Search Results (51,277)

To improve the accuracy of wildfire risk identification in areas adjacent to power transmission corridors, this study proposes a wildfire early warning method that integrates refined land cover segmentation and multimodal feature deep learning. First, an improved bi-branch semantic segmentation network (BuildFormer++) is used to perform refined classification of high-resolution remote sensing images, extracting six types of land cover information, including forest and cultivated land. Second, a multi-dimensional feature set integrating land cover, topography, climate, and human activities is constructed and input into a multimodal wildfire point prediction network for deep feature fusion and probabilistic modeling. Experimental results show that the proposed segmentation network achieves a mean intersection–union ratio (mIoU) of 40.68% in the semantic segmentation task; the early warning model achieves an accuracy of 85.37%, an F1 score of 93.15%, and an ROC-AUC of 85.42% in risk prediction, significantly outperforming comparative methods. The “refined segmentation–feature fusion–risk prediction” framework constructed by this method can provide reliable technical support for the operation and maintenance safety and fire prevention of power transmission corridors. Full article

This paper investigated the mechanism and dynamic process of a significant water and mud inrush disaster that occurred in the Baiyunshan Tunnel, which crosses a karst fault zone. By integrating multi-source data including geological exploration and geophysical surveys, a three-dimensional geological model characterizing the cave–conduit–tunnel system was developed. A numerical approach coupling the Phase-Field and Particle-Tracking Methods was employed, successfully reconstructing the entire disaster process involving the transport of water-air-mud three-phase flow. Simulation results demonstrated that the dynamic viscosity of the mudflow predominantly controls the dynamic characteristics of the particle, such as transport distance and mudflow velocity. Parameter sensitivity analysis revealed quantitative relationships between key mudflow parameters (transport distance, velocity, and drag force) and the Reynolds number, identifying an exponential decay of drag force with increasing Reynolds number in high-viscosity mudflows. This study establishes a comprehensive methodology from geological identification to numerical simulation, providing a theoretical basis and technical support for precise risk assessment and the design of preventive measures for tunnel water and mud inrush disasters. Full article

Accurate and real-time vehicle detection is a fundamental requirement for smart urban traffic monitoring, particularly in densely populated cities where heterogeneous traffic, frequent occlusion, and severe scale variation challenge lightweight vision systems deployed at the edge. To address these issues, this paper proposes BDNet, a lightweight YOLOv12-based vehicle detection framework designed to enhance feature preservation, contextual modeling, and multi-scale representation for intelligent transportation systems. BDNet integrates three complementary architectural components: (i) HyDASE, a hybrid detail-preserving downsampling module that mitigates information loss during resolution reduction; (ii) C3k2_MogaBlock, which strengthens long-range contextual interactions through multi-order gated aggregation; and (iii) an A2C2f_FRFN neck, which refines multi-scale features by suppressing redundancy and emphasizing discriminative responses. To support evaluation under realistic developing-region traffic conditions, we introduce the Bangladeshi Road Vehicle Dataset (BRVD), comprising 10,200 annotated images across 13 native vehicle categories captured under diverse urban scenarios, including daytime, nighttime, fog, and rain. On BRVD, BDNet achieves 85.9% ${m\mathcal{A}\mathcal{P}}_{50}$ and 67.3% ${m\mathcal{A}\mathcal{P}}_{50-95}$, outperforming YOLOv12n by +1.4 and +0.7 percentage points, respectively, while maintaining a compact footprint of 2.5 M parameters, 6.0 GFLOPs, and a real-time inference speed of 285.7 FPS. Cross-dataset evaluation on VisDrone-DET2019, using models trained exclusively on BRVD, further demonstrates improved generalization, achieving 31.9% ${m\mathcal{A}\mathcal{P}}_{50}$ and 17.9% ${m\mathcal{A}\mathcal{P}}_{50-95}$. These results indicate that BDNet provides an effective and resource-efficient vehicle detection solution for smart city–scale urban traffic monitoring. Full article

