Sensors

18 pages, 1693 KiB

Open AccessArticle

AI-Powered Analysis of Eye Tracker Data in Basketball Game

by Daniele Lozzi, Ilaria Di Pompeo, Martina Marcaccio, Michela Alemanno, Melanie Krüger, Giuseppe Curcio and Simone Migliore

Sensors 2025, 25(11), 3572; https://doi.org/10.3390/s25113572 - 5 Jun 2025

Viewed by 245

Abstract

This paper outlines a new system for processing of eye-tracking data in basketball live games with two pre-trained Artificial Intelligence (AI) models. blueThe system is designed to process and extract features from data of basketball coaches and referees, recorded with the Pupil Labs [...] Read more.

This paper outlines a new system for processing of eye-tracking data in basketball live games with two pre-trained Artificial Intelligence (AI) models. blueThe system is designed to process and extract features from data of basketball coaches and referees, recorded with the Pupil Labs Neon Eye Tracker, a device that is specifically optimized for video analysis. The research aims to present a tool useful for understanding their visual attention patterns during the game, what they are attending to, for how long, and their physiological responses, blueas is evidenced through pupil size changes. AI models are used to monitor events and actions within the game and correlate these with eye-tracking data to provide understanding into referees’ and coaches’ cognitive processes and decision-making. This research contributes to the knowledge of sport psychology and performance analysis by introducing the potential of Artificial Intelligence (AI)-based eye-tracking analysis in sport with wearable technology and light neural networks that are capable of running in real time. Full article

(This article belongs to the Special Issue Artificial Intelligence (AI) and Sensors in Sports Safety and NextGen Rehabilitation)

► Show Figures

Figure 1

27 pages, 4448 KiB

Open AccessArticle

Remaining Useful Life Prediction for Rolling Bearings Based on TCN–Transformer Networks Using Vibration Signals

by Xiaochao Jin, Yaping Ji, Shiteng Li, Kailang Lv, Jianzheng Xu, Haonan Jiang and Shengnan Fu

Sensors 2025, 25(11), 3571; https://doi.org/10.3390/s25113571 - 5 Jun 2025

Viewed by 304

Abstract

Remaining useful life (RUL) prediction plays a core role in industrial prognostics and health management (PHM), requiring data-driven models with higher predictive capability for accurate long time series prediction. Developing reliable deep learning-based models based on multi-sensor monitoring data is fundamental for accurately [...] Read more.

Remaining useful life (RUL) prediction plays a core role in industrial prognostics and health management (PHM), requiring data-driven models with higher predictive capability for accurate long time series prediction. Developing reliable deep learning-based models based on multi-sensor monitoring data is fundamental for accurately predicting vibration trends during bearing operation and is crucial for bearing fault diagnosis and RUL prediction. In this work, a method for constructing a health index based on vibration signal is developed to describe the performance features of rolling bearings, which mainly includes feature extraction, sensitive feature index selection, dimensionality reduction, and normalization methods. In addition, a new RUL prediction method, TCN–Transformer, is developed which can efficiently learn and integrate local and global features of vibration signals, addressing the long time series prediction problem in RUL prediction. The TCN extracts local features, while the Transformer learns global features, both of which are seamlessly integrated through a specially designed feature fusion attention module. Both the health indicator (HI) constructed from extracted time domain and frequency domain feature parameters and the RUL prediction method were rigorously validated using the IEEE PHM 2012 Data Challenge dataset for rolling bearing prognostics. By employing the proposed HI construction method, the average comprehensive bearing performance index, used to evaluate RUL prediction accuracy, is improved by 8.69% across the entire dataset compared to the original feature-based composite index. The proposed RUL prediction model can more accurately predict the RUL of rolling bearings under different conditions, reducing the RMSE and MAE by 14.62% and 9.26%, respectively, and improving the SCORE by 13.04%. These results underscore the efficacy and superiority of our approach in RUL prediction of rotating machinery across varying conditions. Full article

(This article belongs to the Section Fault Diagnosis & Sensors)

► Show Figures

Figure 1

20 pages, 5153 KiB

Open AccessArticle

A Practical Method for Red-Edge Band Reconstruction for Landsat Image by Synergizing Sentinel-2 Data with Machine Learning Regression Algorithms

by Yuan Zhang, Zhekui Fan, Wenjia Yan, Chentian Ge and Huasheng Sun

Sensors 2025, 25(11), 3570; https://doi.org/10.3390/s25113570 - 5 Jun 2025

Viewed by 312

Abstract

Red-edge bands are the most essential spectral data for multispectral remote sensing images, with them playing a critical role in monitoring vegetation growth status at regional and global scales. However, the absence of red-edge bands limits the applicability of Landsat images, the most [...] Read more.

