Search Results (213)

Unmanned Aerial Vehicle (UAV) platforms enable flexible and cost-effective vehicle detection for intelligent transportation systems, yet small-scale vehicles in complex aerial scenes pose substantial challenges from extreme scale variations, environmental interference, and single-sensor limitations. We present AMSRDet (Adaptive Multi-Scale Remote Sensing Detector), an adaptive multi-scale detection network fusing infrared (IR) and visible (RGB) modalities for robust UAV-based vehicle detection. Our framework comprises four novel components: (1) a MobileMamba-based dual-stream encoder extracting complementary features via Selective State-Space 2D (SS2D) blocks with linear complexity $O\left(HWC\right)$, achieving 2.1× efficiency improvement over standard Transformers; (2) a Cross-Modal Global Fusion (CMGF) module capturing global dependencies through spatial-channel attention while suppressing modality-specific noise via adaptive gating; (3) a Scale-Coordinate Attention Fusion (SCAF) module integrating multi-scale features via coordinate attention and learned scale-aware weighting, improving small object detection by 2.5 percentage points; and (4) a Separable Dynamic Decoder generating scale-adaptive predictions through content-aware dynamic convolution, reducing computational cost by 48.9% compared to standard DETR decoders. On the DroneVehicle dataset, AMSRDet achieves 45.8% mAP@0.5:0.95 (81.2% mAP@0.5) at 68.3 Frames Per Second (FPS) with 28.6 million (M) parameters and 47.2 Giga Floating Point Operations (GFLOPs), outperforming twenty state-of-the-art detectors including YOLOv12 (+0.7% mAP), DEIM (+0.8% mAP), and Mamba-YOLO (+1.5% mAP). Cross-dataset evaluation on Camera-vehicle yields 52.3% mAP without fine-tuning, demonstrating strong generalization across viewpoints and scenarios. Full article

Accurate river network mapping is essential for hydrological modeling, flood risk assessment, and watershed environment management. However, conventional methods based on either optical imagery or digital elevation models (DEMs) often suffer from river network discontinuity and poor representation of morphologically complex rivers. To overcome this limitation, this study proposes a novel method integrating the river-oriented Gradient Boosting Tree model (RGBT) and adaptive stream burning algorithm for high-precision and topologically consistent river network extraction. Water-oriented multispectral indices and multi-scale linear geometric features are first fused and input for a river-oriented Gradient Boosting Tree model to generate river probability maps. A direction-constrained region growing strategy is then applied to derive spatially coherent river vectors. These vectors are finally integrated into a spatially adaptive stream burning algorithm to construct a conditional DEM for hydrological coherent river network extraction. We select eight representative regions with diverse topographical characteristics to evaluate the performance of our method. Quantitative comparisons against reference networks and mainstream hydrographic products demonstrate that the method achieves the highest positional accuracy and network continuity, with errors mainly focused within a 0–40 m range. Significant improvements are primarily for narrow tributaries, highly meandering rivers, and braided channels. The experiments demonstrate that the proposed method provides a reliable solution for high-resolution river network mapping in complex environments. Full article

Effective and transparent medical diagnosis relies on accurate and interpretable classification of medical images across multiple modalities. This paper introduces an explainable multi-modal image analysis framework based on a dual-stream architecture that fuses handcrafted descriptors with deep features extracted from a custom MobileNet. Handcrafted descriptors include frequency-domain and texture features, while deep features are summarized using 26 statistical metrics to enhance interpretability. In the fusion stage, complementary features are combined at both the feature and decision levels. Decision-level integration combines calibrated soft voting, weighted voting, and stacking ensembles with optimized classifiers, including decision trees, random forests, gradient boosting, and logistic regression. To further refine performance, a hybrid class-specific feature selection strategy is proposed, combining mutual information, recursive elimination, and random forest importance to select the most discriminative features for each class. This hybrid selection approach eliminates redundancy, improves computational efficiency, and ensures robust classification. Explainability is provided through Local Interpretable Model-Agnostic Explanations, which offer transparent details about the ensemble model’s predictions and link influential handcrafted features to clinically meaningful image characteristics. The framework is validated on three benchmark datasets, i.e., BTTypes (brain MRI), Ultrasound Breast Images, and ACRIMA Retinal Fundus Images, demonstrating generalizability across modalities (MRI, ultrasound, retinal fundus) and disease categories (brain tumor, breast cancer, glaucoma). Full article

