Search Results (5,532)

The integration of automated learning and video analysis enables the development of intelligent systems that can operate effectively in uncertain scenarios. These systems can autonomously identify dominant motion dynamics, depending on the theoretical framework used for representation and the learning process used for pattern identification. Current literature offers a state-based approach to describe the key temporal and spatial relationships required to understand motion dynamics. An important aspect of this approach is determining when the number of positively learned rules from a given information source is sufficient to detect dominant motion in automatic surveillance scenarios. This is crucial, as it affects both the variability of movements that monitored subjects can exhibit within the camera’s field of view and the resources needed for effective implementation. This study addresses these gaps through a grammar-based sufficiency criterion, which posits that learning is complete when production rule growth stabilizes, under the assumption of system stationarity. The stability criterion evaluates whether the most probable rules are learned over time, and whenever a high-growth rule is added, it is used to update the criterion. We outline several benefits of having a formal criterion for determining when a symbolic surveillance system has a robust model that explains the observed motion dynamics. Our hypothesis is that a correct model can consistently account for the majority of motion dynamics over time in an automated learning process. The proposed approach is evaluated by modeling motion dynamics in several scenarios using the SEQUITUR algorithm as input and computing the probability of stability along the learning curve, which indicates when the model reaches a steady state of consistent learning. Experimental validation was conducted in real-world scenarios under varying acquisition conditions. The results show that the proposed method achieves robust modeling performance, with accuracy values ranging from 83.56% to 95.92% in dynamic environments. Full article

Background and Objectives: Carotid artery stenosis has traditionally guided therapeutic decision-making; however, plaque morphology and composition are increasingly recognized as more reliable indicators of cerebrovascular risk than luminal narrowing alone. As imaging strategies shift toward vulnerability-based assessment, reproducibility of plaque characterization becomes essential for consistent clinical decision-making. This study aimed to evaluate interobserver agreement in carotid plaque assessment using multidetector computed tomography angiography (MDCTA) and to assess intermodality agreement with duplex ultrasonography (DUS). Materials and Methods: In this single-center study (January–September 2022), 50 patients with ≥60% internal carotid artery stenosis diagnosed by DUS (NASCET criteria), the majority of whom were asymptomatic (90%), were included. MDCTA examinations were independently analyzed by two radiologists, while DUS examinations were evaluated by a third observer. Plaque composition (lipid, fibrous, calcified), surface characteristics (regular, irregular, ulcerated), degree of stenosis, and plaque length were assessed. CT plaque characterization was based on Hounsfield unit (HU) thresholds (<50 HU lipid; 50–120 HU fibrous; >120 HU calcified). Interobserver agreement and intermodality agreement were calculated using Cohen’s kappa coefficient. Results: Good interobserver agreement was observed between the two MDCTA readers (κ = 0.751). Intermodality agreement between MDCTA and DUS was moderate (κ = 0.624 and κ = 0.595). Although significant differences were identified in 3 of 16 HU measurement points, no significant differences were found in overall plaque composition classification between MDCTA observers. DUS yielded significantly higher stenosis values (p = 0.007 and p = 0.005) and greater plaque length measurements (p < 0.0005) compared with MDCTA. Significant differences were also observed in plaque surface assessment between modalities (p = 0.044 and p = 0.033). Conclusions: MDCTA demonstrates good interobserver reproducibility for carotid plaque characterization, while intermodality agreement between MDCTA and DUS is moderate. Minor attenuation measurement differences do not significantly affect plaque classification; however, systematic intermodality differences in stenosis grading, plaque surface evaluation, and plaque length measurement should be considered in clinical decision-making. Full article

Cyber-physical systems (CPSs) represent sophisticated intelligent architectures that tightly couple computational elements, communication networks, and physical processes. Their deployments now span virtually every industrial and civilian domain—from power grids and manufacturing plants to autonomous transportation networks. Ensuring the secure operation of CPSs relies fundamentally on effective time series anomaly detection, which remains a challenging task due to the complex, often unknown system dynamics and non-negligible sensor noise present in real-world environments. To address these challenges, we introduce a Neural Network-Enhanced Kalman Filter (NNEKF), a novel anomaly detection framework that combines model-based filtering with data-driven learning. The NNEKF employs a two-stage trained neural network with a specialized architecture: the first stage learns the underlying dynamics of the CPS, while the second stage optimizes the computation of the Kalman gain during the update step. At inference time, the enhanced Kalman filter recursively estimates the likelihood of observed sensor measurements to identify anomalies, supported by a batched parallel inference scheme that delivers substantial speedups. Extensive experiments on benchmark datasets demonstrate that the NNEKF attains an average F1-score of 0.935, coupled with rapid inference and minimal model footprint—surpassing all competitive baselines and facilitating dependable real-time anomaly detection for CPS environments. Full article