Red-edge bands are the most essential spectral data for multispectral remote sensing images, with them playing a critical role in monitoring vegetation growth status at regional and global scales. However, the absence of red-edge bands limits the applicability of Landsat images, the most widely used remote sensing data, to vegetation monitoring. This study proposes an innovative method to reconstruct Landsat’s red-edge bands. The consistency in corresponding bands of Landsat OLI and Sentinel-2 MSI was first investigated using different resampling approaches and atmospheric correction algorithms. Three machine learning algorithms (ridge regression, gradient boosted regression tree (GBRT), and random forest regression) were then employed to build the red-edge reconstruction model for different vegetation types. With the optimal model, three red-edge bands of Landsat OLI were subsequently obtained in alignment with their derived vegetation indices. Our results showed that bilinear interpolation resampling, in combination with the LaSRC atmospheric correction algorithm, achieved high consistency between the matching bands of OLI and MSI (R² > 0.88). With the GBRT algorithm, three simulated OLI red-edge bands were highly consistent with those of MSI, with an R² > 0.96 and an RMSE < 0.0122. The derived Landsat red-edge indices coincide with those of Sentinel-2, with an R² of 0.78 to 0.95 and an rRMSE of 3.37% to 21.64%. This study illustrates that the proposed red-edge reconstruction method can extend the spectral domain of Landsat OLI and enhance its applicability in global vegetation remote sensing. Meanwhile, it provides potential insight into historical Landsat TM/ETM+ data enhancement for improving time-series vegetation monitoring. Full article

(This article belongs to the Special Issue Machine Learning in Image/Video Processing and Sensing)

► Show Figures

Figure 1

19 pages, 3834 KiB

Open AccessArticle

A Sensitive and Selective Sensor Based on Orthorhombic Copper Molybdate Decorated on Reduced Graphene Oxide for the Detection of Promethazine Hydrochloride

by Venkatachalam Vinothkumar, Yellatur Chandra Sekhar, Shen-Ming Chen, Natesan Manjula and Tae Hyun Kim

Sensors 2025, 25(11), 3569; https://doi.org/10.3390/s25113569 - 5 Jun 2025

Viewed by 199

Abstract

Promethazine hydrochloride (PMH) is a first-generation antipsychotic drug created from phenothiazine derivatives that is widely employed to treat psychiatric disorders in human healthcare systems. However, an overdose or long-term intake of PMH can lead to severe health issues in humans. Hence, establishing a [...] Read more.

Promethazine hydrochloride (PMH) is a first-generation antipsychotic drug created from phenothiazine derivatives that is widely employed to treat psychiatric disorders in human healthcare systems. However, an overdose or long-term intake of PMH can lead to severe health issues in humans. Hence, establishing a sensitive, accurate, and efficient detection approach to detect PMH in human samples is imperative. In this study, we designed orthorhombic copper molybdate microspheres decorated on reduced graphene oxide (Cu₃Mo₂O₉/RGO) composite via the effective one-pot hydrothermal method. The structural and morphological features of the designed hybrid were studied using various spectroscopic methods. Subsequently, the electrochemical activity of the composite-modified screen-printed carbon electrode (Cu₃Mo₂O₉/RGO/SPCE) was assessed by employing voltammetric methods for PMH sensing. Owing to the uniform composition and structural benefits, the combination of Cu₃Mo₂O₉ and RGO has not only improved electrochemical properties but also enhanced the electron transport between PMH and Cu₃Mo₂O₉/RGO. As a result, the Cu₃Mo₂O₉/RGO/SPCE exhibited a broad linear range of 0.4–420.8 µM with a low limit of detection (LoD) of 0.015 µM, highlighting excellent electrocatalytic performance to PMH. It also demonstrated good cyclic stability, reproducibility, and selectivity in the presence of chlorpromazine and biological and metal compounds. Furthermore, the Cu₃Mo₂O₉/RGO/SPCE sensor displayed satisfactory recoveries for real-time monitoring of PMH in human urine and serum samples. This study delivers a promising electrochemical sensor for the efficient analysis of antipsychotic drug molecules. Full article

(This article belongs to the Special Issue Electrochemical Sensors and Biosensors: Materials, Methods and Applications)