Accurate short-lead river discharge forecasting is critical for effective flood risk mitigation in small mountainous basins, where rapid hydrological responses pose significant challenges. In this study, we focus on the Fuhu Stream in Emeishan City, China, and utilize high-resolution 5-min time series of upstream precipitation, stage, and discharge to predict downstream flow. We benchmark three data-driven models—SARIMAX, XGBoost, and LSTM—using a dataset spanning from 7 June 2024 to 25 October 2024. The data were split chronologically, with observations from October 2024 reserved exclusively for testing to ensure rigorous out-of-sample evaluation. Lagged correlation analysis was employed to estimate travel times from upstream to the basin outlet and to inform the selection of time-lagged input features for model training. Results during the test period demonstrate that the LSTM model significantly outperformed both XGBoost and SARIMAX across all regression metrics: it achieved the highest coefficient of determination (R² = 0.994) and the lowest prediction errors (RMSE = 0.016, MAE = 0.011). XGBoost exhibited moderate performance, while SARIMAX showed a tendency toward mean reversion and failed to capture low-flow variability. Accuracy evaluation reveals that LSTM accurately reproduced both baseflow conditions and multiple flood peaks, whereas XGBoost and SARIMAX failed. These results highlight the advantage of sequence-learning architectures in modeling nonlinear hydrological propagation and memory effects in short-term discharge dynamics. Feature importance analysis indicates that the LSTM model was highly effective for real-time forecasting and that the WSQ/LY sites served as critical monitoring nodes for achieving reliable predictions. This research contributes to the operationalization of early warning systems and provides support for decision-making regarding downstream flood disaster prevention. Full article

This study presents a systematic airfoil optimization framework to enhance the hydrodynamic performance of vertical-axis tidal turbines (VATTs) under low-flow conditions. The integrated methodology combines parameterized design, response surface methodology (RSM) optimization, and high-fidelity computational fluid dynamics (CFD) validation to investigate the effects of maximum thickness (Factor A), maximum thickness position (Factor B), and maximum camber (Factor C). The shear stress transport (SST) k-ω turbulence model was employed for flow simulation, with experimental validation conducted across Reynolds numbers from 5.2 × 10⁵ to 8.6 × 10⁵. The tip speed ratio (TSR) predictions demonstrated excellent agreement with experimental measurements, showing a maximum relative error of only 4.5%. From hundreds of Pareto-optimal solutions, five candidate designs were selected for high-fidelity verification. The final optimized airfoil (Optimized Foil 5) achieved a power coefficient (C_P) of 0.1887, representing a 27.5% improvement over the baseline National Advisory Committee for Aeronautics (NACA) 2414 airfoil. This optimal configuration features 23.51% maximum thickness, 30.14% maximum thickness position, and 3.99% maximum camber, with only 0.2% deviation between RSM prediction and CFD validation. The research establishes a reliable design framework for VATTs operating in low-velocity tidal streams, providing significant potential for harnessing previously uneconomical marine energy resources. Full article

Detecting small objects from UAV-captured aerial imagery is a critical yet challenging task, hindered by factors such as small object size, complex backgrounds, and subtle inter-class differences. Single-modal methods lack the robustness for all-weather operation, while existing multimodal solutions are often too computationally expensive for deployment on resource-constrained UAVs. To this end, we propose TAF-YOLO, a lightweight and efficient multimodal detection framework designed to balance accuracy and efficiency. First, we propose an early fusion module, the Two-branch Adaptive Fusion Network (TAFNet), which adaptively integrates visible and infrared information at both pixel and channel levels before the feature extractor, maximizing complementary data while minimizing redundancy. Second, we propose a Large Adaptive Selective Kernel (LASK) module that dynamically expands the receptive field using multi-scale convolutions and spatial attention, preserving crucial details of small objects during downsampling. Finally, we present an optimized feature neck architecture that replaces PANet’s bidirectional path with a more efficient top-down pathway. This is enhanced by a Dual-Stream Attention Bridge (DSAB) that injects high-level semantics into low-level features, improving localization without significant computational overhead. On the VEDAI benchmark, TAF-YOLO achieves 67.2% mAP₅₀, outperforming the CFT model by 2.7% and demonstrating superior performance against seven other YOLO variants. Our work presents a practical and powerful solution that enables real-time, all-weather object detection on resource-constrained UAVs. Full article

