structure, complexity and its storage and regulation capacity. Moreover, shifts in the structural characteristics of river networks reveal that urbanization has a weaker impact on main streams but plays a dominant role in altering tributaries. The analysis demonstrates the extensive burial and modification of rivers across the metropolitan plains. These findings underscore the essence of incorporating river network protection and restoration into sustainable urban planning, providing insights for water resource management and resilient city development in rapidly urbanizing regions. Full article

Fish secondary production integrates multiple demographic parameters, including population density, growth, mortality, and recruitment, and thereby provides a comprehensive measure of ecological performance. It is also a valuable tool for assessing the ecological integrity of stream ecosystems and the responses of fish populations to habitat alteration, climatic variability, and anthropogenic pressures. Despite its relevance, empirical estimates of fish production remain limited due to methodological constraints. In this study, we quantified secondary production and production-to-biomass (P/B) ratios for Mediterranean brown trout (Salmo trutta complex) across six headwater stream reaches in the northeastern Iberian Peninsula, characterized by contrasting hydrological regimes, channel morphology, and water chemistry. Field sampling was conducted over two consecutive annual cycles (2008/2009 and 2009/2010) at all sites, with extended monitoring at two reaches until 2017 to assess long-term variability. Annual trout production, over the two consecutive annual cycles, ranged from 30.9 to 167.8 kg ha⁻¹ year⁻¹ (mean = 82.2 kg ha⁻¹ year⁻¹), and mean P/B ratios ranged from 0.61 to 1.13 (mean = 0.80). These values fall within the intermediate range reported for brown trout globally and reflect the constrained energy dynamics of Mediterranean streams. Higher production was generally associated with strong age-1 recruitment, elevated standing biomass, and greater water alkalinity. Long-term analyses revealed that interannual variation in trout production was significantly correlated with discharge variability, with higher production occurring under more stable flow conditions. However, in addition to flow variability other factors, such as habitat complexity, may modulate local productivity. Consequently, interannual fluctuations at the long-term sites revealed substantial demographic variability influenced by site-specific environmental conditions. These findings offer reference baselines for Mediterranean trout populations and contribute to the ecological basis for their conservation and management. Full article

The increase in Internet of Things (IoT) devices has introduced significant security challenges, primarily due to their inherent constraints in computational power, memory, and energy. This study provides a comparative performance analysis of selected modern cryptographic algorithms on a resource-constrained IoT platform, the Nordic Thingy:53. We evaluated a set of ciphers including the NIST lightweight standard ASCON, eSTREAM finalists Salsa20, Rabbit, Sosemanuk, HC-256, and the extended-nonce variant XChaCha20. Using a dual test-bench methodology, we measured energy consumption and performance under two distinct scenarios: a low-data-rate Bluetooth mesh network and a high-throughput bulk data transfer. The results reveal significant performance variations among the algorithms. In high-throughput tests, ciphers like XChaCha20, Salsa20, and ASCON32 demonstrated superior speed, while HC-256 proved impractically slow for large payloads. The Bluetooth mesh experiments quantified the direct relationship between network activity and power draw, underscoring the critical impact of cryptographic choice on battery life. These findings offer an empirical basis for selecting appropriate cryptographic solutions that balance security, energy efficiency, and performance requirements for real-world IoT applications. Full article