Search Results (280)

Search-and-rescue (SAR) missions following earthquakes and other disasters require aerial video perception systems that do more than detect objects in isolated frames. Operators must maintain the identities of access points, vehicles, responders, hazards, and other mission-relevant targets despite UAV ego-motion, dust, debris, occlusion, scale variation, and abrupt scene transitions. This paper presents E’CHIT (Edge-Oriented Colour Histogram Instance-Guided Tracking), a deployment-oriented, operator-centric UAV tracking framework for real-world disaster-response applications. Its primary scientific contribution is an identity-stabilised, detector-assisted tracking methodology. YOLOv8-seg proposals trained on D’RespNeT initialise and refresh tracks; a Custom-RE3 recurrent module propagates target states through short detector dropouts; and a lightweight EOMC verifier, based on edge orientation, mean colour, and shape consistency, determines whether tracks should be accepted, refreshed, or reacquired. A scene-cut watchdog that combines luminance mean absolute difference (MAD) with HSV histogram divergence prevents stale identities from carrying over after hard edits or sudden feed changes. Custom-RE3 is the continuation module implemented and evaluated in this study. The surrounding E’CHIT wrapper follows an initialise–reseed–verify–reset cycle and is tracker-adaptable at the software-interface level: another compatible SOT or MOT continuation module can be integrated through adapter modifications, state and bounding-box conversion, and method-specific retuning, followed by independent validation. All reported quantitative results therefore apply to the Custom-RE3 implementation. D’RespNeT, the optional reinforcement learning (RL) warm start, the HUD, and the deployment stack support this central tracking contribution. D’RespNeT provides 28 polygon-annotated SAR classes. An author-developed PPO/SAC script is used only during offline detector training. In the reported runs, it produces different early optimisation trajectories for selected difficult or under-represented classes, while the default supervised schedule remains the strongest final global mAP reference. No RL policy runs during deployment; the detector architecture, parameter count, and inference graph remain unchanged. Evaluation on D’RespNeT and authentic disaster-response UAV footage shows that E’CHIT increases Success@IoU ≥ 0.5 from 0.62 to 0.79, reduces identity switches by approximately 71%, and maintains real-time 1080p performance, achieving 164–330 FPS for single-target tracking and 24–100+ FPS for end-to-end multi-target operation on an RTX-class GPU using FP16. The VOT2014, NT-VOT211, and VOTS2024 figures reproduce historical result spaces reported in the literature and include a clearly labelled, non-official E’CHIT operating-point marker solely for context. This marker was not produced using the corresponding official datasets, toolkits, reset rules, or submission routes; it is excluded from the primary quantitative claims and must not be interpreted as a leaderboard rank or a protocol-identical comparison. Overall, the system demonstrates how identity-stable UAV tracks can provide actionable operator cues for target monitoring, entry-point assessment, and UAV–UGV/ground-team coordination in cluttered disaster scenes. Full article

High-precision tunnel breakthroughs depend critically on the north-seeking accuracy of maglev gyroscopes. However, external disturbances during underground construction often introduce abrupt jumps into rotor current signals, significantly reducing the orientation reliability. Existing signal-processing methods either require manually defined segmentation windows or apply identical denoising strategies to both stationary and disturbed signal intervals, resulting in limited adaptability and suboptimal denoising performance. To overcome these limitations, this study proposes an improved rotor current denoising algorithm based on the MAF-ARIMA framework by incorporating the Bayesian ensemble algorithm for abrupt change, seasonality, and trend (BEAST) and an optimized wavelet transform (OWT). First, the BEAST is employed to automatically detect the structural change point of the rotor current signal, enabling the adaptive segmentation of stationary and jump intervals without manual intervention. Subsequently, empirical mode decomposition is performed, and the OWT applies different denoising parameters to the dominant components of the stationary and jump segments according to their distinct fluctuation characteristics. Finally, moving-average smoothing is adopted to preserve the signal continuity at the segmentation boundary, while the autoregressive integrated moving average (ARIMA) model reconstructs the missing trend component of the jump interval to obtain the complete denoised signal. Comparative experiments using 12 field-collected rotor current datasets demonstrated that the proposed method reduced the standard deviation of the denoised signal by 70.96% and the absolute azimuth error by 50.36% compared with the raw signal, outperforming the optimized Hilbert–Huang transform, HSA-KS, and the original MAF-ARIMA algorithm. By introducing adaptive change-point detection and segment-specific denoising into the existing MAF-ARIMA framework, the proposed method significantly improves the adaptability and denoising performance of maglev gyro rotor current processing under complex tunnel construction environments while preserving the signal continuity and reconstruction accuracy. Full article