Sensor networks generate high-dimensional temporally dependent data across healthcare, environmental monitoring, and energy management, which demands robust machine learning for reliable forecasting. While gradient boosting methods have emerged as powerful tools for sensor-based regression, systematic evaluation under realistic deployment conditions remains limited. This work provides a comprehensive review and empirical benchmark of boosting algorithms spanning classical methods (AdaBoost and GBM), modern gradient boosting frameworks (XGBoost, LightGBM, and CatBoost), and adaptive extensions for streaming data and hybrid architectures. We conduct rigorous cross-domain evaluation on continuous glucose monitoring, urban air-quality forecasting, and building-energy prediction, assessing not only predictive accuracy but also robustness under sensor degradation, temporal generalization through proper time-series validation, feature-importance stability, and computational efficiency. Our analysis reveals fundamental trade-offs challenging conventional assumptions. Algorithmic sophistication yields diminishing returns when intrinsic predictability collapses due to exogenous forcing. Random cross-validation (CV) systematically overestimates performance through temporal leakage, with magnitudes varying substantially across domains. Calibration drift emerges as the dominant failure mode, causing catastrophic degradation across all the static models regardless of sophistication. Importantly, feature-importance stability does not guarantee predictive reliability. We synthesize the findings into actionable guidelines for algorithm selection, hyperparameter configuration, and deployment strategies while identifying critical open challenges, including uncertainty quantification, physics-informed architectures, and privacy-preserving distributed learning. Full article

Blockchain-based financial ecosystems generate unprecedented volumes of multi-temporal data streams requiring sophisticated analytical frameworks that leverage both on-chain transaction patterns and off-chain market microstructure dynamics. This study presents an empirical evaluation of a two-class confidence-threshold framework for cryptocurrency direction prediction, systematically integrating macro momentum indicators with microstructure dynamics through unified feature engineering. Building on established selective classification principles, the framework separates directional prediction from execution decisions through confidence-based thresholds, enabling explicit optimization of precision–recall trade-offs for decentralized financial applications. Unlike traditional three-class approaches that simultaneously learn direction and execution timing, our framework uses post-hoc confidence thresholds to separate these decisions. This enables systematic optimization of the accuracy-coverage trade-off for blockchain-integrated trading systems. We conduct comprehensive experiments across 11 major cryptocurrency pairs representing diverse blockchain protocols, evaluating prediction horizons from 10 to 600 min, deadband thresholds from 2 to 20 basis points, and confidence levels of 0.6 and 0.8. The experimental design employs rigorous temporal validation with symbol-wise splitting to prevent data leakage while maintaining realistic conditions for blockchain-integrated trading systems. High confidence regimes achieve peak profits of 167.64 basis points per trade with directional accuracies of 82–95% on executed trades, suggesting potential applicability for automated decentralized finance (DeFi) protocols and smart contract-based trading strategies on similar liquid cryptocurrency pairs. The systematic parameter optimization reveals fundamental trade-offs between trading frequency and signal quality in blockchain financial ecosystems, with high confidence strategies reducing median coverage while substantially improving per-trade profitability suitable for gas-optimized on-chain execution. Full article