The growing need to mitigate climate change has accelerated the development of Carbon Capture, Utilization, and Storage (CCUS) technologies, where the safe transport of supercritical CO₂ (sCO₂) through pipelines is a key challenge. The flow behavior in such systems is strongly influenced by phase-change processes under transient conditions such as decompression and heat transfer and is further complicated by the presence of impurities (e.g., N₂, CH₄, and Ar). These impurities modify thermodynamic properties and phase boundaries, thereby affecting the overall flow dynamics. In this study, the influence of impurities on leakage, mass flow rate, and decompression wave propagation in sCO₂ pipelines is investigated using computational fluid dynamics (CFD) simulations. A real-fluid model (RFM) implemented in the CONVERGE CFD solver is employed, with a tabulation-based approach to accurately capture thermodynamic and transport properties across multiphase regimes. The simulations were validated against available experimental data and performed for varying impurity concentrations to assess their impact on key flow variables, including pressure, temperature, and wave speed. Although simplifying assumptions were used, the results are in fairly good agreement with experimental observations and provide a better understanding of the phase behavior induced by impurities during transient decompression. Additionally, the effects of outlet geometry, pipeline configuration, and the choice of equation of state are examined, highlighting their influence on the predicted flow response. The validity of the RFM modeling framework is further demonstrated by simulations of a large-scale pipeline configuration representative of industrial conditions, which will serve as a benchmark for future improvements. Full article

Background: Efficient parking navigation in large and dynamic parking areas requires systems that can adapt to real-time conditions and provide precise vehicle localization. Methods: This paper presents a smart car parking navigation module that integrates camera-based vehicle perception, homography-based ground-plane localization, mobile GNSS positioning, and dynamic route planning into a unified framework. Instance segmentation (YOLOv8n-seg) is used to detect vehicles and extract ground-contact regions, which are associated with parking slots defined in a GeoJSON-based site model. Mobile GNSS data are fused with visual observations via spatio-temporal proximity scoring to enable robust user–vehicle matching without optical identification. An A* routing algorithm dynamically computes and updates navigation paths, adapting to lane obstructions and slot availability in real time. Results: Experimental evaluation on a real six-camera parking facility shows that the proposed segmentation-based localization reduces mean error from 0.732 m to 0.283 m (61.3% improvement), with the 95th-percentile error dropping from 1.892 m to 0.908 m, and outperforming the bounding-box baseline in 85.3% of detections. Conclusions: These results demonstrate that sub-meter vehicle localization and reliable user–vehicle association are achievable using standard surveillance cameras without specialized infrastructure, offering a scalable and cost-effective solution for intelligent parking navigation. Full article

Position-independent geometric errors crucially impact the accuracy of five-axis machine tools, yet their identification remains challenging due to computational complexities, inadequate measurement pose selection, and disturbances arising from thermal drift and residual uncompensated errors. Existing methods typically rely on linearized kinematic models, heuristic sampling of measurement poses, or computationally expensive global optimization procedures, which collectively limit their effectiveness in industrial environments. This study presents a unified identification framework that overcomes these limitations; it incorporates 3D offset parameters to enhance the decoupling of true geometric errors from non-PIGEs, an observability-driven measurement pose selection strategy to maximize the parameter sensitivity, and an ANN-surrogate model to accelerate high-dimensional global optimization. A genetic algorithm is used to optimize the measurement points based on the observability index of the machine tool. The ANN-surrogate model enhances the identification accuracy of error parameters (11 PIGEs + 3 offsets) through precise kinematic models, global exploration, and final refinement. Experimental validation on a five-axis machine tool demonstrates a volumetric error reduction of 88.615% after compensation, with RMSE decreasing to 0.4337 μm. Sensitivity analysis reveals that PIGEs contribute up to 75.26% of the total inaccuracy, while offset parameters capture 24.74% of the error from thermal and non-PIGE sources. The results confirm the method’s superiority over other techniques in terms of identification accuracy, efficiency, and robustness, providing a practical solution for high-precision applications in the manufacturing industries. Full article