► Show Figures

Graphical abstract

19 pages, 3241 KiB

Open AccessArticle

A Three-Dimensional Target Localization Method for Satellite–Ground Bistatic Radar Based on a Geometry–Motion Cooperative Constraint

by Fangrui Zhang, Hu Xie, She Shang, Hongxing Dang, Dawei Song and Zepeng Yang

Sensors 2025, 25(11), 3568; https://doi.org/10.3390/s25113568 - 5 Jun 2025

Viewed by 222

Abstract

This paper investigates the three-dimensional target localization problem in satellite–ground bistatic radar. In conventional bistatic radar systems, passive receivers struggle to directly acquire the altitude information of the target, making it difficult to achieve effective three-dimensional target localization. This paper uses the bistatic [...] Read more.

This paper investigates the three-dimensional target localization problem in satellite–ground bistatic radar. In conventional bistatic radar systems, passive receivers struggle to directly acquire the altitude information of the target, making it difficult to achieve effective three-dimensional target localization. This paper uses the bistatic distance data obtained after signal processing to construct ellipsoidal constraints, thereafter combining azimuth data to compress the position solution space into a three-dimensional elliptical line. Introducing the assumption of short-term linear uniform motion of the target, the target trajectory and elliptical line constraints are projected onto a two-dimensional plane, establishing an optimization model to determine the target trajectory parameters, ultimately yielding the target’s three-dimensional coordinates and completing the positioning process. The simulation results demonstrate the efficacy and performance of the proposed method. Full article

(This article belongs to the Section Navigation and Positioning)

► Show Figures

Figure 1

20 pages, 1842 KiB

Open AccessArticle

A.A.A. Good Wines WANTED: Blockchain, Non-Destructive Ultrasonic Techniques and Soil Health Assessment for Wine Traceability

by Diego Romano Perinelli, Martina Coletta, Beatrice Sabbatini, Aldo D’Alessandro, Fabio Fabiani, Andrea Passacantando, Giulia Bonacucina and Antonietta La Terza

Sensors 2025, 25(11), 3567; https://doi.org/10.3390/s25113567 - 5 Jun 2025

Viewed by 179

Abstract

The wine industry faces increasing challenges related to authenticity, safety, and sustainability due to recurrent fraud, shifting consumer preferences, and environmental concerns. In this study, as part of the B.I.O.C.E.R.T.O project, we integrated blockchain technology with ultrasonic spectroscopy and soil quality data by [...] Read more.

The wine industry faces increasing challenges related to authenticity, safety, and sustainability due to recurrent fraud, shifting consumer preferences, and environmental concerns. In this study, as part of the B.I.O.C.E.R.T.O project, we integrated blockchain technology with ultrasonic spectroscopy and soil quality data by using the arthropod-based Soil Biological Quality Index (QBS-ar) to enhance traceability, ensure wine quality, and certify sustainable vineyard practices. Four representative wines from the Marche region (Sangiovese, Maceratino, and two Verdicchio PDO varieties) were analyzed across two vintages (2021 and 2022). Ultrasound spectroscopy demonstrated high sensitivity in distinguishing wines based on ethanol and sugar content, comparably to conventional viscosity-based methods. The QBS-ar index was applied to investigate the soil biodiversity status according to the agricultural management practices applied in each vineyard, reinforcing consumer confidence in environmentally responsible viticulture. By recording these data on a public blockchain, we developed a secure, transparent, and immutable certification system to verify the geographical origin of wines along with their unique characteristics. This is the first study to integrate advanced analytical techniques with blockchain technology for wine traceability, simultaneously addressing counterfeiting, consumer demand for transparency, and biodiversity preservation. Our findings support the applicability of this model to other agri-food sectors, with potential for expansion through additional analytical techniques, such as isotopic analysis and further agroecosystem sustainability indicators. Full article

(This article belongs to the Section Chemical Sensors)

► Show Figures

Graphical abstract

22 pages, 6539 KiB

Open AccessArticle

Development of a Multi-Sensor GNSS-IoT System for Precise Water Surface Elevation Measurement

by Jun Wang, Matthew C. Garthwaite, Charles Wang and Lee Hellen

Sensors 2025, 25(11), 3566; https://doi.org/10.3390/s25113566 - 5 Jun 2025

Viewed by 235

Abstract

The Global Navigation Satellite System (GNSS), Internet of Things (IoT) and cloud computing technologies enable high-precision positioning with flexible data communication, making real-time/near-real-time monitoring more economical and efficient. In this study, a multi-sensor GNSS-IoT system was developed for measuring precise water surface elevation [...] Read more.

