Search Results (6,240)

The authentication of wood species is of paramount significance to market regulation and product quality control in the construction industry. While classification based on microscopic wood cell structures serves as a critical reference for this task, the high inter-class similarity of cell structures and the inherent biological anisotropy of fine textures pose significant challenges to existing methods. Due to their generic design, standard deep learning models often struggle to capture these fine-grained directional features and suffer from catastrophic feature loss during global pooling, particularly under limited sample conditions. To bridge this gap between general-purpose architectures and the specific demands of wood texture analysis, this paper proposes CM-EffNet, a lightweight fine-grained classification framework based on an improved EfficientNetV2 architecture. Firstly, a data augmentation strategy is employed to expand a collected dataset of 226 wood species from 3673 to 29,384 images, effectively mitigating overfitting caused by small sample sizes. Secondly, a Coordinate Attention (CA) mechanism is integrated to embed positional information into channel attention. This allows the network to precisely capture long-range dependencies between cell walls and vessels cavities, successfully decoding the challenge of textural anisotropy. Thirdly, a MixPooling strategy is introduced to replace traditional global average pooling, enabling the collaborative extraction of background context and salient fine-grained details to prevent the loss of critical micro-features. Systematic experiments demonstrate that CM-EffNet achieves a recognition accuracy of 96.72% and a training accuracy of 98.18%. Comparative results confirm that the proposed model offers superior learning efficiency and classification precision with a compact parameter size. This makes it highly suitable for deployment on mobile terminals connected to portable microscopes, providing a rapid and accurate solution for on-site timber market regulation and industrial quality control. Full article

Drowsy driving and sudden medical emergencies are major contributors to traffic accidents, necessitating continuous, non-intrusive driver monitoring. Since current technologies often struggle to balance accuracy with practicality, this study presents the design, fabrication, and validation of a smart steering wheel prototype. The device integrates dry-contact electrocardiogram (ECG), photoplethysmography (PPG), and inertial sensors to facilitate multimodal physiological monitoring. The system underwent a two-stage evaluation involving a single participant: laboratory validation benchmarking acquired signals against medical-grade equipment, followed by real-world testing in a custom electric research vehicle to assess performance under dynamic conditions. Laboratory results demonstrated that the prototype captured high-quality signals suitable for reliable heart rate variability analysis. Furthermore, on-road evaluation confirmed the system’s operational functionality; despite increased noise from motion artifacts, the ECG signal remained sufficiently robust for continuous R-peak detection. These findings confirm that the multimodal smart steering wheel is a feasible solution for unobtrusive driver monitoring. This integrated platform provides a solid foundation for developing sophisticated machine-learning algorithms to enhance road safety by predicting fatigue and detecting adverse health events. Full article

Electromagnetic flowmeters face significant challenges in measuring complex fluids, characterized by weak flow signals and severe noise interference. Conventional solutions, such as dual-frequency rectangular wave excitation, suffer from multiple drawbacks including rich harmonic components, high electromagnetic noise during switching transitions, a propensity for resonance which shortens stabilization time, reduced sampling windows, and complex circuit implementation. Similarly, traditional single-frequency excitation struggles to balance zero stability with the suppression of slurry noise. To address these limitations, this paper proposes a novel converter design based on dual-frequency sinusoidal wave excitation. A pure hardware circuit is used to generate the composite excitation signal, which superimposes low-frequency and high-frequency components. This approach eliminates the need for a master control chip in signal generation, thereby reducing both circuit complexity and computational resource allocation. The signal processing chain employs a technique of “high-order Butterworth separation filtering combined with synchronous demodulation,” effectively suppressing power frequency, orthogonal, and in-phase interference, achieving an improvement in interference rejection by approximately three orders of magnitude (1000×). Experimental results show that the proposed converter featured simplified circuitry, achieved a measurement accuracy of class 0.5, and validated the overall feasibility of the scheme. Full article