In order to explore the co-evolutionary relationship between the functions of microbial communities and the chemical composition of groundwater in a hyporheic zone affected by groundwater exploitation along riverbank, we have taken the Huangjia water source area on the Liao River main stream in Shenyang as an example. DNA was extracted from microorganisms in the hyporheic zone affected by groundwater exploitation along the riverbank, and we conducted high-throughput sequencing to select the dominant bacterial strains from the indigenous bacteria. They are classified as the Proteobacteria phylum, the Actinobacteria phylum, the Firmicutes phylum, the Bacteroidetes phylum, the Chloroflexi phylum, and the Acidobacteria phylum. The dominant bacteria have a good correlation with Fe, Mn, and environmental factors (such as DO—dissolved oxygen, Eh—oxidation-reduction potential, etc.) in the hyporheic zone. The functions and activities of the superior bacterial strains exhibit a feature of co-evolution with the water’s chemical environment, which has certain response characteristics to redox zoning. Studying the co-evolution relationship between the microbial community structure and function in the hyporheic zone and the chemical composition of the groundwater can provide a microbiological theoretical basis for the redox zonation. It also offers reference for understanding the process of Fe and Mn migration and transformation in the hyporheic zone under the hydrodynamic conditions of groundwater exploitation along the riverbank. Full article

Wireless connectivity plays a pivotal role in enabling real-time telemetry, sensor feedback, and autonomous navigation within Industry 4.0 environments. This paper presents a ROS 2-based mobile robotic platform designed to perform real-time network diagnostics across both private 5G and Wi-Fi technologies in a live smart manufacturing testbed. The system integrates high-frequency telemetry acquisition with spatial localization, multi-protocol connection analysis, and detailed performance monitoring. Metrics such as latency, packet loss, bandwidth, and IIoT (Industrial Internet of Things) data stream health are continuously logged and analysed. Telemetry is captured during motion and synchronously stored in an InfluxDB time-series database, enabling live visualization through Grafana dashboards. A key feature of the platform is its dual-path transmission architecture, which provides communication redundancy and allows side-by-side evaluation of network behaviour under identical physical conditions. Experimental trials demonstrate the platform’s ability to detect roaming events, characterize packet loss, and reveal latency differences between Wi-Fi and 5G networks. Results show that Wi-Fi suffered from roaming-induced instability and packet loss, whereas 5G maintained stable and uninterrupted connectivity throughout the test area. This work introduces a modular, extensible framework for mobile network evaluation in industrial settings and provides practical insights for infrastructure tuning, protocol selection, and wireless fault detection. Full article

Teleoperation in agricultural environments requires enhanced situational awareness for effective architectural scouting and decision-making for orchard management applications. The dynamic complexity of orchard structures presents challenges for remote visualization during architectural scouting operations. This study presents an adaptive streaming and rendering pipeline for real-time point cloud visualization in Virtual Reality (VR) teleoperation systems. The proposed method integrates selective streaming that localizes teleoperators within live maps, an efficient point cloud parser for Unity Engine, and an adaptive Level-of-Detail rendering system utilizing dynamically scaled and smoothed polygons. The implementation incorporates pseudo-coloring through LiDAR reflectivity fields to enhance the distinction between materials and geometry. The pipeline was evaluated using datasets containing LiDAR point cloud scans of orchard environments captured during spring and summer seasons, with testing conducted on both standalone and PC-tethered VR configurations. Performance analysis demonstrated improvements of 10.2–19.4% in runtime performance compared to existing methods, with a framerate enhancement of up to 112% achieved through selectively streamed representations. Qualitative assessment confirms the method’s capability to maintain visual continuity at close proximity while preserving the geometric features discernible for architectural scouting operations. The results establish the viability of VR-based teleoperation for precision agriculture applications, while demonstrating the critical relationship between Quality-of-Service parameters and operator Quality of Experience in remote environmental perception. Full article

The unequal status between manufacturers and live-streamers often undermines supply chain profitability and social welfare. However, the “volume guarantee” commission mode, designed to mitigate this issue, has proven ineffective in practice. This paper adopts a Nash bargaining fairness framework to analyze this paradox, incorporating two defining features of live-streaming commerce: the social network effect and the streamer’s cost of purchasing public domain traffic. We develop a dynamic game model involving the platform, manufacturer, streamer, and consumers to examine commission mode selection and supply chain decision-making. Our analysis yields four key findings: (1) Under Nash bargaining fairness, the “volume guarantee” mode is invariably redundant, regardless of who sets the sales threshold. Bargaining power only influences profit distribution via commission rates without distorting optimal product pricing or traffic acquisition decisions. (2) The social network effect boosts product prices, traffic purchases, total profit, and social welfare, with its impact amplified by the streamer’s fanbase size. Thus, collaborating with top-streamers is advantageous for manufacturers. (3) While higher platform traffic costs do not affect the optimal product price, they reduce traffic purchase volume, thereby decreasing supply chain profits and social welfare. (4) To enhance social welfare, platforms can implement differentiated traffic pricing, offering discounts to top-streamers. This study provides critical managerial insights for designing fair contracts and fostering equitable cooperation in live-streaming ecosystems. Full article