The rapid rise in the use of Liquefied Petroleum Gas (LPG) as an alternative vehicle fuel in Ghana presents both opportunities and risks within the national energy transition agenda. This study investigates LPG safety as well as environmental and regulatory implications using a multi-method quantitative approach that combines structured survey data, exploratory multivariate analysis (MCA), and machine learning classification (Random Forest) to uncover emerging associations and patterns in LPG safety practices. Primary data were obtained from 384 respondents, including vehicle operators, auto-technicians, regulatory officials, and LPG station attendants across five major transport zones: Kejetia, Asafo, Ahodwo, Bantama, and Suame Magazine. The MCA identified four distinct behavioural and safety profiles—At-Risk, Proactive Safety, Compliant and Equipped, and Formal and Reported—reflecting diverse compliance and risk patterns across socio-occupational groups. The Random Forest classifier achieved a predictive accuracy of 96.5% based on cross-validated performance. Sensitivity and specificity values were high, indicating reliable discrimination among incident types. To reduce the risk of overfitting, k-fold cross-validation and monitored error convergence were performed across increasing numbers of trees. While the model shows strong predictive capability, we present these results cautiously and emphasize observed associations and emerging patterns rather than definitive predictive conclusions. The findings reveal that while economic motivations underpin LPG adoption, weak institutional enforcement and widespread informal installations heighten safety vulnerabilities. Comparisons with sub-Saharan and Asian contexts underscore the need for a structured regulatory framework, mandatory certification of installers, and periodic vehicle inspections. The study contributes to the broader discourse on informal energy transitions in developing economies by demonstrating how technical and behavioural determinants interact within weak regulatory systems. Policy recommendations emphasize the integration of data-driven risk assessment tools into regulatory oversight to enhance vehicular LPG safety and sustainability. Full article

Natural fractures are critical controls on shale oil storage and migration in the Upper Triassic Chang 7 Member of the Ordos Basin. However, conventional identification techniques—such as mud-invasion correction, R/S rescaled range analysis, and radioactive element analysis—are time-consuming, computationally intensive, and highly dependent on specialized logging data, limiting their large-scale application. To overcome these challenges, this study develops a multi-modal deep neural network that integrates conventional well logs with borehole imaging data. A coupled convolutional neural network (CNN) and deep neural network (DNN) architecture was constructed to predict fracture occurrence, dip angle, and aperture. The model achieves dip-angle prediction accuracies of 98.82% for both training and testing datasets, while aperture prediction accuracies reach 95.97% and 95.91%, respectively. Predicted dip angles are concentrated between 65° and 80°, deviating by less than 0.48° from measured values, whereas apertures fall mainly within 0.5–4.5 cm, with deviations below 0.21 cm except in extreme cases. The CNN branch effectively extracts spatial features from imaging logs, while the DNN branch captures nonlinear relationships in conventional logs. The integrated framework substantially improves fracture characterization accuracy and efficiency. This study provides a scalable and cost-effective approach for rapid fracture identification based on conventional logging data, reducing reliance on specialized imaging logs and supporting integrated geological and engineering evaluations in shale oil reservoirs. Full article

Multi-outlet air-assisted sprayers are increasingly used for directional and zoned airflow to match varying canopy structures. In this study, a self-developed multi-outlet orchard air-assisted sprayer was investigated. Airflow velocity and direction were tested at different inlet areas, heights, and downstream horizontal distances using a three-dimensional ultrasonic anemometer. Analysis of variance (ANOVA) and regression modeling were applied to elucidate the effects of these three factors on airflow velocity, horizontal angle (θ), and elevation angle (Φ). The results showed that a stable alternating “primary jet–interaction zone” structure was formed in the spatial airflow field under all operating conditions, indicating that the fundamental airflow pattern was mainly governed by the sprayer layout. Varying the inlet area did not alter the basic airflow structure; however, the intensity and directional stability of the primary jets were significantly modified. Larger inlet openings produced higher airflow velocities, with a maximum near-field velocity of 19.7 m s⁻¹, whereas smaller inlet openings resulted in faster far-field attenuation and more pronounced diffusion. Increasing the inlet area caused the θ distribution peak to converge toward 0°, thereby improving axial coherence and directional stability. In contrast, decreasing the inlet area shifted Φ toward more negative values, with Φ reaching approximately −20° in the far field; moreover, far-field differences in Φ were more pronounced. Under the minimum inlet opening area condition (S1), the airflow velocity within the region 80–100 cm from the outlet can be stably maintained above 3 m/s, with a relatively uniform velocity distribution. This is beneficial for improving droplet deposition uniformity within the canopy and reducing droplet drift in non-target areas. Based on the experimental data, a regression model for mean airflow velocity was established (R² = 0.873), demonstrating good predictive performance and indicating that inlet-opening regulation is feasible. These findings provide a basis for airflow matching and spray-parameter optimization for different canopy structures. Full article

