Search Results (12,147)

With the strict requirements of national policies on truck dimensions, axle loads, and weight limits, along with the implementation of tolls based on vehicle types, rapid and accurate identification of vehicle axle types has become essential for toll station management. To address the limitations of existing methods in distinguishing between drive and driven axles, complex equipment setup, and image evidence retention, this article proposes a panoramic image detection technology for vehicle chassis based on enhanced ORB and YOLOv11. A portable vehicle chassis image acquisition system, based on area array cameras, was developed for rapid on-site deployment within 20 min, eliminating the requirement for embedded installation. The FeatureBooster (FB) module was employed to optimize the ORB algorithm’s feature matching, and combined with keyframe technology to achieve high-quality panoramic image stitching. After fine-tuning the FB model on a domain-specific area scan dataset, the number of feature matches increased to 151 ± 18, substantially outperforming both the pre-trained FB model and the baseline ORB. Experimental results on axle type recognition using the YOLOv11 algorithm combined with ORB and FB features demonstrated that the integrated approach achieved superior performance. On the overall test set, the model attained an mAP@50 of 0.989 and an mAP@50:95 of 0.780, along with a precision (P) of 0.98 and a recall (R) of 0.99. In nighttime scenarios, it maintained an mAP@50 of 0.977 and an mAP@50:95 of 0.743, with precision and recall both consistently at 0.98 and 0.99, respectively. The field verification shows that the real-time and accuracy of the system can provide technical support for the axle type recognition of toll stations. Full article

We address surrogate-assisted multi-objective optimization for computationally expensive structural designs. The testbed is an axisymmetric laminated composite shell whose geometry, ply angles, and plywise materials are optimized to simultaneously (i) maximize separation of selected natural frequencies from a known excitation and (ii) minimize material cost. To reduce high-fidelity (HF) finite element evaluations, we develop a deep neural network surrogate framework with three variants: an HF-only baseline; a multi-fidelity (MF) pipeline using an auxiliary refinement network to convert abundant low-fidelity (LF) data into pseudo-HF labels for a single-fidelity evaluator; and a cascaded ensemble that emulates HF responses and then maps them to pseudo-experimental targets. During optimization, only surrogates are queried—no FEM calls—while final designs are verified by FEM. Pareto-front quality is quantified primarily by a normalized relative hypervolume indicator computed against an envelope approximation of the True Pareto Front, complemented where appropriate by standard indicators. A controlled training protocol and common validation regime isolate the effect of fidelity strategy from architectural choices. Results show that MF variants markedly reduce HF data requirements and improve Pareto-front quality over the HF-only baseline, offering a practical route to scalable, accurate design under strict computational budgets. Full article

To address the technical challenges in the educational domain named entity recognition (NER), such as ambiguous entity boundaries and difficulties with nested entity identification, this study proposes an enhanced semantic BERT model (ES-BERT). The model innovatively adopts an education domain, vocabulary-assisted semantic enhancement strategy that (1) applies the term frequency–inverse document frequency (TF-IDF) algorithm to weight domain-specific terms, and (2) fuses the weighted lexical information with character-level features, enabling BERT to generate enriched, domain-aware, character–word hybrid representations. A complete bidirectional long short-term memory-conditional random field (BiLSTM-CRF) recognition framework was established, and a novel focal loss-based joint training method was introduced to optimize the process. The experimental design employed a three-phase validation protocol, as follows: (1) In a comparative evaluation using 5-fold cross-validation on our proprietary computer-education dataset, the proposed ES-BERT model yielded a precision of 90.38%, which is higher than that of the baseline models; (2) Ablation studies confirmed the contribution of domain-vocabulary enhancement to performance improvement; (3) Cross-domain experiments on the 2016 knowledge base question answering datasets and resume benchmark datasets demonstrated outstanding precision of 98.41% and 96.75%, respectively, verifying the model’s transfer-learning capability. These comprehensive experimental results substantiate that ES-BERT not only effectively resolves domain-specific NER challenges in education but also exhibits remarkable cross-domain adaptability. Full article

We investigate the applicability of visual foundation models, a recent advancement in artificial intelligence, for archaeological remote sensing. In contrast to earlier approaches, we employ a strictly zero-shot methodology, testing the hypothesis that such models can perform archaeological feature detection without any fine-tuning or other adaptation for the remote sensing domain. Across five experiments using satellite imagery, aerial LiDAR, and drone video data, we assess the models’ ability to detect archaeological features. Our results demonstrate that such foundation models can achieve detection performance comparable to that of human experts and established automated methods. A key advantage lies in the substantial reduction of required human effort and the elimination of the need for training data. To support reproducibility and future experimentation, we provide open-source scripts and datasets and suggest a novel workflow for remote sensing projects. If current trends persist, foundation models may offer a scalable and accessible alternative to conventional archaeological prospection. Full article

