Road Obstacle Detection Method Based on Improved YOLOv5 DOI Creative Commons

Pengliu Tan,

Zhi Wang, Xin Chang

et al.

Algorithms, Journal Year: 2025, Volume and Issue: 18(6), P. 300 - 300

Published: May 22, 2025

Road obstacle detection is essential for ensuring the smooth operation of roads and safeguarding lives property travelers. However, current methods face challenges such as missed detections false positives. To address these issues, an enhanced algorithm based on YOLOv5 (YOLOv5-EC3F) proposed. First, effective multi-scale feature fusion module (EMFF) introduced to extract features from input map, providing richer semantic information enhancing perceptual range. Second, SPPF replaced with C3SPPF improve model’s understanding contextual increase its adaptability. Experimental results demonstrate that, custom dataset, YOLOv5-EC3F raises mAP by 3 percentage points 82% recall 7 78%, without compromising precision. This study offers a valuable optimization strategy practical application road detection.

Language: Английский

SC-AttentiveNet: Lightweight Multiscale Feature Fusion Network for Surface Defect Detection on Copper Strips DOI Open Access

Zeteng Li,

Guo Zhang,

Qi Yang

et al.

Electronics, Journal Year: 2025, Volume and Issue: 14(7), P. 1422 - 1422

Published: April 1, 2025

Small defects on the surface of copper strips have a significant impact key properties such as electrical conductivity and corrosion resistance, existing inspection techniques struggle to meet demand in terms accuracy generalisability. Although there been some studies metal defect detection, is relative lack research highly reflective strips. In this paper, lightweight efficient strip detection algorithm, SC-AttentiveNet, proposed, aiming solve problems large model size, slow speed, insufficient poor generalisability models. The algorithm based ConvNeXt V2, combines SCDown module group normalisation design SCGNNet feature extraction network, which significantly reduces computational overhead while maintaining excellent capability. addition, introduces SPPF-PSA enhance multi-scale capability, constructs new neck fusion network via HD-CF Fusion Block module, further enhances diversity fine granularity. experimental results show that SC-AttentiveNet has mAP 90.11% 64.14% KUST-DET VOC datasets, respectively, with parameter count only 6.365 MB complexity 14.442 GFLOPs. Tests NEU-DET dataset an generalisation performance, 76.41% speed 78 FPS, demonstrating wide range practical application potential.

Language: Английский

Citations

0

Two methodologies for brain signal analysis derived from Freeman Neurodynamics DOI Creative Commons

Jeffery Jonathan Joshua Davis,

Ian J. Kirk, Róbert Kozma

et al.

Frontiers in Systems Neuroscience, Journal Year: 2025, Volume and Issue: 19

Published: April 15, 2025

Here, Freeman Neurodynamics is explored to introduce the reader challenges of analyzing electrocorticogram or electroencephalogram signals make sense two things: (a) how brain participates in creation knowledge and meaning (b) differentiate between cognitive states modalities dynamics. The first addressed via a Hilbert transform-based methodology second Fourier transform methodology. These methodologies, it seems us, conform with systems' neuroscience views, models, signal analysis methods that Walter J. III used left for us as his legacy.

Language: Английский

Citations

0

Applications for Predicting Cracking Outlet Temperatures for Ethylene Cracker Furnaces Based on the GAC-BiLSTM-AM Method DOI Open Access
Yanwei Dong, Qirui Li,

Delong Cui

et al.

Processes, Journal Year: 2025, Volume and Issue: 13(5), P. 1269 - 1269

Published: April 22, 2025

In order to achieve a good generalization ability across different prediction tasks and ensure the reliability of results, this paper proposes GAC-BiLSTM-AM method based on ensemble learning. This employs dual-channel convolutional neural network comprehensively extract integrate sample features, which are then input into composite analysis module consisting bidirectional long short-term memory, an attention mechanism, fully connected layer in-depth processing information, jointly constructing efficient base learner. Meanwhile, during parameter iteration tuning process learner, mistake correction mechanism is introduced dynamically adjust weights in correct potential biases, further combined with gold rush optimizer optimize key structural parameters model. The application evaluation multiple benchmark confirms that possesses excellent performance.

Language: Английский

Citations

0

Experimental and Analytical Studies on Electromagnetic Wave Propagation in DC Circuits DOI Open Access
Mingyu Lu,

Charan Litchfield

Published: April 7, 2025

It is well known that two pieces of electrical conductors behave as a waveguide when they are employed to transmit AC signals. Some experimental and analytical studies reported in this paper demonstrate also DC signals practice. Specifically, the speed wave propagation measured experiments, based on theory transmission line.

Language: Английский

Citations

0

A lightweight deep learning framework for wild berry detection in complex natural environments DOI
Xiaorong Zhang, Fei Li, Xuting Hu

et al.

Engineering Applications of Artificial Intelligence, Journal Year: 2025, Volume and Issue: 154, P. 110918 - 110918

Published: April 29, 2025

Language: Английский

Citations

0

Hierarchical Structure of the Program Used by Filamentous Fungi to Navigate in Confining Microenvironments DOI Creative Commons
Gala Montiel Rubies, Marie Held,

Kristi L. Hanson

et al.

Biomimetics, Journal Year: 2025, Volume and Issue: 10(5), P. 287 - 287

Published: May 2, 2025

The spatial navigation of filamentous fungi was compared for three species, namely Pycnoporus cinnabarinus, Neurospora crassa wild type and ro-1 mutant, Armillaria mellea, in microfluidic structures. analysis the these open especially confining environments suggests that they perform space exploration using a hierarchical, three-layered system information processing. output single hypha is result coordination competition between programs with their corresponding subroutines: (i) sensing narrow passages (remote- or contact-based); (ii) directional memory; (iii) branching (collision-induced stochastic). One information-processing level up, distribution multiple, closely collocated hyphae combination negative autotropism cytoplasm reallocation related branches (with anastomosis as an alternative subroutine to increase robustness). Finally, mycelium sum quasi-autonomous sub-populations performing parallel based on different conditions constraints found locally. efficiency by appears be synergy various biological algorithms integrated into hierarchical architecture processing, balancing complexity specialization.