Wampee (Clausena lansium) harvesting currently relies heavily on manual labor, but automation is significantly hindered by clustered fruit growth patterns, small fruit sizes, and complex orchard backgrounds, which make accurate detection highly challenging. This study proposes Wampee-YOLO, a lightweight and high-precision model based on the YOLO11n architecture, specifically designed for real-time wampee detection in natural orchard environments. The proposed model integrates several architectural enhancements: the RFEMAConv module for expanded receptive fields, an AIFI module for improved small target interaction, and a C2PSA-MSCADYT structure to boost multi-scale adaptability. Additionally, a Triplet Attention mechanism strengthens multi-dimensional feature representation, while an AFPN-Pro2345 neck structure optimizes cross-scale feature fusion. Experimental results demonstrate that Wampee-YOLO achieves an mAP₅₀ of 90.3%, a precision of 92.1%, and F1 score of 87%. This represents a significant 3.4% mAP₅₀ improvement over the YOLO11n baseline, with a slight increase to 3.28 M parameters. Ablation studies further confirm that the AFPN-Pro2345 module provides the most substantial performance gain, increasing mAP₅₀ by 2.4%. The model effectively balances computational efficiency with detection accuracy. These findings indicate that Wampee-YOLO offers a robust and efficient visual detection solution suitable for deployment on resource-constrained edge devices in smart orchard applications. Full article

This paper proposed a stacking-based ensemble model to replace the traditional single machine learning model prediction approach, significantly improving the evaluation efficiency of SoC and SoH of lithium batteries. Firstly, a dataset was constructed including three input variables (temperature, current, and voltage) and two output variables (SoC and SoH). Pearson correlation coefficients and histograms were used for preliminary analysis of the correlations and distributions of the dataset. The multi-layer perceptron (MLP), support vector machine (SVM), random forest (RF), and extreme gradient boosting tree (XGB) were used as base prediction models. Bayesian optimization (BO) was used to fine-tune the parameters of these models, then three statistical indicators were compared to assess the prediction accuracy of the four ML models. Furthermore, MLP, SVM, and RF were selected as base models, while XGB was used as the meta-model, enhancing the integrated performance of the prediction models. SHAP was used to quantify the influence of the output variables on SoC. Finally, linked measures for the prediction model were proposed to achieve autonomous monitoring of drones. The results showed that XGB exhibited superior prediction accuracy, with R² of 0.93 and RMSE of 0.14. The ensemble model obtained using stacking reduced the number of outliers by 89.4%. Current was identified as the key variable influencing both SoC and SoH. Furthermore, the intelligent prediction model proposed in this paper can be integrated with controllers, visualization web pages, and other systems to enable the health status assessment of drones. Full article

Multimodal named-entity recognition (MNER) aims to identify entity information by leveraging multimodal features. With recent research shifting to multi-image scenarios, existing methods overlook modality noise and lack effective cross-modal interaction, leading to prominent semantic gaps. This study innovatively integrates symmetric multimodal fusion with contrastive learning, proposing a novel model with a symmetric-encoder collaborative architecture. To mitigate the noise, a modality refinement encoder maps each modality to an exclusive space, while an aligned encoder bridges gaps via contrastive learning in a shared space, surpassing the superficial cross-modal mapping of existing models. Building on these encoders, the symmetric fusion module achieves deep bidirectional fusion, breaking traditional one-way or concatenation-based limitations. Experiments on two datasets show the model outperforms state-of-the-art methods, with ablation experiments validating the symmetric encoder’s uniqueness for consistent multimodal learning. Full article