Flood risk in small streams is rising under climate change, as small catchments are highly vulnerable to short, intense storms. We develop a high-resolution assessment that integrates a Digital Surface Model (DSM), a Digital Elevation Model (DEM), and airborne LiDAR within a MATLAB (2025b) hydraulic workflow. A hybrid elevation model uses the DEM as baseline and selectively retains DSM-derived structures (levees, bridges, embankments), while filtering vegetation via DSM–DEM differencing with a 1.0 m threshold and a 2-pixel kernel. We simulate 10-, 30-, 50-, 100-, and 200-year return periods and calibrate the 200-year case to the July 2025 Sancheong event (793.5 mm over 105 h; peak 100 mm h⁻¹). The hybrid approach improves predictions over DEM-only runs, capturing localized depth increases of 1.5–2.0 m behind embankments and reducing false positives in vegetated areas by 12–18% relative to raw DSM use. Multi-frequency maps show progressive expansion of inundation; in the 100-year scenario, 68% of the inundated area exceeds 2.0 m depth, while 0–1.0 m zones comprise only 13% of the footprint. Unlike previous DSM–DEM studies, this work introduces a selective integration approach that distinguishes structural and vegetative features to improve the physical realism of small-stream flood modeling. This transferable framework supports climate adaptation, emergency response planning, and sustainable watershed management in small-stream basins. Full article

Healthcare systems worldwide have faced unprecedented pressure during crises such as the COVID-19 pandemic, exposing limits in managing scarce hospital resources. Many predictive models remain static, unable to adapt to new variants, shifting conditions, or diverse patient populations. This work proposes a dynamic prioritization framework that recalculates severity scores in batch mode when new factors appear and applies them instantly through a streaming pipeline to incoming patients. Unlike approaches focused only on fixed mortality or severity risks, our model integrates dual datasets (survivors and non-survivors) to refine feature selection and weighting, enhancing robustness. Built on a big data infrastructure (Spark/Databricks), it ensures scalability and responsiveness, even with millions of records. Experimental results confirm the effectiveness of this architecture: The artificial neural network (ANN) achieved 98.7% accuracy, with higher precision and recall than traditional models, while random forest and logistic regression also showed strong AUC values. Additional tests, including temporal validation and real-time latency simulation, demonstrated both stability over time and feasibility for deployment in near-real-world conditions. By combining adaptability, robustness, and scalability, the proposed framework offers a methodological contribution to healthcare analytics, supporting fair and effective hospitalization prioritization during pandemics and other public health emergencies. Full article

The widespread integration of Internet-connected devices into industrial environments has enhanced connectivity and automation but has also increased the exposure of industrial cyber–physical systems to security threats. Detecting anomalies is essential for ensuring operational continuity and safeguarding critical assets, yet the dynamic, real-time nature of such data poses challenges for developing effective defenses. This paper introduces DataSense, a comprehensive dataset designed to advance security research in industrial networked environments. DataSense contains synchronized sensor and network stream data, capturing interactions among diverse industrial sensors, commonly used connected devices, and network equipment, enabling vulnerability studies across heterogeneous industrial setups. The dataset was generated through the controlled execution of 50 realistic attacks spanning seven major categories: reconnaissance, denial of service, distributed denial of service, web exploitation, man-in-the-middle, brute force, and malware. This process produced a balanced mix of benign and malicious traffic that reflects real-world conditions. To enhance its utility, we introduce an original feature selection approach that identifies features most relevant to improving detection rates while minimizing resource usage. Comprehensive experiments with a broad spectrum of machine learning and deep learning models validate the dataset’s applicability, making DataSense a valuable resource for developing robust systems for detecting anomalies and preventing intrusions in real time within industrial environments. Full article