The prediction of dynamic responses of offshore wind turbine foundations under wind-wave-current multi-field coupled loads is the cornerstone of safety in offshore wind power engineering. The currently widely adopted equivalent load application method, while computationally efficient, simplifies loads into concentrated forces applied at the pile top and tower top, neglecting fluid-structure dynamic interaction mechanisms, which leads to deviations in response predictions. To overcome this limitation, this paper proposes a high-precision bidirectional fluid-structure interaction numerical framework. The fluid domain employs computational fluid dynamics (CFD) to construct an air-seawater two-phase flow model, utilizing the standard k-ε turbulence model and nonlinear wave theory to accurately simulate complex marine environments. The solid domain establishes a wind turbine-stratified seabed system via the finite element method (FEM), describing soil-rock mechanical properties based on the Mohr-Coulomb constitutive model. Comparative studies indicate that the equivalent static method significantly underestimates the displacement response of pile foundations, particularly under the extreme shutdown conditions examined in this study. This value should be interpreted as a case-specific observation rather than a universal deviation, and the discrepancy may vary with sea state, wind speed, current velocity, and wind–wave misalignment, thereby leading to non-conservative estimates of stress distribution. In contrast, the fluid-structure interaction method can reveal key physical processes such as local flow acceleration and wake–interference effects around the tower and the parked rotor under shutdown conditions, and the nonlinear interaction and resistance-increasing mechanisms between waves and currents. This model provides a reliable tool for safety assessment and damage evolution analysis of wind turbine foundations under extreme marine conditions, promoting the transformation of offshore wind power structure design from empirical formulas to mechanism-driven approaches. Full article

Pig behavior statistics can reflect their health status. Conventional approaches depend on manual observation to derive behavioral information from video recordings, a process that demands substantial time and human effort. To overcome these limitations in indoor intensive farming environments, this study introduces an effective approach for recognizing pig behaviors, employing an enhanced YOLOv8n architecture. The approach utilizes advanced object detection algorithms to automatically identify pig behaviors, including stand, lie, eat, fight, and tail-bite, from overhead video footage of the enclosure. First, images of daily pig behaviors are collected using cameras to build a pig behavior dataset. To boost detection accuracy, the SE attention mechanism is embedded within the feature extraction backbone of the YOLOv8n network to enhance its representational capacity, strengthening the model’s capacity to grasp overarching contextual information and improve the expressiveness of extracted features. The GIoU loss function is employed during training to reduce computational cost and accelerate model convergence. Moreover, integrating Ghost convolution into the backbone significantly reduces both computational complexity and the total number of parameters. The experimental findings reveal that the optimized YOLOv8n model contains just 1.71 million parameters, marking a 42.93% reduction relative to the baseline model. Its floating-point operations total 5.0 billion, indicating a 38.27% decrease, while the mean average precision (mAP@50) reaches 96.8%, surpassing the original by 2.6 percentage points. Compared with other widely used YOLO-based object detection frameworks, the proposed approach achieves notably higher accuracy while requiring significantly lower computational resources and model complexity. Full article

This paper introduces a new hybrid forecasting architecture that combines Nonlinear Autoregressive (NAR) models, the proxy Standardized Precipitation-Evapotranspiration Index (SPEI), and a Nonlinear Autoregressive with Exogenous Inputs (NARX) framework for Soil Moisture Index (SMI) prediction. The suggested methodology solves the crucial difficulty of combining future climatic knowledge into soil moisture forecasting by using a cascaded approach. Stage 1 uses univariate NAR models to create multi-step-ahead predictions of precipitation and temperature. Stage 2 converts these forecasts into proxy SPEI values using a physically based water balance computation, and Stage 3 employs a NARX model that uses observed historical SMI and forecast-derived proxy SPEI as exogenous inputs. The framework is assessed using high-frequency observations from 2014 to 2020, with training data through 2019 and validation covering the whole 2020 horizon. The study combining forecast-driven climatic indicators with autoregressive soil moisture dynamics resulted in prediction accuracy (R² = 0.9888, RMSE = 0.0827, MAE = 0.0567). This study presents a new NAR–SPEI–NARX model for SMI prediction forecasting, based on three-stage modeling, where NAR models forecast precipitation and temperature and then turn them into SPEI-proxy as an exogenous input for NARX. Full article