The Global Navigation Satellite System (GNSS), Internet of Things (IoT) and cloud computing technologies enable high-precision positioning with flexible data communication, making real-time/near-real-time monitoring more economical and efficient. In this study, a multi-sensor GNSS-IoT system was developed for measuring precise water surface elevation (WSE). The system, which includes ultrasonic and accelerometer sensors, was deployed on a floating platform in Googong reservoir, Australia, over a four-month period in 2024. WSE data derived from the system were compared against independent reference measurements from the reservoir operator, achieving an accuracy of 7 mm for 6 h averaged solutions and 28 mm for epoch-by-epoch solutions. The results demonstrate the system’s potential for remote, autonomous WSE monitoring and its suitability for validating satellite Earth observation data, particularly from the Surface Water and Ocean Topography (SWOT) mission. Despite environmental challenges such as moderate gale conditions, the system maintained robust performance, with over 90% of solutions meeting quality assurance standards. This study highlights the advantages of combining the GNSS with IoT technologies and multiple sensors for cost-effective, long-term WSE monitoring in remote and dynamic environments. Future work will focus on optimizing accuracy and expanding applications to diverse aquatic settings. Full article

(This article belongs to the Topic Advances in Earth Observation Technologies to Support Water-Related Sustainable Development Goals (SDGs))

► Show Figures

Figure 1

22 pages, 4582 KiB

Open AccessArticle

Enhanced Object Detection in Thangka Images Using Gabor, Wavelet, and Color Feature Fusion

by Yukai Xian, Yurui Lee, Te Shen, Ping Lan, Qijun Zhao and Liang Yan

Sensors 2025, 25(11), 3565; https://doi.org/10.3390/s25113565 - 5 Jun 2025

Viewed by 289

Abstract

Thangka image detection poses unique challenges due to complex iconography, densely packed small-scale elements, and stylized color–texture compositions. Existing detectors often struggle to capture both global structures and local details and rarely leverage domain-specific visual priors. To address this, we propose a frequency- [...] Read more.

Thangka image detection poses unique challenges due to complex iconography, densely packed small-scale elements, and stylized color–texture compositions. Existing detectors often struggle to capture both global structures and local details and rarely leverage domain-specific visual priors. To address this, we propose a frequency- and prior-enhanced detection framework based on YOLOv11, specifically tailored for Thangka images. We introduce a Learnable Lifting Wavelet Block (LLWB) to decompose features into low- and high-frequency components, while LLWB_Down and LLWB_Up enable frequency-guided multi-scale fusion. To incorporate chromatic and directional cues, we design a Color-Gabor Block (CGBlock), a dual-branch attention module based on HSV histograms and Gabor responses, and embed it via the Color-Gabor Cross Gate (C2CG) residual fusion module. Furthermore, we redesign all detection heads with decoupled branches and introduce center-ness prediction, alongside an additional shallow detection head to improve recall for ultra-small targets. Extensive experiments on a curated Thangka dataset demonstrate that our model achieves 89.5% mAP@0.5, 59.4% mAP@[0.5:0.95], and 84.7% recall, surpassing all baseline detectors while maintaining a compact size of 20.9 M parameters. Ablation studies validate the individual and synergistic contributions of each proposed component. Our method provides a robust and interpretable solution for fine-grained object detection in complex heritage images. Full article

(This article belongs to the Section Sensing and Imaging)

► Show Figures

Figure 1

27 pages, 3526 KiB

Open AccessArticle

Addressing Sensor Data Heterogeneity and Sample Imbalance: A Transformer-Based Approach for Battery Degradation Prediction in Electric Vehicles

by Bi Wu, Shi Qiu and Wenhe Liu

Sensors 2025, 25(11), 3564; https://doi.org/10.3390/s25113564 - 5 Jun 2025

Viewed by 230

Abstract

Battery health monitoring and remaining useful life (RUL) estimation for electric vehicles face two significant challenges: battery data heterogeneity and sample imbalance. This study presents a novel approach based on Transformer architecture to specifically address these issues. We utilized the NASA lithium-ion battery [...] Read more.