Due to the complexity of aerodynamic coupling between the duct and propeller, the overall design and optimization of ducted fans often require extensive experience and time. Meanwhile, traditional design methods based on Blade Element Momentum Theory, Lifting Surface Theory, Vortex Lattice Methods, and Panel Method usually exhibit certain deviations between their design results and actual outcomes. This is because these approaches struggle to accurately calculate the aerodynamic coupling effects between the duct and propeller, coupled with numerous simplifications inherent in the methods themselves. Considering the strong nonlinear coupling relationship between the duct and propeller, the Response Surface Method (RSM), which enables efficient and accurate analysis of multi-variable coupling effects, was selected for the parameter design and optimization of ducted fans. Computational Fluid Dynamics (CFD) was applied to evaluate the impact of design parameters on overall aerodynamic performance. This approach addresses the limitations of traditional methods, including low design accuracy, high computational cost, and insufficient multi–objective optimization capability. It explicitly models multi-parameter coupling and nonlinear effects using a small number of experimental points, combined with the Multi-Objective Genetic Algorithm (MOGA) to find the global optimum. Compared to the baseline duct fan, the optimized duct fan achieved a 9.6% increase in overall lift and a 9.5% improvement in lift efficiency. Full article

Accurate detection of oil and gas (O&G) facilities in high-resolution remote sensing imagery is critical for infrastructure surveillance and sustainable resource management, yet conventional detectors struggle with severe class imbalance, extreme scale variation, and arbitrary orientation. In this work, we propose OGF Oriented R-CNN (Oil and Gas Facility Detection Oriented Region-based Convolutional Neural Network), an enhanced oriented detection model derived from Oriented R-CNN that integrates three improvements: (1) O&G Loss Function, (2) Class-Aware Hard Example Mining (CAHEM) module, and (3) Feature Pyramid Network with Feature Enhancement Attention (FPNFEA). Working in synergy, they resolve the coupled challenges more effectively than any standalone fix and do so without relying on rigid one-to-one matching between modules and individual issues. Evaluated on the O&G facility dataset comprising 3039 high-resolution images annotated with rotated bounding boxes across three classes (well sites: 3006, industrial and mining lands: 692, drilling: 244), OGF Oriented R-CNN achieves a mean average precision (mAP) of 82.9%, outperforming seven state-of-the-art (SOTA) models by margins of up to 27.6 percentage points (pp) and delivering a cumulative gain of +10.5 pp over Oriented R-CNN. Full article

Fog computing is one of the rapidly growing platforms with an exponentially increasing demand for real-time data processing. The fog computing market is expected to reach USD 8358 million by the year 2030 with a compound annual growth of 50%. The wide adaptation of fog computing by the industries worldwide is due to the advantages like reduced latency, high operational efficiency, and high-level data privacy. The highly distributed and heterogeneous nature of fog computing leads to significant challenges related to resource management, data security, task scheduling, data privacy, and interoperability. The task typically represents a job generated by the IoT device. The action indicates the way of executing the tasks whose decision is taken by the scheduler. Task scheduling is one of the prominent issues in fog computing which includes the process of effectively scheduling the tasks among fog devices to effectively utilize the resources and meet the Quality of Service (QoS) requirements of the applications. Improper task scheduling leads to increased execution time, overutilization of resources, data loss, and poor scalability. Hence there is a need to do proper task scheduling to make optimal task distribution decisions in a highly dynamic resource-constrained heterogeneous fog computing environment. Flow Q learning (FQL) is a potential form of reinforcement learning algorithm which uses the flow matching policy for action distribution. It can handle complex forms of data and multimodal action distribution which make it suitable for the highly volatile fog computing environment. However, flow Q learning struggles to achieve a proper trade-off between the expressive flow model and a reduction in the Q function, as it relies on a one-step optimization policy that introduces bias into the estimated Q function value. The Tiny Language Model (TLM) is a significantly smaller form of a Large Language Model (LLM) which is designed to operate over the device-constrained environment. It can provide fair and systematic guidance to disproportionally biased deep learning models. In this paper a novel TLM guided flow Q learning framework is designed to address the task scheduling problem in fog computing. The neutrality and fine-tuning capability of the TLM is combined with the quick generable ability of the FQL algorithm. The framework is simulated using the Simcan2Fog simulator considering the dynamic nature of fog environment under finite and infinite resources. The performance is found to be good with respect to parameters like execution time, accuracy, response time, and latency. Further the results obtained are validated using the expected value analysis method which is found to be satisfactory. Full article

