Search Results (2,627)

Early detection of agricultural fires with Unmanned Aerial Vehicles (UAVs) is important for environmental safety, yet it remains difficult because ignition cues are extremely small, smoke patterns vary widely, and farmland scenes often contain strong background interference such as specular reflections. Model development is further constrained by the scarcity of data from the early ignition stage. To address these challenges, we propose a joint data and model optimization framework. We first build a hybrid dataset through an ROI-guided synthesis pipeline, in which latent diffusion models are used to insert high-fidelity, carefully screened fire samples into real farmland backgrounds. We then introduce EF-YOLO, a detector designed for high sensitivity to small targets. The network uses SPD-Conv to reduce feature loss during spatial downsampling and includes a high-resolution P2 head to improve the detection of minute objects. To reduce background clutter, a Dual-Path Frequency–Spatial Enhancement (DP-FSE) module serves as a lightweight statistical surrogate that extracts global contextual cues and local salient features in parallel, thereby suppressing high-frequency noise. Experimental results show that EF-YOLO achieves an AP_s of 40.2% on sub-pixel targets, exceeding the YOLOv8s baseline by 15.4 percentage points. With a recall of 88.7% and a real-time inference speed of 78 FPS, the proposed framework offers a strong balance between detection performance and efficiency, making it well suited for edge-deployed agricultural fire early-warning systems. Full article

The environmental persistence of β-lactam antibiotics represents a growing ecological concern, requiring materials capable of combined adsorption and catalytic degradation. Herein, pure ZnS and 1% Fe-doped ZnS nanoparticles were synthesized via microwave-assisted treatment and evaluated for the removal of ceftaroline fosamil from aqueous media. Transmission electron microscopy revealed quasi-spherical nanoparticles below 10 nm, while selected area electron diffraction confirmed a face-centered cubic structure retained after Fe incorporation. UV-Vis spectroscopy showed similar absorption edges (~316 nm), indicating negligible band-gap variation, whereas photoluminescence analysis demonstrated strong emission quenching in Fe-ZnS, indicating suppressed electron–hole recombination. Point-of-zero charge measurements (pH_PZC ≈ 4.6 for ZnS; 4.5 for Fe-ZnS) indicated negatively charged surfaces under circumneutral conditions, influencing interfacial interactions with the antibiotic. Adsorption experiments followed the Langmuir isotherm model, with Fe-ZnS exhibiting a higher maximum adsorption capacity (156 mg g⁻¹) compared to ZnS (115 mg g⁻¹). Under UV irradiation (302 nm), Fe-ZnS achieved near-complete degradation at a catalyst loading of 500 ppm. Liquid chromatography–mass spectrometry analysis revealed the transformation of ceftaroline fosamil (m/z 685.01) into ceftaroline (m/z 605.05) via phosphate group loss, followed by the formation of intermediate fragments at m/z 492.08 and 308.03, associated with cleavage of the thiadiazol-amine moiety and subsequent opening of the cephalosporin ring. After extended irradiation, these intermediates diminished, and a fragment at m/z 356.01 was detected, suggesting further breakdown through thioether bond cleavage. These results support a degradation pathway involving sequential dephosphorylation and fragmentation of the cephalosporin core. Overall, the enhanced performance of Fe-ZnS arises from the synergistic interplay between surface charge characteristics and dopant-modulated charge carrier dynamics, highlighting its potential for antibiotic remediation in aquatic environments. Full article