Mining disturbances can promote roof separation, fracture propagation, strata collapse, and water-conducting fracture-zone development, increasing roof water-hazard risk. Conventional apparent-resistivity pseudosection imaging is useful for rapid display; however, restricted electrode deployment limits effective coverage and representation of anomaly position and spatial continuity. Time-lapse resistivity inversion can characterize progressive fracture development, but representation of discontinuous anomalies caused by rupture, fracture connection, or collapse can be affected by inversion model constraints. To address these limitations, this study proposes a full-space apparent-resistivity rapid imaging method based on point-source attenuation fields. Each current electrode is regarded as a point current source, and potential attenuation with distance is used to construct attenuation curves, map responses to target-region grids, fuse multi-source estimates, and extract representative apparent-resistivity values. Numerical simulations and a scaled physical model experiment show that the method improves the spatial continuity of electrical anomaly responses and provides a more direct representation of abrupt electrical changes. Field application indicates that the method can identify mining-related electrical anomalies in roofs and anomalous ranges potentially associated with fracture development. The maximum vertical extent of the electrical anomaly was approximately 37 m, which was broadly consistent with the empirical estimate of approximately 40 m. The proposed approach provides an efficient geoelectrical monitoring tool for roof water-hazard identification and fracture-zone delineation in coal mining. Full article

As the core power equipment of the deep-sea mining system, the deep-sea mineral lifting pump continuously operates under harsh service conditions coupled with high hydrostatic pressure, complex marine environments, and solid particle media. Its operating state shows obvious strong nonlinearity, time-varying fluctuation, and local abrupt change features. Therefore, time series health state prediction of the deep-sea mineral lifting pump is of vital engineering significance for realizing predictive maintenance and ensuring the safety of offshore trials and mining operations. Taking the 500 m-level offshore sea trial conducted in the Xisha area of the South China Sea as the engineering background, four critical health characteristic parameters, including shaft power, pump efficiency, motor winding temperature, and outlet radial vibration, are selected to construct a hybrid CNN-LSTM time series prediction model. Comprehensive model evaluation metrics and ablation comparison experiments are adopted to analyze the multi-step-ahead prediction performance of the proposed model. The results show that the CNN-LSTM model achieves optimal comprehensive evaluation indices in one-step prediction and possesses excellent tracking capability for inflection points and amplitude fluctuations of time series data. Although the prediction accuracy decreases gradually with the increase in prediction steps, the model can still effectively characterize the evolutionary trend of pump operating states, and its overall prediction performance is significantly superior to that of single models. This study provides model support and technical reference for the health evaluation, early fault alarm, and maintenance optimization of deep-sea mineral lifting pumps in offshore trials. Full article