Data prefetching is essential for modern file storage systems operating in large-scale cloud and data-intensive environments, where high performance increasingly depends on intelligent, adaptive mechanisms. Traditional rule-based methods and recently proposed machine learning-based techniques often struggle to cope with the complex and rapidly evolving data access patterns characteristic of big-data workloads. In this paper, we introduce an online, streaming machine learning (SML) approach for predictive data prefetching that retrieves useful data into the cache ahead of time. We present a novel online training framework that extracts features in real time and continuously updates streaming ML models to learn and adapt from large and dynamic access streams. Building on this framework, we design new SML-driven prefetching algorithms that decide when, how, and what data to prefetch into the cache with minimal overhead. Extensive experiments using production traces from Huawei Technologies Inc. and Google workloads from the SNIA IOTTA repository demonstrate that our intelligent policies consistently deliver the highest byte hits among competing approaches, achieving 97% prefetch byte precision and reducing data access latency by up to 2.8 times. These results show that streaming ML can deliver immediate performance gains and offers a scalable foundation for future adaptive storage systems. Full article

Lakes on the Qinghai–Tibet Plateau are important indicators of global climate change, and variations in their water storage strongly influence regional hydrological cycles and ecosystems. However, existing studies have largely focused on relative changes in lake volume, while the precise quantification of absolute water storage remains insufficient, largely due to the lack of long-term, high-accuracy water storage time series. Constrained by harsh natural conditions and limited in situ observations, conventional approaches struggle to achieve the accurate long-term monitoring of lake water storage across the Plateau. To address this challenge, we propose a DEM-based underwater topography extrapolation method. Under the assumption of continuity between surrounding onshore terrain and submerged lakebed morphology, nearshore DEM data are extrapolated to reconstruct lake bathymetry. By integrating multi-source remote sensing observations of lake area and water level, we estimate and reconstruct 30-year absolute water storage time series for 120 Plateau lakes larger than 50 km². This method does not require measured water depth data and is particularly suitable for data-scarce, topographically complex, high-altitude lake regions, effectively overcoming key limitations of conventional methods used for absolute water storage monitoring. Validation shows strong agreement between our estimates and an independent validation dataset, with an overall correlation coefficient of 0.95; the reconstructed time series are highly reliable, with correlation coefficients exceeding 0.6. During the study period, the total lake water storage of the Qinghai–Tibet Plateau exhibited a significant increasing trend, with a cumulative growth of approximately 137.297 billion m³, representing a 20.73% increase, and showing notable spatial heterogeneity. The water storage dataset constructed in this study provides reliable data support for research on water cycles, climate change assessment, and regional water resource management on the Qinghai–Tibet Plateau. Full article

Undergraduate civil engineering students frequently struggle to transition from deterministic to probabilistic reasoning, a conceptual shift essential for modern structural design practice governed by reliability-based codes. This paper presents a design-based research (DBR) contribution and a theoretically grounded pedagogical framework that integrates AI-powered conversational tutoring with interactive simulations to scaffold this transition. The framework synthesizes cognitive load theory, scaffolding principles, self-regulated learning research, and threshold concepts theory. The design incorporates three novel elements: (1) a structured misconception inventory specific to structural reliability, derived from literature and expert elicitation, with each misconception linked to targeted intervention strategies; (2) an integration architecture connecting large language model tutoring with domain-specific simulations, where simulation states inform tutoring and misconception detection triggers targeted activities; and (3) a scaffolded module sequence building systematically from deterministic foundations through probability concepts to reliability analysis methods. Sequential modules progress from uncertainty recognition through Monte Carlo simulation and design applications. We provide technical specifications for the implementation of AI tutoring, including prompt engineering strategies, accuracy safeguards that address known limitations of large language models (LLMs), and protocols for escalation to human instructors. An assessment framework specifies concept inventory items, process measures, and practical competence tasks. Ultimately, this paper provides testable conjectures and identifies conditions under which the framework might fail, structuring subsequent empirical validation with student participants following institutional ethics approval. Full article