Battery health monitoring and remaining useful life (RUL) estimation for electric vehicles face two significant challenges: battery data heterogeneity and sample imbalance. This study presents a novel approach based on Transformer architecture to specifically address these issues. We utilized the NASA lithium-ion battery cycling dataset, which contains charge-discharge and impedance measurement data under various temperature conditions. To tackle data heterogeneity, we developed a multimodal feature fusion strategy that effectively integrates battery sensor data from different sources and formats, including time-series charge-discharge sensor data and spectral impedance sensor measurements. To mitigate sample imbalance, we implemented an adaptive resampling technique and hierarchical attention mechanism, enhancing the model’s ability to recognize rare degradation patterns. Our Transformer-based model captures long-term dependencies in the battery degradation process through its self-attention mechanism. Experimental results demonstrate that the proposed solution significantly improves battery degradation prediction accuracy, achieving a 21.3% increase in accuracy when processing heterogeneous data and a 24.5% improvement in prediction capability for imbalanced samples compared to traditional methods. Additionally, through case studies, we validate the applicability of this method in actual electric vehicle battery management systems, providing reliable data support for battery preventive maintenance and replacement decisions. The findings have important implications for enhancing the reliability and economic efficiency of electric vehicle battery management systems. Full article

(This article belongs to the Section Vehicular Sensing)

► Show Figures

Figure 1

2 pages, 482 KiB

Open AccessCorrection

Correction: Hannotier et al. A CFO-Assisted Algorithm for Wireless Time-Difference-of-Arrival Localization Networks: Analytical Study and Experimental Results. Sensors 2024, 24, 737

by Cédric Hannotier, François Horlin and François Quitin

Sensors 2025, 25(11), 3563; https://doi.org/10.3390/s25113563 - 5 Jun 2025

Viewed by 154

Abstract

In the original publication [...] Full article

(This article belongs to the Special Issue Wireless Communications in Vehicular Network)

► Show Figures

Figure 10

27 pages, 4944 KiB

Open AccessArticle

Study on Electric Power Fittings Identification Method for Snake Inspection Robot Based on Non-Contact Inductive Coils

by Zhiyong Yang, Jianguo Liu, Shengze Yang and Changjin Zhang

Sensors 2025, 25(11), 3562; https://doi.org/10.3390/s25113562 - 5 Jun 2025

Viewed by 249

Abstract

In power inspection fields, snake-like robots are often used for transmission line inspection tasks, requiring accurate identification of various power fittings. However, traditional visual sensors perform poorly under varying light intensity and complex background conditions. This paper proposes a non-visual perception method for [...] Read more.

In power inspection fields, snake-like robots are often used for transmission line inspection tasks, requiring accurate identification of various power fittings. However, traditional visual sensors perform poorly under varying light intensity and complex background conditions. This paper proposes a non-visual perception method for the high-precision classification of different power fittings (e.g., vibration dampers, suspension clamps, and tension clamps) in snake-like robot transmission line inspection for high-voltage lines. This method, unaffected by light intensity changes, uses machine learning to classify the magnetic induction electromotive force signals around the fittings. First, the Dodd–Deeds eddy current model is used to analyse the magnetic field changes around the transmission line fittings and determine the induction coil distribution. Then, the concept of condition number and singular value decomposition (SVD) are introduced to analyse the impact of detection position on classification accuracy, with optimal detection positions found using the particle swarm optimization algorithm. Finally, a BP neural network optimised by a genetic algorithm is used for power fitting identification. Experiments show that this method successfully identifies vibration dampers, tension clamps, suspension clamps, and transmission lines at detection distances of 5 cm, 10 cm, 15 cm, and 20 cm, with accuracies of 99.8%, 97.5%, 95.1%, and 92.5%, respectively. Full article

(This article belongs to the Section Electronic Sensors)

► Show Figures

Figure 1

12 pages, 1993 KiB

Open AccessArticle

Determination of the Precision of Glucometers Used in Saudi Arabia

by Shoug A. Al-Othman, Zahra H. Al-Zaidany, Shahad H. Al-Ghannam, Ahmed M. Al-Turki, Abdulrahman A. Al-Abdulazeem, Chittibabu Vatte, Alawi Habara, Amein K. Al-Ali and Mohammed F. Al-Awami

Sensors 2025, 25(11), 3561; https://doi.org/10.3390/s25113561 - 5 Jun 2025

Viewed by 684

Abstract

Background: Efforts have been joined to set the parameters for the reliability of glucometers, yet once they are on the market, they are not further tested for the maintenance of accuracy, specificity, or precision. Methods: This comparative analytical study investigated the precision of [...] Read more.