Medical image classification is a fundamental task in medical image analysis and underpins a wide range of clinical applications, including dermatological screening, retinal disease assessment, and malignant tissue detection. In recent years, diffusion models have demonstrated promising potential for medical image classification owing to their strong representation learning capability. However, existing diffusion-based classification methods often rely on oversimplified prior modeling strategies, which fail to adequately capture the intrinsic multi-scale semantic information and contextual dependencies inherent in medical images. As a result, the discriminative power and stability of feature representations are constrained in complex scenarios. In addition, fixed noise injection strategies neglect variations in sample-level prediction confidence, leading to uniform perturbations being imposed on samples with different levels of semantic reliability during the diffusion process, which in turn limits the model’s discriminative performance and generalization ability. To address these challenges, this paper proposes a Confidence-Guided Adaptive Diffusion Network (CGAD-Net) for medical image classification. Specifically, a hybrid prior modeling framework is introduced, consisting of a Hierarchical Pyramid Context Modeling (HPCM) module and an Intra-Scale Dilated Convolution Refinement (IDCR) module. These two components jointly enable the diffusion-based feature modeling process to effectively capture fine-grained structural details and global contextual semantic information. Furthermore, a Confidence-Guided Adaptive Noise Injection (CG-ANI) strategy is designed to dynamically regulate noise intensity during the diffusion process according to sample-level prediction confidence. Without altering the underlying discriminative objective, CG-ANI stabilizes model training and enhances robust representation learning for semantically ambiguous samples.Experimental results on multiple public medical image classification benchmarks, including HAM10000, APTOS2019, and Chaoyang, demonstrate that CGAD-Net achieves competitive performance in terms of classification accuracy, robustness, and training stability. These results validate the effectiveness and application potential of confidence-guided diffusion modeling for two-dimensional medical image classification tasks, and provide valuable insights for further research on diffusion models in the field of medical image analysis. Full article

Fog computing extends cloud services to the network edge, enabling low-latency processing for Internet of Things (IoT) applications. However, this distributed approach is vulnerable to a wide range of attacks, necessitating advanced intrusion detection systems (IDSs) that operate under resource constraints. This study proposes integrating self-awareness (online learning and concept drift adaptation) into a lightweight RL (reinforcement learning)-based IDS for fog networks and quantitatively comparing it with non-RL static thresholds and bandit-based approaches in real time. Novel self-aware reinforcement learning (RL) models, the Hierarchical Adaptive Thompson Sampling–Reinforcement Learning (HATS-RL) model, and the Federated Hierarchical Adaptive Thompson Sampling–Reinforcement Learning (F-HATS-RL), were proposed for real-time intrusion detection in a fog network. These self-aware RL policies integrated online uncertainty estimation and concept-drift detection to adapt to evolving attacks. The RL models were benchmarked against the static threshold (ST) model and a widely adopted linear bandit (Linear Upper Confidence Bound/LinUCB). A realistic fog network simulator with heterogeneous nodes and streaming traffic, including multi-type attack bursts and gradual concept drift, was established. The models’ detection performance was compared using metrics including latency, energy consumption, detection accuracy, and the area under the precision–recall curve (AUPR) and the area under the receiver operating characteristic curve (AUROC). Notably, the federated self-aware agent (F-HATS-RL) achieved the best AUROC (0.933) and AUPR (0.857), with a latency of 0.27 ms and the lowest energy consumption of 0.0137 mJ, indicating its ability to detect intrusions in fog networks with minimal energy. The findings suggest that self-aware RL agents can detect traffic–dynamic attack methods and adapt accordingly, resulting in more stable long-term performance. By contrast, a static model’s accuracy degrades under drift. Full article

The convergence of Lean manufacturing and Industry 4.0 requires digital infrastructures capable of transforming high-frequency telemetry into actionable insights. However, architectures that integrate near real-time data with closed-loop process control remain scarce, particularly in the food-processing industry. This study proposes a “Lean 4.0” framework based on a six-layer Digital Twin (DT) architecture to digitise waste detection and optimise a medium-scale bakery. The methodology integrates a heterogeneous Industrial Internet of Things (IIoT) network comprising 17 ESP32 (Espressif Systems, Shanghai, China)-based monitoring nodes. Data collection is managed via an edge-centric MQTT–InfluxDB (version 2.7, InfluxData, San Francisco, CA, USA) data pipeline. Furthermore, the analytics layer employs discrete-event simulation in Siemens Plant Simulation (version 2302, Siemens Digital Industries Software, Plano, TX, USA), constraint programming with Google OR-Tools (version 9.8, Google LLC, Mountain View, CA, USA), and machine learning models (Isolation Forest and SARIMA). Multi-month validation in a brownfield bakery, including a 60-day continuous monitoring test, demonstrated that the proposed architecture reduced production cycle time by 24.4% and inter-operational waiting time by 51.2%. Moreover, manual planning time decreased by 87.4% through the use of low-code scheduling interfaces. In addition, state-based control of critical ovens reduced energy consumption by 23.06%. These findings indicate that combining deterministic simulation and combinatorial optimisation with data-driven analytics provides a scalable blueprint for implementing cyber-physical systems in food-processing SMEs. This approach effectively bridges the gap between traditional Lean principles and data-driven smart manufacturing. Full article