In the current paper, pulmonary emphysema is hypothesized to emerge from a nonlinear breakdown of cooperation across two tightly coupled systems: the extracellular matrix (ECM) crosslink network and the cellular populations responsible for its maintenance. To formalize this concept, we construct a game-theoretic model that unifies the mechanical failure, inflammatory changes, and percolation-driven tissue collapse that are recognized features of the disease. At the ECM level, elastin and collagen crosslinks are modeled as players in an iterated Prisoner’s Dilemma, where cooperation corresponds to maintaining structural integrity, and defection corresponds to rupture under mechanical stress. At the cellular level, fibroblasts, macrophages, and neutrophils engage in a parallel strategic game in which repair reflects cooperative activity, and protease- or oxidant-producing phenotypes are indicative of defection. These parallel games are coupled through bidirectional payoff modulation, generating a dynamical system with bistability, tipping points, and runaway positive feedback. As the fraction of intact crosslinks falls below a critical percolation threshold, global network connectivity collapses and lung function drops precipitously. This framework explains the characteristic features of pulmonary emphysema, including spatial heterogeneity, abrupt acceleration, and irreversibility as emergent properties of coupled cooperation–defection dynamics, and identifies new leverage points for stabilizing cooperation and preventing catastrophic network failure in early disease. In support of this hypothesis, we present previously published studies from our laboratory involving measurements of elastin-specific desmosine crosslinks in human postmortem emphysematous lungs showing a marked increase in tissue crosslink density at the early stage of the disease, and accelerating loss of these crosslinks as airspace enlargement progresses, consistent with initial cooperation followed by defection. This conceptual framework is then applied to the poorly understood lung disease, Combined Pulmonary Fibrosis and Emphysema, to provide a potential mechanism for its pathogenesis. Full article

Near-surface geometric defects and local electrical nonuniformities in semiconductor wafers are difficult to identify simultaneously using conventional far-field inspection methods. In this study, a three-dimensional electromagnetic simulation model based on AFM tip-enhanced near-field coupling was developed to investigate the local defect responses of AlN wide-bandgap semiconductors at 110 GHz. A finite-conductivity Pt80Ir20 metallic AFM probe, a low-loss AlN sample, and a point field probe in CST were used to extract the localized electric-field response near the tip apex. Lateral scanning response and normalized response variation were introduced to evaluate near-field perturbations induced by different defects. The results show that the metallic AFM tip produces a strongly localized electric-field enhancement within the tip–sample gap. Surface cracks, subsurface voids, and local Drude-AlN electrical anomaly blocks all lead to distinguishable near-field response variations. Geometric defects mainly cause local field redistribution and abrupt changes in lateral scanning curves, whereas the Drude-AlN anomaly produces interface transition and depth-dependent attenuation without changing the surface morphology. These findings indicate that AFM-enhanced near-field electromagnetic responses can provide a useful simulation basis for potential nondestructive characterization and signal interpretation of near-surface defects in AlN and related wide-bandgap semiconductors. Full article

Wind power ramp events (abrupt swings in output driven by frontal passages, sea-breeze transitions, and turbulence) are among the hardest problems for operators integrating renewables. Forecasting models are usually judged by aggregate error metrics (MAE, RMSE, overall directional accuracy) that average stable and ramp periods together, masking how a model behaves during the ramps that actually stress the grid. We address this on two fronts. First, we propose a regime-stratified evaluation that reports ramp event directional accuracy (ramp-DA) separately from stable-period accuracy and argue that ramp-DA should be a primary metric for grid-integration forecasting. Second, we build an ensemble of five regime-specialised sub-models trained with a direction-focused loss that penalises sign errors in the forecast power change, using only on-site SCADA wind speed and power. On 33,411 held-out samples from three onshore Greek farms, the ensemble reaches 76.5% ramp-DA, against 70.1% for a two-layer LSTM (+6.4 pp) and 74.3% and 74.1% for the PatchTST and iTransformer baselines. A strict leave-one-farm-out test retains 77.4% ramp-DA on a fully unseen farm. Overall directional accuracy rises 3.4 points, evidence that aggregate metrics understate the ramp-focused gain, while mean absolute error falls 19% compared to the LSTM (Diebold–Mariano p < 0.001). Full article

Traditional non-invasive harmonic impedance estimation methods often suffer from significant accuracy degradation or even failure under conditions of large background harmonic voltage fluctuations and abrupt changes in harmonic impedance. To address these challenges, a novel estimation method for utility harmonic impedance is proposed. First, a sliding-window-based isolation forest (IF) method is proposed to extract data segments exhibiting a strong correlation between harmonic voltage and current, thereby reducing the adverse impact of background harmonic voltage fluctuations on the impedance estimation results. Then, the Bernaola Galvan (BG) method is proposed to detect abrupt change points in harmonic impedance, and data are grouped accordingly based on the detected change points. Finally, a binary linear regression method is employed to estimate the utility harmonic impedance. The proposed method is validated using both simulation and field data, achieving relative errors below 4% and outperforming conventional methods under complex conditions. Full article