This paper presents HE-DMDeception, a novel adversarial attack network that integrates human visual deception with deep model deception to enhance the security of 3D object detection. Existing patch-based and camouflage methods can mislead deep learning models but struggle to generate visually imperceptible, high-quality textures. Our framework employs a CycleGAN-based camouflage network to generate highly camouflaged background textures, while a dedicated deception module disrupts non-maximum suppression (NMS) and attention mechanisms through optimized constraints that balance attack efficacy and visual fidelity. To overcome the scarcity of annotated vehicle data, an image segmentation module based on the pre-trained Segment Anything (SAM) model is introduced, leveraging a two-stage training strategy combining semi-supervised self-training and supervised fine-tuning. Experimental results show that the minimum P@0.5 values (50%, 55%, 20%, 25%, 25%) were achieved by HE-DMDeception across You Only Look Once version 8 (YOLOv8), Real-Time Detection Transformer (RT-DETR), Fast Region-based Convolutional Neural Network (Faster-RCNN), Single Shot MultiBox Detector (SSD), and MaskRegion-based Convolutional Neural Network (Mask RCNN) detection models, while maintaining high visual consistency with the original camouflage. These findings demonstrate the robustness and practicality of HE-DMDeception, offering new insights into 3D object detection adversarial attacks. Full article

Flange bolts are vital fasteners in civil, mechanical, and aerospace structures, where preload stability directly affects overall safety. Conventional methods for bolt loosening detection often suffer from missed detections, weak feature representation, and insufficient cross-scale fusion under complex backgrounds. This paper presents an integrated detection and angle estimation framework using a lightweight deep learning detection network. A MobileViT backbone is employed to balance local texture with global context. In the spatial pyramid pooling stage, large separable convolutional kernels are combined with a channel and spatial attention mechanism to highlight discriminative features while suppressing noise. Together with content-aware upsampling and bidirectional multi-scale feature fusion, the network achieves high accuracy in detecting small and low-contrast targets while maintaining real-time performance. For angle estimation, the framework adopts an efficient training-free pipeline consisting of oriented FAST and rotated BRIEF feature detection, approximate nearest neighbor matching, and robust sample consensus fitting. This approach reliably removes false correspondences and extracts stable rotation components, maintaining success rates between 85% and 93% with an average error close to one degree, even under reflection, blur, or moderate viewpoint changes. Experimental validation demonstrates strong stability in detection and angular estimation under varying illumination and texture conditions, with a favorable balance between computational efficiency and practical applicability. This study provides a practical, intelligent, and deployable solution for bolt loosening detection, supporting the safe operation of large-scale equipment and infrastructure. Full article

Fabric flattening is a critical pre-processing step for automated garment manufacturing. Most existing approaches employ single-arm robotic systems that act at a single contact point. Due to the nonlinear and deformable dynamics of fabric, such systems often require multiple actions to achieve a fully flattened state. This study introduces a dual-arm fabric-flattening method based on a cascaded Proposal–Action network with a hybrid training framework. The PA network is first trained through imitation learning from human demonstrations and is subsequently refined through reinforcement learning with real-world flattening feedback. Experimental results demonstrate that the hybrid training framework substantially improves the overall flattening success rate compared with a policy trained only on human demonstrations. The success rate for a single flattening operation increases from 74% to 94%, while the overall success rate improves from 82% to 100% after two rounds of training. Furthermore, the learned policy, trained exclusively on baseline fabric, generalizes effectively to fabrics with varying thicknesses and stiffnesses. The approach reduces the number of required flattening actions while maintaining a high success rate, thereby enhancing both efficiency and practicality in automated garment manufacturing. Full article