Despite the aesthetic potential of free-form envelopes in large-scale public buildings, geometric interlacing complexity, ambiguous façade boundaries, and constructability translation gaps persist as systemic barriers. This study addresses these challenges through a Design Science Research (DSR) approach, developing a rule-based digital twin methodology that maintains parametric intelligence across the building life cycle. Implemented via a five-layer integrated framework, i.e., geometric, parametric, BIM, coordination, and fabrication, the methodology was validated through a revelatory case study of the Shenzhen Bay Culture Plaza. Results demonstrate 91.2% clash resolution prior to construction, 20.3 million RMB in cost savings (10.8% reduction), and 35.4% schedule compression, while preserving rule-based relationships into operational facility management. The study advances BIM theory by operationalizing life-cycle digital twins for non-standard geometries, offering a replicable framework for future special-shaped construction projects. Full article

Persistent cloud cover and frequent rainfall in subtropical regions throughout the year significantly limit the applicability of optical remote sensing for tree species identification, thereby constraining dynamic forest monitoring and precise management of forest resources. To address this challenge, this study proposes a tree species identification method that integrates multi-source remote sensing temporal features. By combining multi-temporal optical imagery from Sentinel-2 and dual-polarisation Synthetic Aperture Radar (SAR) data from Sentinel-1, we constructed a comprehensive feature set that incorporates spectral, structural, and phenological attributes, including various vegetation indices, backscatter coefficients, and polarimetric decomposition parameters. Through correlation analysis and assessment of temporal feature variability, five distinct integration strategies (T1-T5) were developed to classify six typical subtropical tree species: Pinus massoniana, Pinus elliottii, Acacia, Eucalyptus grandis, Mangrove, and Other hardwoods, using a random forest classifier. The results indicate that the multi-source feature fusion approach significantly outperforms single-source models, with the T5 strategy achieving the highest overall accuracy (OA) of 95.33% and a Kappa coefficient of 0.94. The red-edge vegetation indices and SAR polarimetric features were identified as major contributors to improving the classification accuracy of hardwood species. This study demonstrates that multi-source remote sensing data fusion can effectively mitigate the spatiotemporal constraints of optical imagery, providing a viable solution and technical framework for high-accuracy remote sensing classification in complex subtropical forest environments. Full article

Urban community children’s play spaces play a crucial role in promoting both physical and mental health, significantly influencing children’s development and fostering a sense of belonging to the community. However, existing design practices often fail to adequately address the complex behavioral and emotional needs of children in these spaces. To overcome this gap, there is an urgent need for a system that can effectively respond to these complexities, thereby enhancing children’s play experiences and their attachment to the space. This study seeks to optimize the design of children’s play spaces in urban communities through a quantitative approach based on Exploratory Factor Analysis (EFA) and Structural Equation Modeling (SEM). First, multi-dimensional data concerning children’s physical environment, subjective perceptions, play behaviors, and satisfaction were gathered through field surveys and questionnaires. Reliability and validity assessments were conducted to ensure data quality. Subsequently, EFA was applied to perform dimensionality reduction and identify the underlying structure, resulting in the extraction of six key factors that influence children’s play experiences. Finally, SEM was utilized to construct a structural model, test hypotheses, and quantify the relationships between the identified dimensions. The results demonstrate that the EFA-SEM framework effectively transforms subjective concepts into actionable design parameters, meeting user needs and providing a solid scientific foundation for the design of children’s play spaces in urban communities. Full article