Hydrogen refueling station (HRS) requires continuous safety monitoring, yet conventional management relies largely on periodic inspection and manual oversight, limiting proactive risk mitigation. This study presents a data-driven intelligent analysis platform for real-time monitoring and anomaly detection of HRS safety data, including pressure, temperature, and flow-rate measurements from compressors, storage tanks, and dispensers. The platform integrates data collection adapters, a time-series database, and machine learning-based diagnostic modules (regression, clustering, and classification) into a unified reference software framework. For anomaly detection, an unsupervised LSTM-Variational Autoencoder trained on normal operating data is combined with DBSCAN-based clustering and a Mann–Kendall trend test to jointly identify point anomalies and pattern-level drifts, addressing the scarcity of labeled abnormal data in HRS environments. A continual learning mechanism further adapts detection thresholds to gradual and abrupt pattern changes without full retraining. The system was deployed and validated at BAM’s demonstration hydrogen refueling station in Germany, integrated with a remote safety-monitoring system and confirmed through performance testing, demonstrating reliable, proactive hydrogen safety management. Full article

This paper presents an experimental comparative study of four advanced maximum power point tracking (MPPT) strategies—backstepping, adaptive backstepping, sliding mode control, and vision-based MPPT—combined with three wireless communication technologies: IEEE 802.15.4, Wi-Fi, and 3G. The comparison is performed on a distributed photovoltaic (PV) platform under common power-stage conditions in harmonized irradiance scenarios, including abrupt uniform-irradiance transients and controlled partial shading patterns. Under uniform irradiance, adaptive backstepping achieved the best overall dynamic behavior, with the shortest average convergence time of 0.045 s using IEEE 802.15.4, compared with 0.052 s for Wi-Fi and 0.115 s for 3G, while all tested configurations maintained tracking efficiencies above 98.8%. Under partial shading, the ranking changed substantially: the vision-based MPPT provided the best GMPP-oriented performance, reaching the highest tracking success rate and the lowest energy loss. In particular, its lost energy increased from 0.20% with IEEE 802.15.4 to 0.45% with 3G, whereas conventional backstepping increased from 0.65% to 1.35% over the same communication range. Latency measurements showed that IEEE 802.15.4 exhibited the lowest median end-to-end delay and the smallest dispersion, Wi-Fi showed intermediate behavior, and 3G introduced the largest latency and variability. The results demonstrate that MPPT performance in distributed PV systems depends on both the control strategy and the communication architecture and that a unified experimental assessment of both layers is required to identify the optimal MPPT–communication combination for each operating scenario. Full article

Identifying shear band locations is essential for precursor recognition and early warning of progressive failure in gravelly soil slopes. However, conventional displacement monitoring methods mainly capture macroscopic slope deformation and remain limited in detecting internal localized deformation and the spatial evolution of shear bands. To address this limitation, this study proposes an MEMS data-driven framework for predicting spatial rotation-angle responses and locating potential shear bands in gravelly soil slopes, with the aim of enhancing the perception of internal shear deformation and detecting potential instability zones. First, scaled laboratory model tests were conducted under different gravel contents, and MEMS sensors were embedded within the slope to measure cumulative rotation-angle responses during shear band formation. Second, based on a DEM model incorporating particle geometric morphology, the spatial differentiation of the rotation-angle field during shear band evolution was analyzed, and the experimental results were further validated. Finally, a shear band localization framework integrating PDL-GAN-based data augmentation with PCA + Gaussian regional rotation-angle field prediction was established. Potential shear band locations were then indirectly localized based on the positive–negative partitioning and abrupt amplitude changes in the predicted rotation angles. The results show that the DEM simulations agree well with the cumulative rotation-angle responses obtained from the laboratory tests, with mean absolute percentage errors of 9.79%, 6.03%, and 3.84% under the T1, T2, and T3 conditions, respectively. Within the shear band influence zone, the upper monitoring points mainly exhibit negative rotation-angle accumulation, whereas the lower monitoring points primarily show positive rotation-angle responses. A larger absolute rotation angle indicates a stronger controlling effect of the shear band on the corresponding monitoring point. The PCA + Gaussian model demonstrates strong overall performance in regional rotation-angle prediction, with MAE, RMSE, and CRPS values of 0.0803, 0.1024, and 0.0801, respectively. The model preserves the dominant deformation mode of the rotation-angle field and provides probabilistic prediction outputs. The proposed method enables the prediction of internal rotation-angle responses and facilitates the localization of potential shear band locations in gravelly soil slopes, providing data-driven technical support for precursor recognition and intelligent early warning of progressive slope failure. Full article