Accurate programming of sagittal condylar inclination (SCI) and Bennett angle (BA) is important for prosthodontic treatment, yet evidence directly comparing conventional and digital recording approaches remains limited. This single-participant pilot feasibility study compared SCI and BA obtained using a digital jaw motion tracking system, a conventional facebow transfer method, and a cone-beam computed tomography (CBCT)-based registration method. Ten repeated datasets were generated for each method from one healthy adult participant. The digital system recorded mandibular motion using optical tracking and automatically calculated SCI and BA in a virtual articulator. The conventional method used a mechanical facebow and check-bite records, whereas the CBCT-based method combined one centric-relation CBCT scan with repeated protrusive and lateral interocclusal records after digital alignment. Significant differences were observed for left SCI (p = 0.036), left BA (p = 0.049), and right BA (p < 0.001), whereas right SCI was not significantly different (p = 0.197). The digital method showed the lowest standard deviations across all variables and lower coefficients of variation for left SCI, right SCI, and left BA. Within the limitations of this single-participant pilot study, digital jaw motion tracking demonstrated favorable repeatability and clinically comparable measurements, supporting its potential utility in digitally integrated prosthodontic workflows. Full article

As a widely used computing substrate, the side-channel security of FPGAs has attracted considerable attention, yet a systematic understanding of how FPGA device types contribute to exploitable leakage remains limited. This work presents a device-centric evaluation that maps an S-box-like function onto common FPGA primitives, including look-up table (LUT), flip-flop (FF), block RAM (BRAM), and distributed RAM (LUTRAM), and assesses Correlation Power Analysis (CPA) outcomes under the Hamming Weight (HW) and Hamming Distance (HD) power models. The results show pronounced leakage differences across device types: FF- and BRAM-based implementations exhibit substantially stronger leakage than LUT- and LUTRAM-based ones, and they frequently achieve $\mathrm{GE}=0$ in our configurations, while the HD model is generally more effective than the HW model in the performed CPA evaluations. Notably, FF-, BRAM-, and LUTRAM-based implementations can already be breakable starting from one instance under the HD model in our device-level tests, indicating that exploitable leakage may manifest in real FPGA applications. These device-level observations are further validated on a practical cipher by analyzing two SM4 encryption modules that differ only in the S-box implementation style; the BRAM-based design shows significantly stronger leakage than the LUT-based design, achieving $\mathrm{GE}=2.58$ versus $\mathrm{GE}=78.3$ at 10,000 traces. This work highlights the critical role of device selection and implementation style in FPGA side-channel security, and it provides practical insights for designing secure FPGA applications against power side-channel analysis. Full article

Accurate and stable extrinsic calibration is the foundation of high-quality fusion sensing and positioning of camera and Light Detection and Ranging (LiDAR). However, traditional targetless calibration methods suffer from limitations such as poor scene adaptability and unstable convergence, which significantly restrict calibration accuracy and robustness in complex environments. Aiming at solving those problems, we propose an online cascade-optimization-based extrinsic calibration method of combining motion trajectory alignment and edge feature alignment. In the initial calibration stage, a hand–eye calibration algorithm is designed by minimizing the residual discrepancies between camera odometry and LiDAR odometry sequences. It establishes a robust initialization for subsequent optimization. Then, in order to extract robust edge line features from sparse point clouds, we employ depth difference and planar edges of point clouds in the optimization process. Subsequently, principal component analysis (PCA) is applied to compute the principal direction of the extracted line features, enabling a decoupled optimization scheme that accounts for directional observability. This approach effectively mitigates the adverse effects of uneven environmental feature distributions. Experimental validation on typical urban datasets demonstrates the method’s generalizability and competitive accuracy: rotational parameter errors are constrained within 0.25°, and translational errors are maintained below 0.05 m. This affirms the method’s suitability for high-accuracy engineering applications. Full article