Background: Efforts have been joined to set the parameters for the reliability of glucometers, yet once they are on the market, they are not further tested for the maintenance of accuracy, specificity, or precision. Methods: This comparative analytical study investigated the precision of commonly used glucometers in Saudi Arabia, namely Accu-Chek Instant^®, On-Call Sharp^®, and ConTour^®, as well as the effects of vitamin C, acetaminophen, and maltose on glucose readings. Ten milliliters of blood was drawn in lithium heparin from healthy volunteers (n = 9). Six samples were divided into two groups of three. One group was designed for normal glucose levels. The second group was designed for high glucose levels by adding a dextrose solution. The last three samples were designed for low glucose levels by leaving the sample for 24 h at room temperature and then following with centrifuge and plasma extraction. Results: This study showed that only Accu-Chek Instant met the International Organization for Standardization (ISO) standard for precision across all dextrose concentrations, along with intra-class correlation values ranging from 0.95–1 (p < 0.001). By spiking the plasma samples with sub-therapeutic, therapeutic, and overdose concentrations of the metabolites, we found that vitamin C had a more evident interference on glucose readings compared to acetaminophen and maltose. Conclusions: The ascertainment of the precision of glucometers and the effects of interferences on them are vital in preventing the improper administration of insulin, which can lead to serious complications. Full article

(This article belongs to the Section Biosensors)

► Show Figures

Figure 1

1 pages, 126 KiB

Open AccessCorrection

Correction: Luna-Perejón et al. Smart Shoe Insole Based on Polydimethylsiloxane Composite Capacitive Sensors. Sensors 2023, 23, 1298

by Francisco Luna-Perejón, Blas Salvador-Domínguez, Fernando Perez-Peña, José María Rodríguez Corral, Elena Escobar-Linero and Arturo Morgado-Estévez

Sensors 2025, 25(11), 3560; https://doi.org/10.3390/s25113560 - 5 Jun 2025

Viewed by 186

Abstract

There was an error in the original publication [...] Full article

(This article belongs to the Section Intelligent Sensors)

9 pages, 1591 KiB

Open AccessCommunication

Highly Sensitive Dissolved Oxygen Sensor with High Stability in Seawater Based on Silica-Encapsulated Platinum(II) Porphyrin

by Hang Lv, Siyuan Cheng, Song Hu and Guohong Zhou

Sensors 2025, 25(11), 3559; https://doi.org/10.3390/s25113559 - 5 Jun 2025

Viewed by 224

Abstract

This study utilized tetramethylammonium hydroxide (TMAH) as a substitute for traditional catalysts and successfully incorporated platinum octaethylporphyrin (PtOEP) into SiO₂ nanoparticles (PtOEP@SiO₂) via the Stöber method. Methyl silicone resin was employed as the matrix material, and a drop-coating technique was [...] Read more.

This study utilized tetramethylammonium hydroxide (TMAH) as a substitute for traditional catalysts and successfully incorporated platinum octaethylporphyrin (PtOEP) into SiO₂ nanoparticles (PtOEP@SiO₂) via the Stöber method. Methyl silicone resin was employed as the matrix material, and a drop-coating technique was applied to fabricate thin films of PtOEP@SiO₂ particles for dissolved oxygen (DO) sensing in seawater. By optimizing the concentrations of TMAH and PtOEP, a highly sensitive oxygen-sensing film with a quenching ratio (I₀/I₁₀₀) of 28 was ultimately developed, with a wide linear detection range (0~20 mg/L, R² = 0.994). Stability tests revealed no significant performance degradation during five oxygen–nitrogen cycle tests. After 30 days of immersion in East China Sea seawater, the quenching ratio decreased by only 6%, confirming its long-term stability and excellent resistance to ion interference. This research provides a novel strategy for developing highly reliable in situ marine DO sensors. Full article

(This article belongs to the Section Optical Sensors)

► Show Figures

Figure 1

22 pages, 3223 KiB

Open AccessArticle

An EMG-Based GRU Model for Estimating Foot Pressure to Support Active Ankle Orthosis Development

by Praveen Nuwantha Gunaratne and Hiroki Tamura

Sensors 2025, 25(11), 3558; https://doi.org/10.3390/s25113558 - 5 Jun 2025

Viewed by 267

Abstract

As populations age, particularly in countries like Japan, mobility impairments related to ankle joint dysfunction, such as foot drop, instability, and reduced gait adaptability, have become a significant concern. Active ankle–foot orthoses (AAFO) offer targeted support during walking; however, most existing systems rely [...] Read more.