Accurate next-day stock price forecasting remains challenging because daily price changes have a low signal-to-noise ratio and can be strongly affected by short-lived news shocks, order-flow imbalances, and abrupt changes in volatility or market sentiment. This study presents a controlled empirical comparison of deep learning architectures for next-day stock price forecasting using technical indicators. Using a decade-long daily dataset covering four large-cap NASDAQ equities (AAPL, META, SBUX, and TSLA), multivariate input sequences are constructed by combining historical prices with five widely used technical indicators: exponential moving average (EMA), relative strength index (RSI), moving average convergence divergence (MACD), on-balance volume (OBV), and average true range (ATR). Four deep sequence architectures—long short-term memory (LSTM), bidirectional LSTM (BiLSTM), gated recurrent unit (GRU), and convolutional LSTM (ConvLSTM)—are evaluated across multiple lookback windows (5, 15, and 30 trading days) and chronological train/validation/test splits (60–20–20, 70–15–15, and 80–10–10). Hyperparameters are optimized through random search, and forecasting performance is assessed on held-out test sets using normalized-scale root mean squared error (RMSE) and out-of-sample $R^2$. Within the examined fixed chronological partitions, ConvLSTM records the lowest observed RMSE for all four equities, attaining values between 0.0256 and 0.0394 and out-of-sample $R^2$ values above 0.90. Because the evaluation does not include walk-forward validation or formal statistical significance testing, these results should be interpreted as descriptive evidence within the present experimental setting rather than as proof of general architectural superiority. To assess practical utility, forecasts are translated into a transparent long-only trading rule that enters the market when the predicted next-day closing price exceeds the current closing price. Out-of-sample backtesting shows that the frictionless forecast-driven strategy achieves higher terminal cumulative returns than Buy-and-Hold for AAPL, SBUX, and TSLA, while Buy-and-Hold remains superior for META. Approximate five-day-frequency risk-adjusted estimates generally reinforce these relative patterns: the ConvLSTM strategy improves the Sharpe, Sortino, and Calmar ratios for AAPL, SBUX, and TSLA, although TSLA remains exposed to substantial drawdown risk. Transaction-cost sensitivity analysis further indicates that the terminal-return gains weaken under trading frictions and are particularly sensitive for AAPL. The findings demonstrate the value of evaluating forecasting architectures through both statistical and financial criteria, while emphasizing that lower point-forecast error does not necessarily translate into superior economic or risk-adjusted performance. Full article

Arid inland saline lakes are key components of basin ecosystems. As the largest saline lake in Xinjiang and a critical ecological barrier in northwest China, Ebinur Lake’s area dynamics are vital to regional sustainable development. This study integrates Landsat imagery (1992–2024) with meteorological and socio-economic data to investigate optimal water extraction methods, spatio-temporal lake area variations, and driving mechanisms. Multiple methods were employed, including water index comparison, Mann–Kendall test, Pearson correlation, and ridge regression. Results show that: (1) the Normalized Difference Water Index (NDWI) maintains high, stable classification accuracy across years and months, making it suitable for long-term monitoring; (2) from 1992 to 2024, lake area demonstrates a significant fluctuating downward trend without abrupt change points, indicating continuous degradation. During the growing season (April–October), it first decreases and then increases, with larger early-season areas, minima in August and September, coinciding with peak agricultural irrigation demand; (3) regarding driving mechanisms, socio-economic factors dominate (approximately 70%), while meteorological factors play a weakly regulatory role (about 30%). Population growth and increased water consumption are the primary drivers, with obvious seasonal differences. Meteorological changes, socio-economic development, and ecological measures jointly influence lake area. Although extreme events (e.g., anomalous precipitation) induce short-term fluctuations, they do not alter the long-term degradation trend dominated by human activities. This study provides methodological support for long-term monitoring of arid saline lakes and scientific evidence for ecological conservation and water resource management in the Ebinur Lake Basin. Full article