Mathematics is a diverse discipline that requires a variety of cognitive abilities and presents varying levels of difficulty. Understanding how different cognitive profiles relate to specific patterns of mathematical performance is important for developing effective educational interventions. This study extends our previous research, in which we identified subgroups of children with mathematical difficulties based on their cognitive abilities. We examined 146 Israeli elementary school children in grades 3 and 4, classified into four subgroups: Reading Accuracy Difficulties (RAD), Mild Mathematical Difficulties (MMD), Non-Verbal Reasoning Difficulties (NVRD), and Typically Developing children (TD). Participants were assessed on arithmetic facts, computational fluency, procedural skills, estimation, and numeration. We observed varied performance patterns among subgroups. The RAD group showed the most severe impairments across all mathematical domains, along with reading comorbidity and cognitive difficulties. The MMD group, which maintained intact cognitive skills, faced notable challenges in computation, performing significantly below the TD group but better than the RAD group. The NVRD group, despite limitations in nonverbal reasoning, outperformed other difficulty groups on fact retrieval and estimation. Performance on multiplication and division tasks consistently followed a hierarchical pattern across all difficulty groups, with the RAD group facing the greatest challenges. These findings demonstrate that mathematical difficulties vary across cognitive profiles and that distinguishing between profiles through targeted assessment enables the development of differentiated interventions tailored to each learner’s specific cognitive profile. Full article

Accurate brain tumor segmentation from magnetic resonance imaging (MRI) is crucial for brain tumor diagnosis, clinical treatment decisions, and advancing research. CNNs and Transformers have dominated this area, but CNNs struggle with long-range modeling, whereas Transformers are limited by the high computational costs of self-attention. Recently, Mamba has garnered significant attention due to its remarkable performance in long sequence modeling. However, the original Mamba architecture, designed primarily for 1D sequence modeling, fails to effectively capture the spatial and structural relationships essential for brain tumor segmentation. In this paper, we propose FocusMamba, a Mamba-based model inspired by human visual observation patterns, which jointly enhances local detail modeling and global contextual understanding. FocusMamba consists of three components: (i) a novel hierarchical and tri-directional Mamba unit that elevates attention from the global to the window level, reinforcing local semantic feature extraction, while simultaneously achieving window-level interactions to maintain broader global awareness, (ii) a large kernel convolution unit that captures long-range dependencies within whole-volume features, overcoming the limitations of Mamba’s single-scale context modeling, and (iii) a fusion unit that enhances the overall feature representation by fusing information from different levels. Extensive experiments on the BraTS 2023 and BraTS 2020 datasets demonstrate that FocusMamba achieves superior segmentation performance compared with several advanced methods. Full article

This study investigated neural oscillatory dynamics underlying visual word recognition in university students with dyslexia using a portable brain–computer interface (BCI) EEG system. The sample included university students with dyslexia (N = 12) and matched controls (N = 14) who completed auditory discrimination and visual word recognition tasks, with and without musical accompaniment. Through these experimental conditions, the researchers assessed (a) the cortical activation across frequency bands, (b) the modulatory effect of background music, and (c) the relationship between emotional states and brain activity. Results revealed significant group differences in oscillatory patterns, with reduced β- and γ-band activity in the left occipito-temporal cortex among participants with dyslexia, confirming disrupted temporal coordination in posterior reading networks. Compensatory right-hemisphere activation was observed, particularly under musical conditions, accompanied by increased α-band power and reduced δ activity, indicating enhanced attentional engagement and reduced cognitive fatigue. Emotional assessment using the DASS-21 revealed higher stress and anxiety scores in the dyslexic group, suggesting that affective factors may modulate oscillatory dynamics. The presence of background music appeared to attenuate these effects, supporting improved emotional regulation and cognitive focus. These findings demonstrate that dyslexia reflects a distributed disruption in neural synchrony and cross-frequency coupling, influenced by both cognitive and affective mechanisms. The integration of portable EEG technology with rhythmic auditory stimulation offers new insights into the neurophysiological and emotional aspects of dyslexia, highlighting the potential of rhythm- and music-based approaches for both diagnostic and therapeutic applications. Full article