This article develops a boundary-adaptive nonparametric methodology for estimating the geometric conditional quantiles of a multivariate response when the conditioning covariate is supported on the simplex—an important case, as it is the natural domain of compositional data. The statistical difficulty addressed here is twofold. First, geometric conditional quantiles for multivariate responses must be defined and estimated through a genuinely directional and convex framework rather than through any scalar ordering. Second, when the covariate is compositional or otherwise simplex-constrained, conventional symmetric kernel procedures suffer from intrinsic support mismatch and severe boundary distortion, thereby compromising both estimation accuracy and inferential validity near faces and edges of the simplex. The method proposed in this paper is designed precisely to overcome this combined obstacle. Our main innovation consists in embedding the spatial quantile formalism of Chaudhuri within a Dirichlet–Kernel smoothing scheme whose shape parameters depend deterministically on the evaluation point. This produces a convex M-estimator that respects the simplex geometry exactly, automatically adapts its local shape to the position of the target point, and removes the need for artificial boundary corrections. To the best of our knowledge, this is the first contribution to provide a complete asymptotic treatment of geometric conditional quantile estimation under simplex-supported covariates with location-adaptive asymmetric kernels. We establish a Bahadur-type linear representation with an explicit negligible remainder, from which we derive refined asymptotic bias and variance expansions. The variance analysis reveals a distinctive geometric phenomenon: each coordinate direction approaching the simplex boundary induces an additional $b^{-1/2}$ inflation factor, so that the variance at a face of codimension $\left|\mathcal{J}\right|$ scales as $n^{-1}{b}^{-(s+|\mathcal{J}\left|\right)/2}$. We further obtain the asymptotic mean squared error, an explicit optimal bandwidth rate, asymptotic normality under the nonstandard normalization $n^{1/2}{b}^{-s/4}$, and consistent plug-in covariance estimators yielding valid confidence ellipsoids. Numerical experiments and a real-data illustration based on the GEMAS data confirm the practical merit of the approach, especially in boundary regions where classical methods are known to deteriorate. Full article

To address the urgent need for enhancing landslide risk monitoring in reservoir embankment slopes, a core component of water conservancy projects, this paper proposes the LRES-YOLO algorithm for real-time landslide detection on reservoir embankments. In LRES-YOLO, we first integrate coordinate attention into basic feature extraction convolutional blocks to form the CACBS attention module, which enhances the model’s ability to identify and locate landslide targets in complex reservoir terrain, overcoming positional information insensitivity in deep networks. Second, we add novel downsampling DP modules and ELAN-W modules to the backbone network, improving feature recognition efficiency for embankment slopes with diverse hydrological and topographical interference. Third, we optimize the feature fusion network with targeted concatenation and pooling operations, balancing semantic information enhancement with computational load reduction to mitigate overfitting in variable reservoir environments. Finally, we adopt Focal Loss and EIoU Loss to accelerate training convergence and strengthen target feature representation for small or obscured landslides on embankments. Experimental results show that LRES-YOLO outperforms traditional algorithms in detecting landslides across diverse reservoir embankment scenarios: it achieves an average improvement of 8.4 percentage points in mean mAP over the best-performing baseline across five independent trials, a detection speed of 8.2 ms per image, and memory usage of 139 MB. This lightweight design makes it suitable for edge computing devices, providing robust technical support for intelligent monitoring systems in water conservancy projects. 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

Millimeter-wave (mmWave) radar point clouds offer a privacy-preserving solution for Human Activity Recognition (HAR), but their inherent sparsity and noise limit single-modal performance. While multimodal fusion mitigates this issue, existing methods often suffer from severe negative transfer during visual degradation and incur high computational costs, unsuitable for edge devices. To address these challenges, we propose Tac-Mamba, a lightweight cross-modal state space model. First, we introduce a topology-guided distillation scheme that uses a Spatial Mamba teacher to extract structural priors from visual skeletons. These priors are then explicitly distilled into a Point Transformer v3 (PTv3) radar student with a modality dropout strategy. We also developed a Trust-Aware Cross-Modal Attention (TACMA) module to prevent negative transfer. It evaluates the reliability of visual features through a SiLU-activated cross-modal bilinear interaction, smoothly degrading to a pure radar-driven fallback projection when visual inputs are corrupted. Finally, a Lightweight Temporal Mamba Block (LTMB) with a Zero-Parameter Cross-Gating (ZPCG) mechanism captures long-range kinematic dependencies with linear complexity. Experiments on the public MM-Fi dataset under strict cross-environment protocols demonstrate that Tac-Mamba achieves competitive accuracies of 95.37% (multimodal) and 87.54% (radar-only) with only 0.86M parameters and 1.89 ms inference latency. These results highlight the model’s exceptional robustness to modality missingness and its feasibility for edge deployment. Full article