Industrial-style 3D point clouds frequently exhibit intricate geometric structures such as abrupt curvature changes, multi-surface junctions, and fine surface seams. Extracting fine boundary cues from these data remains challenging due to three intertwined factors: cross-boundary neighborhood contamination caused by purely spatial neighbor selection, orientation-sensitive boundary contrast in arbitrary poses, and over-smoothing from uniform feature aggregation. To address these issues, we present GeoRefine-Net, a geometry-consistent boundary refinement framework that progressively refines boundary predictions through three dedicated mechanisms. The geometry-consistent graph construction (GCG) module incorporates multi-order geometric statistics and a learnable boundary prior to actively discourage erroneous cross-surface neighbor associations during graph construction. The Rotation-aware Boundary Contrast Encoding (RBC) module captures local directional disparities using a composite descriptor incorporating normal cross-product interaction, angular divergence, curvature discrepancy, and tangent-plane projection, thereby providing rotation-consistency-enhanced boundary cues across varying object orientations. The Confidence-Guided Boundary Refinement (CBR) module introduces a gating mechanism that adaptively amplifies or suppresses neighbor contributions based on learned consistency scores, producing sharper boundary activation while suppressing noisy responses. The main novelty of GeoRefine-Net lies in the coupled design of geometry-consistent graph construction, rotation-consistency-enhanced boundary contrast encoding, and confidence-guided aggregation, rather than in any single handcrafted descriptor or isolated network component. Additional evaluations using SymBLE, PR-AUC, directed-versus-symmetric GCG, normal and curvature perturbations, normal-free variants, coefficient sensitivity analysis, and inference-time profiling further characterize the robustness and limitations of the proposed framework. Comprehensive evaluations on the ABC CAD dataset demonstrate that GeoRefine-Net outperforms the strongest directly compared baseline by 2.8 percentage points in the F1 score. Full article

Staged upward seepage can trigger particle migration and pore-structure adjustment in graded sandy soils, but the resulting transition behavior remains difficult to identify quantitatively. In this study, three representative sandy soils from a deep overburden deposit in southeastern Tibet were tested using a laboratory vertical upward seepage apparatus. Eight specimens with different nominal preparation states were subjected to stepwise increases in hydraulic head difference. Local hydraulic gradient, seepage velocity, hydraulic conductivity, and macroscopic outflow phenomena were monitored. Apparent seepage-transition hydraulic gradients were identified by combining abrupt changes in k–i curves, conductivity ratios between eligible staged records, and observed seepage responses. The clearest transition occurred in the nominal loose specimen of Soil 2, where the temperature-corrected hydraulic conductivity k₂₀ increased from 1.76 × 10⁻³ to 2.25 × 10⁻² cm s⁻¹ as i increased from 0.20 to 0.25, giving k_20,2/k_20,1 = 12.80 and i_c = 0.225. A clear transition was also identified for the nominal dense specimen of Soil 3, with k_20,2/k_20,1 = 6.61 and i_c = 0.583. Clear transitions were identified in the tested specimens only for Groups D and H, whereas the remaining specimens showed weak or phenomenon-assisted responses, local high-gradient fluctuations, or anomalous loading-path records rather than uniformly identifiable transition points. These results show that apparent transition gradients are path-dependent and should be evaluated together with loading history, seepage-velocity evolution, conductivity ratios, and macroscopic observations. Full article