In the Internet of Vehicles (IoV), zero-day attacks pose a significant security threat. These attacks are characterized by unknown patterns and limited sample availability. Traditional anomaly detection methods often fail because they rely on oversimplified assumptions, hindering their ability to model complex normal IoV behavior. This limitation results in low detection accuracy and high false alarm rates. To overcome these challenges, we propose a novel zero-day attack detection framework based on Feature-Enhanced Inverse Mapping (IM-ZDD). The framework introduces a two-stage process. In the first stage, a feature enhancement module mitigates data scarcity by employing an innovative multi-generator, multi-discriminator Conditional GAN (CGAN) with dynamic focusing loss to generate a large-scale, high-quality synthetic normal dataset characterized by sharply defined feature boundaries. In the second stage, a learning-based inverse mapping module is trained exclusively on this synthetic data. Through adversarial training, the module learns a precise inverse mapping function, thereby establishing a compact and expressive representation of normal behavior. During detection, samples that cannot be effectively mapped are identified as attacks. Experimental results on the F2MD platform show IM-ZDD achieves superior accuracy and a low false alarm rate, yielding an average AUC of 98.25% and F1-Score of 96.41%, surpassing state-of-the-art methods by up to 4.4 and 10.8 percentage points. Moreover, with a median detection latency of only 3 ms, the framework meets real-time requirements, providing a robust solution for zero-day attack detection in data-scarce IoV environments. Full article

Diesel engines power much of the heavy-duty equipment used in underground mines, where exhaust emissions pose acute environmental and occupational health challenges. However, predicting the amount of air required to dilute these emissions is difficult because exhaust mass flow and pollutant concentrations vary nonlinearly with multiple operating parameters. We apply deep learning to predict the total exhaust mass flow and carbon monoxide (CO) concentration of a six-cylinder gas–diesel (dual-fuel) turbocharged KAMAZ 910.12-450 engine under controlled operating conditions. We trained artificial neural networks on the preprocessed experimental dataset to capture nonlinear relationships between engine inputs and exhaust responses. Model interpretation with Shapley additive explanations (SHAP) identifies torque, speed, and boost pressure as dominant drivers of exhaust mass flow, and catalyst pressure, EGR rate, and boost pressure as primary contributors to CO concentration. In addition, symbolic regression yields an interpretable analytical expression for exhaust mass flow, facilitating interpretation and potential integration into control. The results indicate that deep learning enables accurate and interpretable prediction of key exhaust parameters in dual-fuel engines, supporting emission assessment and mitigation strategies relevant to underground mining operations. These findings support future integration with ventilation models and real-time monitoring frameworks. Full article

This paper focuses on the efficient fine-tuning of large language models within the federated learning framework. To address the performance bottlenecks caused by multi-source heterogeneity and structural inconsistency, a structure-aware federated fine-tuning method is proposed. The method incorporates a graph representation module (GRM) to model internal structural relationships within text and employs a segmentation mechanism (SM) to reconstruct and align semantic structures across inputs, thereby enhancing structural robustness and generalization under non-IID (non-Independent and Identically Distributed) settings. During training, the method ensures data locality and integrates structural pruning with gradient encryption (SPGE) strategies to balance privacy preservation and communication efficiency. Compared with representative federated fine-tuning baselines such as FedNLP and FedPrompt, the proposed method achieves consistent accuracy and F1-score improvements across multiple tasks. To evaluate the effectiveness of the proposed method, extensive comparative experiments are conducted across tasks of text classification, named entity recognition, and question answering, using multiple datasets with diverse structures and heterogeneity levels. Experimental results show that the proposed approach significantly outperforms existing federated fine-tuning strategies on most tasks, achieving higher performance while preserving privacy, and demonstrating strong practical applicability and generalization potential. Full article

Modern subway platforms are generally equipped with platform screen door systems to enhance safety, but the gap between the platform screen doors and train doors may cause passengers or objects to become trapped, leading to accidents. Addressing the issues of excessive parameter counts and computational complexity in existing foreign object intrusion detection algorithms, as well as false positives and false negatives for small objects, this article introduces a lightweight deep learning model based on YOLOv11n, named GA-YOLOv11. First, a lightweight GhostConv convolution module is introduced into the backbone network to reduce computational resource waste in irrelevant areas, thereby lowering model complexity and computational load. Additionally, the GAM attention mechanism is incorporated into the head network to enhance the model’s ability to distinguish features, enabling precise identification of object location and category, and significantly reducing the probability of false positives and false negatives. Experimental results demonstrate that in comparison to the original YOLOv11n model, the improved model achieves 3.3%, 3.2%, 1.2%, and 3.5% improvements in precision, recall, mAP@0.5, and mAP@0.5: 0.95, respectively. In contrast to the original YOLOv11n model, the number of parameters and GFLOPs were reduced by 18% and 7.9%, respectfully, while maintaining the same model size. The improved model is more lightweight while ensuring real-time performance and accuracy, designed for detecting foreign objects in subway platform gaps. Full article