Boundary Layer Ingestion propulsors operate in an adverse aerodynamic environment with high levels of distortion. With the purpose of extending the operating range of transonic fan rotors for BLI applications, in this paper we present an optimisation study focused on blade profiles design under different working conditions. Quasi-2D blade sections are optimised using a genetic algorithm and numerical simulations, by varying the camberline and thickness distribution. A method to efficiently achieve a combination of total pressure ratio at a given relative inlet Mach number is devised. The isentropic efficiency is optimised at the design point, concurrently with the stall total pressure ratio at a lower inlet Mach number, in a multi-objective fashion. Pareto-optimal profiles exhibit a moderate leading edge concavity for high efficiency and a straighter fore part with increased trailing edge deflection for higher compression at stall. Optimised airfoils are used in a preliminary three-dimensional evaluation with a realistic BLI inflow, in which the unsteady full-annulus analysis corroborates the approach of the sectional optimisation, also showing the possibility of estimating the integral performance of the machine with a simplified approach based on a single-passage simulation with a circumferential-averaged inflow distribution. Full article

While Digital Twins offer transformative potential, their efficacy for real-time control is constrained by the slow data acquisition and the high computational intensity required to process raw datasets like point clouds. This paper identifies these critical bottlenecks—specifically the latency between data capture and actionable insight—and proposes a refined theoretical framework for near-real-time automated progress monitoring and quality reporting. Building on the findings of the NORMENG project and informing the subsequent AutoGreenTraC project, this research synthesizes state-of-the-art advancements in reality capture, including LIDAR, SfM-MVS, and 360-degree vision. The study highlights a fundamental divergence in stakeholder requirements: the need for millimeter-level precision in quality control versus the demand for high-velocity documentation for progress monitoring. A key innovation presented is the shift toward neural rendering techniques to bypass the computational delays of traditional photogrammetry and enable immediate on-site visualization. By structuring a tiered processing hierarchy that combines lightweight edge analysis for immediate safety and progress monitoring with asynchronous high-fidelity Digital Twin updates, the framework aims to establish a single source of truth. Full article

Owing to its exceptional mechanical and electrical properties, graphene is regarded as an ideal sensing material for piezoresistive pressure sensors. However, vacancy defects inevitably introduced during graphene preparation and transfer significantly alter its electrical characteristics and piezoresistive performance. Based on first-principles calculations, this work systematically investigates the influence of mono-, di-, and tri-vacancy defects on the electrical and piezoresistive properties of graphene. The results indicate that di- and tri-vacancy defects reconstruct into 5-8-5 and 5-10-5 configurations during relaxation. Mono-, di-, and tri-vacancy defects effectively open bandgaps in graphene, yielding values of 0.62, 0.48, and 0.72 eV, respectively. The mono-vacancy introduces localized defect states near the Fermi level, the di-vacancy shifts the Dirac point from K to M, and the tri-vacancy moves it along the K-Γ path, eventually placing it between K and Γ. The application of strain not only widens the bandgap of defective graphene but also induces the movement of defect energy levels toward the band edges in the mono-vacancy system. All three defect types enhance the piezoresistive effect, with the tri-vacancy defect showing the most pronounced enhancement—boosting the gauge factor by a factor of 5.58. These findings provide a theoretical foundation for optimizing graphene-based pressure sensors. Full article

The optimal deployment of Low-Power Wide-Area Networks (LPWANs) such as LoRaWAN in complex urban environments remains an NP-Hard Set Covering Problem. Traditional network planning often relies on 2D mathematical grids that ignore physical RF barriers, leading to topographic shadowing and single points of failure. This research proposes the Native 3D Memetic Spatially Aware Genetic Algorithm (3D-M-SAGA), an optimization framework that operates over a Morphological Digital Twin. By fusing OpenStreetMap (OSM) vector topologies with NASA SRTM elevation data and autonomous urban clutter classification, the framework evaluates physical constraints—including ITU-R knife-edge diffraction and dielectric absorption—directly within the evolutionary loop. To counteract the epistatic variance inherent to standard genetic algorithms, the 3D-M-SAGA integrates a vectorized memetic “Smart Repair” operator driven by heuristic attraction and repulsion forces. Formulated as a multi-objective optimization problem balancing Capital Expenditure (CAPEX) and topological Quality of Service (QoS) through K-coverage, the framework is evaluated using a 36-scenario parametric grid search and a 50-iteration Monte Carlo benchmark. Results show that the 3D-M-SAGA tightly bounds stochastic CAPEX variance ($\sigma =\pm 0.51$ gateways) while reducing single-point-of-failure network fragility ($K=1$) by up to 20%, guaranteeing fault tolerance ($K\ge 2$) without over-provisioning civic infrastructure. Full article