As populations age, particularly in countries like Japan, mobility impairments related to ankle joint dysfunction, such as foot drop, instability, and reduced gait adaptability, have become a significant concern. Active ankle–foot orthoses (AAFO) offer targeted support during walking; however, most existing systems rely on rule-based or threshold-based control, which are often limited to sagittal plane movements and lacking adaptability to subject-specific gait variations. This study proposes an approach driven by neuromuscular activation using surface electromyography (EMG) and a Gated Recurrent Unit (GRU)-based deep learning model to predict plantar pressure distributions at the heel, midfoot, and toe regions during gait. EMG signals were collected from four key ankle muscles, and plantar pressures were recorded using a customized sandal-integrated force-sensitive resistor (FSR) system. The data underwent comprehensive preprocessing and segmentation using a sliding window method. Root mean square (RMS) values were extracted as the primary input feature due to their consistent performance in capturing muscle activation intensity. The GRU model successfully generalized across subjects, enabling the accurate real-time inference of critical gait events such as heel strike, mid-stance, and toe off. This biomechanical evaluation demonstrated strong signal compatibility, while also identifying individual variations in electromechanical delay (EMD). The proposed predictive framework offers a scalable and interpretable approach to improving real-time AAFO control by synchronizing assistance with user-specific gait dynamics. Full article

(This article belongs to the Special Issue Sensor-Based Human Activity Recognition)

► Show Figures

Figure 1

22 pages, 14388 KiB

Open AccessArticle

A Dual-Band Flexible MIMO Array Antenna for Sub-6 GHz 5G Communications

by Deepthi Mariam John, Tanweer Ali, Shweta Vincent, Sameena Pathan, Jaume Anguera, Bal Virdee, Rajiv Mohan David, Krishnamurthy Nayak and Sudheesh Puthenveettil Gopi

Sensors 2025, 25(11), 3557; https://doi.org/10.3390/s25113557 - 5 Jun 2025

Viewed by 273

Abstract

This paper presents a novel dual-band flexible antenna, uniquely designed and extended to array as well as MIMO configurations for the Sub-6 GHz band. The single-element monopole antenna features a modified rectangular radiator with two L-strips and a reduced ground plane, enabling a [...] Read more.

This paper presents a novel dual-band flexible antenna, uniquely designed and extended to array as well as MIMO configurations for the Sub-6 GHz band. The single-element monopole antenna features a modified rectangular radiator with two L-strips and a reduced ground plane, enabling a compact dual-band response. The proposed four-element, two-port MIMO configuration is extended from the 1 × 2 array antenna, achieving an overall dimension of 57 × 50 × 0.1 mm³, making it exceptionally compact and flexible compared to existing rigid and bulkier designs. Operating in the 3.6–3.8 GHz and 5.65–5.95 GHz bands, the antenna delivers a high gain of 5.2 dBi and 7.7 dBi, outperforming many designs in terms of gain while maintaining the superior isolation of >22 dB utilizing a defected ground structure (DGS). The design satisfies key MIMO diversity metrics (ECC < 0.05, DG > 9.99) and demonstrates low SAR values (0.0702/0.25 W/kg at 3.75 GHz and 0.175/0.507 W/kg at 5.9 GHz), making it highly suitable for wearable and on-body communication, unlike many rigid counterparts. Fabricated on a flexible polyimide substrate, the antenna addresses challenges such as size, bandwidth, isolation, and safety in MIMO antenna design. The performance, validated through fabrication and measurement, establishes the proposed antenna as a superior alternative to existing MIMO designs for compact, high-performance Sub-6 GHz 5G applications. Full article

(This article belongs to the Section Communications)

► Show Figures

Figure 1

14 pages, 1433 KiB

Open AccessArticle

Evaluation of Optical and Thermal Properties of NIR-Blocking Ophthalmic Lenses Under Controlled Conditions

by Jae-Yeon Pyo, Min-Cheul Kim, Seung-Jin Oh, Ki-Choong Mah and Jae-Young Jang

Sensors 2025, 25(11), 3556; https://doi.org/10.3390/s25113556 - 5 Jun 2025

Viewed by 253

Abstract

This study evaluates the optical and thermal performance of near-infrared (NIR)-blocking spectacle lenses at luminous transmittance grades of 0, 2, and 3. Ten lens types were tested, including clear, tinted, and NIR-blocking spectacle lenses (NIBSL). The NIR blocking rate was measured across 780–1100 [...] Read more.