Existing video inpainting methods typically combine optical flow propagation with Transformer architectures, achieving promising inpainting results. However, they lack adaptive inpainting strategy optimization in diverse scenarios, and struggle to capture high-level temporal semantics, causing temporal inconsistencies and quality degradation. To address these challenges, we make one of the first attempts to introduce reinforcement learning into the video inpainting domain, establishing a closed-loop framework named CLIP-RL that enables adaptive strategy optimization. Specifically, video inpainting is reformulated as an agent–environment interaction, where the inpainting module functions as the agent’s execution component, and a pre-trained inpainting detection module provides real-time quality feedback. Guided by a policy network and a composite reward function that incorporates a weighted temporal alignment loss, the agent dynamically selects actions to adjust the inpainting strategy and iteratively refines the inpainting results. Compared to ProPainter, CLIP-RL improves PSNR from 34.43 to 34.67 and SSIM from 0.974 to 0.986 on the YouTube-VOS dataset. Qualitative analysis demonstrates that CLIP-RL excels in detail preservation and artifact suppression, validating its superiority in video inpainting tasks. Full article

Accurate global mapping of lunar iron oxide (FeO) abundance is essential for understanding the Moon’s geological evolution and for supporting future in situ resource utilization (ISRU). While hyperspectral data from the Moon Mineralogy Mapper (M3) provide a unique combination of high spectral dimensionality, hectometre-scale spatial resolution, and near-global coverage, existing FeO retrieval approaches struggle to fully exploit the high dimensionality, nonlinear spectral variability, and planetary-scale volume of the Global Mode dataset. To address these limitations, we present an integrated machine learning pipeline for estimating lunar FeO abundance from M3 hyperspectral observations. Unlike traditional methods based on raw reflectance or empirical spectral indices, the proposed framework combines Discrete Wavelet Transform (DWT), deep autoencoder-based feature compression, and ensemble regression to achieve robust and scalable FeO prediction. M3 spectra (83 bands, 475–3000 nm) are transformed using a Daubechies-4 (db4) DWT to extract 42 representative coefficients per pixel, capturing the dominant spectral information while filtering high-frequency noise. These features are further compressed into a six-dimensional latent space via a deep autoencoder and used as input to a Random Forest regressor, which outperforms kernel-based and linear Support Vector Regression (SVR) as well as Lasso regression in predictive accuracy and stability. The proposed model achieves an average prediction error of 1.204 wt.% FeO and demonstrates consistent performance across diverse lunar geological units. Applied to 806 orbital tracks (approximately $3.5\times {10}^9$ pixels), covering more than 95% of the lunar surface, the pipeline produces a global FeO abundance map at 150 m per pixel resolution. These results demonstrate the potential of integrating multiscale wavelet representations with nonlinear feature learning to enable large-scale, geochemically constrained planetary mineral mapping. Full article

Rainfall infiltration is one of the main factors inducing slope instability, while the spatial heterogeneity and uncertainty of soil parameters have profound impacts on slope response characteristics and stability evolution. Traditional deterministic analysis methods struggle to reveal the dynamic risk evolution process of the system under heavy rainfall. Therefore, this paper proposes an uncertainty analysis framework combining Karhunen–Loève Expansion (KLE) random field theory, Polynomial Chaos Kriging (PCK) surrogate modeling, and Monte Carlo simulation to efficiently quantify the probabilistic characteristics and spatial risks of rainfall-induced slope instability. First, for key strength parameters such as cohesion and internal friction angle, a two-dimensional random field with spatial correlation is constructed to realistically depict the regional variability of soil mechanical properties. Second, a PCK surrogate model optimized by the LARS algorithm is developed to achieve high-precision replacement of finite element calculation results. Then, large-scale Monte Carlo simulations are conducted based on the surrogate model to obtain the probability distribution characteristics of slope safety factors and potential instability areas at different times. The research results show that the slope enters the most unstable stage during the middle of rainfall (36–54 h), with severe system response fluctuations and highly concentrated instability risks. Deterministic analysis generally overestimates slope safety and ignores extreme responses in tail samples. The proposed method can effectively identify the multi-source uncertainty effects of slope systems, providing theoretical support and technical pathways for risk early warning, zoning design, and protection optimization of slope engineering during rainfall periods. Full article