Transient-source detection without relying on difference images still faces challenges in achieving high accuracy, especially under practical space-based astronomical survey conditions where the data volume is enormous, on-orbit transmission bandwidth is limited, and real-time response is required for rapid follow-up observations. To address these issues, this paper proposes a lightweight detection network that integrates multi-scale feature fusion with contextual feature extraction, enabling efficient real-time processing on resource-constrained edge devices. The proposed model enhances robustness to point-spread-function variations across observation conditions and to complex background environments, while simultaneously improving detection accuracy. To evaluate performance comprehensively, lightweight VGG and lightweight ResNet architectures and other baseline models—commonly used as baselines for transient-source detection—are adopted for comparison. Experimental results show that under the condition that the models have approximately the same number of parameters, the proposed network achieves the best accuracy, obtaining nearly 1% improvement compared with the best-performing baseline model. Based on this design, an ultra-lightweight version with only 7k parameters is further developed by incorporating a compact multi-scale module, improving accuracy by 1% over the version without the multi-scale structure. Moreover, through heterogeneous knowledge distillation and adaptive iterative training, the accuracy of the ultra-lightweight model is further increased from 93.3% to 94.0%. Finally, the model is deployed and validated on an AI hardware acceleration platform. The results demonstrate that the proposed method substantially improves inference throughput while maintaining high accuracy, providing a practical solution for real-time, low-latency, on-device transient-source detection under large data volume and limited transmission conditions. Specifically, the proposed models are trained offline on a high-performance GPU and subsequently deployed on the Fudan Microelectronics 7100 AI board to evaluate their real-world inference efficiency on resource-constrained edge devices. Full article

The global shipping network, which handles over 80% of international trade volume, is increasingly exposed to disruptions from typhoons and other extreme weather events under climate change. However, conventional network vulnerability assessments often overlook the geographically heterogeneous nature of such natural hazards. Here, we introduce a typhoon-related systemic vulnerability model (GMSN-TV) that integrates three core components: typhoon exposure, port network sensitivity, and national adaptive capacity, to quantify the Typhoon Vulnerability Index (TVI) of 1075 major ports across 2017 and 2021. Our analysis reveals four key findings. First, the global shipping network became structurally sparser between 2017 and 2021, with edges declining by 17.84% and network efficiency decreasing by 4.22%, rendering it more susceptible to climate-related disruptions. Second, simulated TVI-based natural attacks and conventional degree-based deliberate attacks induce fundamentally different risk patterns: removing the top 10% high-TVI ports in 2021 caused a 6.3% decline in network efficiency, whereas removing the top 10% hub ports resulted in a 20.1% decline, a difference of 13.8 percentage points; however, natural attacks proved more effective at isolating peripheral ports, generating an isolated node ratio of 1.16% compared to 0.00% under deliberate attacks. Third, when removing the top 50% high TVI ports, the contribution of typhoon vulnerability to network degradation increased from 13.77% in 2017 to 15.87% in 2021. Fourth, high-vulnerability ports exhibit significant spatial clustering, with the Northwest Pacific region (50.8%) and the North Atlantic region (29.5%) collectively accounting for over 80% of global high-vulnerability ports in 2021. Compared to conventional topology-based assessments, the GMSN-TV analytical framework proposed in this study integrates typhoon hazard data with network topology, providing a novel scientific tool with enhanced identification efficacy and accuracy. It successfully captures local network disintegration effects entirely missed by traditional deliberate attacks, revealing an isolated node ratio of 12.5% after removing 70% of high-TVI ports. This demonstrates the tool’s precision in identifying hidden high-risk peripheral nodes, enabling decision-makers to prioritize climate adaptation investments for critical maritime infrastructure more accurately. Full article