Accurate inversion of the T₂ spectrum of shale oil reservoir fluids is crucial for reservoir evaluation. However, traditional nuclear magnetic resonance inversion methods face challenges in extracting features from multi-exponential decay signals. This study proposed an inversion method that combines autoencoder (AE) and Fourier transform, aiming to enhance the accuracy and stability of T₂ spectrum estimation for shale oil reservoirs. The autoencoder is employed to automatically extract deep features from the echo train, while the Fourier transform is used to enhance frequency domain features of multi-exponential decay information. Furthermore, this paper designs a customized weighted loss function based on a self-attention mechanism to focus the model’s learning capability on peak regions, thereby mitigating the negative impact of zero-value regions on model training. Experimental results demonstrate significant improvements in inversion accuracy, noise resistance, and computational efficiency compared to traditional inversion methods. This research provides an efficient and reliable new approach for precise evaluation of the T₂ spectrum in shale oil reservoirs. Full article

The optimal utilization of UAV-integrated photovoltaic (PV) systems demands accurate modeling that accounts for dynamic flight conditions. This paper introduces a novel computational intelligence-based framework that models the behavior of a moving PV system mounted on a UAV. A unique mathematical approach is developed to translate UAV flight dynamics, specifically roll, pitch, and yaw, into the tilt and azimuth angles of the PV module. To adaptively estimate the diode ideality factor under varying conditions, the Grey Wolf Optimization (GWO) algorithm is employed, outperforming traditional methods like Particle Swarm Optimization (PSO). Using a one-year environmental dataset, multiple machine learning (ML) models are trained to predict maximum power point (MPP) parameters for a commercial PV panel. The best-performing model, Rational Quadratic Gaussian Process Regression (RQGPR), demonstrates high accuracy and low computational cost. Furthermore, the proposed ML-based model is experimentally integrated into an incremental conductance (IC) MPPT technique, forming a hybrid MPPT controller. Hardware and experimental validations confirm the model’s effectiveness in real-time MPP prediction and tracking, highlighting its potential for enhancing UAV endurance and energy efficiency. Full article

The Internet of Vehicles (IoV) is becoming increasingly vulnerable to zero-day (ZD) cyberattacks, which often bypass conventional intrusion detection systems. To mitigate this challenge, this study proposes Zero-Day Bidirectional Encoder Representations from Transformers approach (ZDBERTa), a zero-shot learning (ZSL)-based framework for ZD attack detection, evaluated on the CICIoV2024 dataset. Unlike conventional AI models, ZSL enables the classification of attack types not previously encountered during the training phase. Two dataset variants are formed: Variant 1, created through synthetic traffic generation using a mixture of pattern-based, crossover, and mutation techniques, and Variant 2, augmented with a Generative Adversarial Network (GAN). To replicate realistic zero-day conditions, denial-of-service (DoS) attacks were omitted during training and introduced only at testing. The proposed ZDBERTa incorporates a Byte-Pair Encoding (BPE) tokenizer, a multi-layer transformer encoder, and a classification head for prediction, enabling the model to capture semantic patterns and identify previously unseen threats. The experimental results demonstrate that ZDBERTa achieves 86.677% accuracy on Variant 1, highlighting the complexity of zero-day detection, while performance significantly improves to 99.315% on Variant 2, underscoring the effectiveness of GAN-based augmentation. To the best of our knowledge, this is the first research to explore ZD detection within CICIoV2024, contributing a novel direction toward resilient IoV cybersecurity. Full article

Efficient routing optimization remains a pivotal challenge in Electronic Design Automation (EDA), as it profoundly influences circuit performance, power consumption, and manufacturing cost. The Rectilinear Steiner Minimum Tree (RSMT) problem plays a crucial role in this process by minimizing the routing length through the introduction of Steiner points. This paper proposes a reinforcement learning-driven RSMT construction model that incorporates a novel Selective Kernel Transformer Network (SKTNet) encoder to enhance feature representation. SKTNet integrates a Selective Kernel Convolution (SKConv) and an improved Macaron Transformer to improve multi-scale feature extraction and global topology modeling. Additionally, Self-Critical Sequence Training (SCST) is employed to optimize the policy by leveraging a greedy-decoded baseline sequence for the advantage computation. Experimental results demonstrate superior performance over state-of-the-art methods in wirelength optimization. Ablation studies further validate the contribution of this model, highlighting its effectiveness and scalability for routing. Full article