To address the lack of inertia in full-power converter wind turbines and the inability of existing mechanical speed regulation technologies to achieve power smoothing without converters, this paper proposes a novel hybrid wind energy storage system integrating a Continuously Variable Transmission (CVT) and an electromechanical flywheel. This system establishes a cascaded topology featuring “CVT-based source-side speed regulation and electromechanical flywheel-based terminal power stabilization.” By utilizing the CVT for speed decoupling and introducing the flywheel via a planetary differential branch, the system retains physical inertia by eliminating large-capacity converters and overcomes the bottleneck of traditional mechanical transmissions, which struggle to balance constant frequency with stable power output. Simulation results demonstrate that the proposed system reduces the active power fluctuation range by 47.60% compared to the raw wind power capture. Moreover, the required capacity of the auxiliary motor is only about 15% of the rated power, reducing the reliance on power electronic converters by approximately 85% compared to full-power converter systems. Furthermore, during a grid voltage dip of 0.6 p.u., the system restricts rotor speed fluctuations to within 0.5%, significantly enhancing Low Voltage Ride-Through (LVRT) capability. Full article

Deep learning-based network traffic anomaly detection, particularly using Recurrent Neural Networks (RNNs), often struggles with high computational overhead and difficulties in capturing long-range temporal dependencies. To address these limitations, this paper proposes a Bidirectional Temporal Attention Convolutional Network (Bi-TACN) for robust and efficient network traffic anomaly detection. Specifically, dilated causal convolutions with expanding receptive fields and residual modules are employed to capture multi-scale temporal patterns while effectively mitigating the vanishing gradient. Furthermore, a bidirectional structure integrated with Efficient Channel Attention (ECA) is designed to adaptively weight contextual features, preventing sparse attack indicators from being overwhelmed by dominant normal traffic. A Softmax-based classifier then leverages these refined representations to execute high-performance anomaly detection. Extensive experiments on the NSL-KDD and UNSW-NB15 datasets demonstrate that Bi-TACN achieves average accuracies of 88.51% and 82.5%, respectively, significantly outperforming baseline models such as Bi-TCN and Bi-GRU in terms of both precision and convergence speed. Full article

Accurate identification of the combustion state in municipal solid waste incineration (MSWI) processes is crucial for achieving efficient, low-emission, and safe operation. However, existing methods often struggle with stable and reliable recognition due to insufficient feature extraction capabilities when confronted with challenges such as complex flame morphology, blurred boundaries, and significant noise in flame images. To address this, this paper proposes a novel hybrid architecture model named PRTNet, which aims to enhance the accuracy and robustness of combustion state recognition through multi-scale feature enhancement and adaptive fusion mechanisms. First, a local-semantic enhanced residual network is constructed to establish spatial correlations between fine-grained textures and macroscopic combustion patterns. Subsequently, a feature-adaptive fusion Transformer is designed, which models long-range dependencies and high-frequency details in parallel via deformable attention and local convolutions, and achieves adaptive fusion of global and local features through a gating mechanism. Finally, a cross-scale feature guided aggregation module is proposed to fuse shallow detailed information with deep semantic features under dual-attention guidance. Experiments conducted on a flame image dataset from an MSWI plant in Beijing show that PRTNet achieves an accuracy of 96.29% in the combustion state classification task, with precision, recall, and F1-score all exceeding 96%, significantly outperforming numerous mainstream baseline models. Ablation studies further validate the effectiveness and synergistic effects of each module. The proposed method provides a reliable solution for intelligent flame state recognition in complex industrial scenarios, contributing to the advancement of intelligent and sustainable development in municipal solid waste incineration processes. Full article