Language: Английский

Citations

0

Operator Expertise in Bilateral Teleoperation: Performance, Manipulation, and Gaze Metrics DOI Open Access
Harun Tugal, İhsan TUĞAL,

Fumiaki Abe

et al.

Electronics, Journal Year: 2025, Volume and Issue: 14(10), P. 1923 - 1923

Published: May 9, 2025

This paper presents a comprehensive user study aimed as assessing and differentiating operator expertise within bilateral teleoperation systems. The primary objective is to identify key performance metrics that effectively distinguish novice from expert users. Unlike prior approaches focus primarily on psychological evaluations, this emphasizes direct analysis across range of telerobotic tasks. Ten participants (six novices four experts) were assessed based task completion time difficulty, error rates, manipulator motion characteristics, gaze behaviour, subjective feedback via questionnaires. results show experienced operators outperformed by completing tasks faster, making fewer errors, demonstrating smoother control, reflected reduced jerks higher spatial precision. Also, experts maintained consistent even complexity increased, whereas sharp decline, particularly at difficulty levels. Questionnaire responses further revealed mental physical demands, especially in unfamiliar tasks, while demonstrated concentration arousal Additionally, the introduces transition entropy (GTE) stationary (SGE) quantify visual attention strategies, with exhibiting more focused, goal-oriented patterns, showed erratic inefficient behaviour. These findings highlight both quantitative qualitative measures critical for evaluating informing future training programs.

Language: Английский

Citations

0

Impact of Polymers on Sand Sedimentation Characteristics of Shale Oil-Produced Fluid DOI Open Access

Yongbin Shang,

Qiaosheng Zhang,

Wanrui Li

et al.

Materials, Journal Year: 2025, Volume and Issue: 18(10), P. 2269 - 2269

Published: May 14, 2025

The introduction of polymers has significantly altered the properties sand particles in shale oil production fluids, leading to a more complex sedimentation mechanism. However, specific ways which influence dynamics remain poorly understood. In this study, Soxhlet extraction and supercritical water oxidation techniques were employed compare particle size distribution polymer-containing with that actual sand. results show fluids involves two mechanisms: gravity-dominated single-particle polymer-induced multi-particle flocculation-sedimentation. Additionally, induce both flocculation-sedimentation hindering effects. Specifically, content temperature can promote group by adjusting rheology, polymer content, stability fluid. experimental rates processes increased 38.05% 54.76%, respectively. Based on these findings, characteristics under obtained, offering valuable insights for management control fluids.

Language: Английский

Citations

0

Explainable Machine Learning with Two-Layer Multi-Objective Optimization Algorithm Applied to Sealing Structure Design DOI Open Access

Weiru Zhou,

Zonghong Xie

Materials, Journal Year: 2025, Volume and Issue: 18(10), P. 2307 - 2307

Published: May 15, 2025

This study addresses the challenge of optimizing seal structure design through a novel two-stage interpretable optimization framework. Focusing on O-ring waterproof performance under hyperelastic material behavior, this proposes double-layer method integrating explainable machine learning with hierarchical clustering algorithms. The key innovation lies in employing modified to categorize parameters into two groups: bolt preload and groove depth. enables dimensionality reduction while maintaining physical interpretability critical parameters. In first layer, systematic parameter screening are applied variable reduce database, six remaining data points that constitute one-seventh original data. second layer subsequently refines configurations using E-TOPSIS (Entropy Weight—Technique for Order Preference by Similarity Ideal Solution) optimization. All evaluations performed FEA (finite element analysis) considering nonlinear responses. optimal is depth 0.8 mm 80 N. experimental validation demonstrates efficiently identifies designs meeting IPX8 requirements, zero leakage observed both surfaces motor interiors. proposed methodology provides physically meaningful guidelines.

Language: Английский

Citations

0

Simple Design of Mechanical Ventilator for Mass Production May Offer Excellent Performance, Precise Monitoring, and Advanced Safety DOI Creative Commons
Šimon Walzel, Ladislav Bís, Václav Ort

et al.

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(10), P. 5631 - 5631

Published: May 18, 2025

The COVID-19 pandemic raised global concerns about the shortage of ventilators and revealed challenges rapidly scaling up production to meet emergency needs. In response, numerous teams worldwide attempted develop simple mechanical ventilators. Among these, CoroVent ventilator was developed urgent need for ventilatory support in Czech Republic. aim this study describe innovative design ventilator, evaluate its compliance with international safety performance standards, verify reliability under simulated clinical conditions, demonstrate suitability use crisis scenarios. designed a focus on needs patients respiratory failure ensure safe ventilation while maintaining simplified design. It features volume-controlled, pressure-limited mandatory supports key adjustable parameters such as tidal volume, rate, inspiratory-to-expiratory time ratio, inspired oxygen fraction, positive end-expiratory pressure (PEEP). incorporates robust mechanisms, including alarms relief valve, protect against excessive airway pressures. Results confirmed ability maintain consistent volumes, stable PEEP, precise limitation over extended periods use. results showed that met essential standards accuracy, those set by UK Medicines Healthcare products Regulatory Agency, U.S. Food Drug Administration, ISO 80601-2-12. Although these stopped 2021 Republic managed ventilators, validate their indicate potential critical care situations.

Language: Английский

Citations

0