Vegetation resilience plays a key role in ecosystem stability as climate change and human disturbance intensify. We quantified resilience via AR(1) from kNDVI data over mainland China (2000–2024), and assessed its spatiotemporal patterns, long-term causal drivers (Causal Forest), and breakpoint-related mechanism shifts (non-stationary causal networks). Resilience varied strongly across space, with higher AR(1) values concentrated in northern transition belts and inland regions. Breakpoints clustered in 2010–2018 and showed broad synchronicity nationwide. Long-term effects were dominated by environmental background states: mean variables generally outweighed variability (CV) and memory terms, suggesting that persistent climate–environment conditions primarily shaped resilience gradients. Temperature emerged as the strongest national-scale control and acted asymmetrically across metrics—TMX strongly suppressed resilience, whereas TMN tended to enhance it—while precipitation and CO₂ gained importance regionally. Driver networks reorganized markedly across breakpoints, exhibiting high edge turnover and heterogeneous lag shifts—pointing to stage-dependent restructuring that goes beyond changes in driver strength. This framework links net effects with mechanism reorganization to help diagnose vegetation resilience under non-stationary conditions. Full article

Visual perception in Unmanned Surface Vehicles (USVs) suffers from drastic lighting changes and missing texture features. These factors lead to depth scale drift and motion estimation bias. Moreover, existing multi-modal fusion models are computationally complex and unfit for resource-limited edge devices. To address these problems, a lightweight Radar–Vision Fusion (Geo-RVF) algorithm is proposed. To supplement spatial information, point clouds are projected to build sparse depth maps. A probability consistency-based depth correction module is designed to suppress water noise. This helps extract accurate geometric anchors to guide visual depth propagation. Subsequently, a Recurrent Autoregressive Network (RAN) fuses radar and image features in the temporal dimension. This resolves dynamic positional deviations caused by texture degradation and distant small targets. After real-time optimization, Geo-RVF achieves 23 FPS on the Jetson Orin NX. On a collected dataset, the method attains a mean average precision (mAP) 50–90 of 44.2% and a mean intersection over union (mIoU) of 99%, outperforming HybridNets and Achelous. Full article

This study has explored the effectiveness of machine learning methods for high-precision spatial interpolation of meteorological variables, aiming to provide accurate atmospheric delay corrections for high-precision edge and corner nets observation in complex-terrain environments such as the Xiluodu Hydropower Station, thereby enhancing the accuracy of deformation monitoring. Considering the significant limitations of traditional interpolation methods such as Inverse Distance Weighting (IDW) and Ordinary Kriging (OK) in capturing spatial variability under complex topographic conditions, we systematically introduced machine learning algorithms including Random Forest (RF)and eXtreme Gradient Boosting (XGBoost, XGB) to compare their performance with traditional methods for high-density interpolation of sparsely distributed temperature, relative humidity, and surface pressure, respectively. Concurrently, we proposed an enhanced XGB model incorporating center-point features (XGB-C) which frames spatial interpolation as a supervised learning problem that learns physical mapping from synoptic backgrounds to local microclimates instead of relying on geometric distances alone. The interpolation performance indices (RMSE, MAE, and R²) were evaluated with daily meteorological observations from 47 stations (38 for training, 9 for testing) during 2023–2024. Results demonstrate that machine learning methods significantly outperform traditional approaches, with XGB-C achieving the highest accuracy (R² ≈ 1.00 for pressure, 0.97 for humidity, 0.83 for temperature). Moreover, the interpolation performance also exhibits a dependence on seasons and the station location. Greater challenges are shown in the summer season and in the “Urban and Built-Up” and “Croplands” areas. These findings highlight the substantial advantages of machine learning, particularly the proposed XGB-C, for meteorological interpolation in mountainous hydropower station environments where accurate atmospheric correction is crucial for deformation monitoring. This also lays a solid foundation for developing operational ML-based interpolation models trained with high-quality labels derived from unmanned aerial vehicle (UAV) remote sensing data. Full article