This study evaluates the optical and thermal performance of near-infrared (NIR)-blocking spectacle lenses at luminous transmittance grades of 0, 2, and 3. Ten lens types were tested, including clear, tinted, and NIR-blocking spectacle lenses (NIBSL). The NIR blocking rate was measured across 780–1100 nm and 1100–1400 nm wavelength bands. Color reproduction was assessed using sharpness (MTF 50), point spread function (PSF), and color accuracy (ΔE₀₀) under 1000 lux outdoor illumination. Thermal insulation was analyzed by monitoring porcine skin temperature at 36 °C and 60 °C under each lens type. As a result, the NIBSL showed better near-infrared blocking performance than other types of lenses in both wavelength ranges, and the coated NIBSL blocked near-infrared more effectively than the polymerized lenses. Compared with other types of lenses, NIBSL showed no difference in object identification, color recognition, and reproducibility, so there is no problem in using them together. Strong correlations were observed between lens surface temperature and underlying pig skin temperature, and inverse correlations between NIR blocking rate and pig skin temperature gradient. These findings confirm that NIBSL offer enhanced protection against NIR-induced thermal effects without compromising optical performance, supporting their use in daily environments for ocular and skin safety. Full article

(This article belongs to the Section Optical Sensors)

► Show Figures

Figure 1

15 pages, 7307 KiB

Open AccessArticle

GRACE-FO Satellite Data Preprocessing Based on Residual Iterative Correction and Its Application to Gravity Field Inversion

by Shuhong Zhao and Lidan Li

Sensors 2025, 25(11), 3555; https://doi.org/10.3390/s25113555 - 5 Jun 2025

Viewed by 225

Abstract

To address the limited inversion accuracy caused by low-fidelity data in satellite gravimetry, this study proposes a data preprocessing framework based on iterative residual correction. Utilizing Level-1B observations from the Gravity Recovery and Climate Experiment Follow-On (GRACE-FO) satellite (January 2020), outliers were systematically [...] Read more.

To address the limited inversion accuracy caused by low-fidelity data in satellite gravimetry, this study proposes a data preprocessing framework based on iterative residual correction. Utilizing Level-1B observations from the Gravity Recovery and Climate Experiment Follow-On (GRACE-FO) satellite (January 2020), outliers were systematically detected and removed, while data gaps were compensated through spline interpolation. Experimental results demonstrate that the proposed method effectively mitigates data discontinuities and anomalous perturbations, achieving a significant improvement in data quality. Furthermore, a 60-order Earth gravity field model derived via the energy balance approach was validated against contemporaneous models published by the University of Texas Center for Space Research (CSR), German Research Centre for Geosciences (GFZ), and Jet Propulsion Laboratory (JPL). The results reveal a two-order-of-magnitude enhancement in inversion precision, with model accuracy improving from 10⁻⁶–10⁻⁷ to 10⁻⁸–10⁻⁹. This method provides a robust solution for enhancing the reliability of gravity field recovery in satellite-based geodetic missions. Full article

(This article belongs to the Section Remote Sensors)

► Show Figures

Figure 1

30 pages, 853 KiB

Open AccessArticle

Privacy-Conducive Data Ecosystem Architecture: By-Design Vulnerability Assessment Using Privacy Risk Expansion Factor and Privacy Exposure Index

by Ionela Chereja, Rudolf Erdei, Daniela Delinschi, Emil Pasca, Anca Avram and Oliviu Matei

Sensors 2025, 25(11), 3554; https://doi.org/10.3390/s25113554 - 5 Jun 2025

Viewed by 323

Abstract

The increasing complexity of data ecosystems demands advanced methodologies for systematic privacy risk assessment. This work introduces two complementary metrics—the privacy risk expansion factor (PREF) and the privacy exposure index (PEI)—to evaluate how architectural decisions influence the exposure and distribution of sensitive data. [...] Read more.

The increasing complexity of data ecosystems demands advanced methodologies for systematic privacy risk assessment. This work introduces two complementary metrics—the privacy risk expansion factor (PREF) and the privacy exposure index (PEI)—to evaluate how architectural decisions influence the exposure and distribution of sensitive data. Several representative use cases validate the methodology, demonstrating how the metrics provide structured insights into the privacy impact of distinct design choices. By enabling comparative analysis across architectures, this approach supports the development of privacy-first data ecosystems and lays the groundwork for future research on dynamic, AI-driven risk monitoring. Full article

(This article belongs to the Special Issue Privacy and Cybersecurity in IoT-Based Applications)

► Show Figures

Figure 1

20 pages, 6031 KiB

Open AccessArticle

Comparison of Two System Identification Approaches for a Four-Wheel Differential Robot Based on Velocity Command Execution

by Diego Guffanti, Moisés Filiberto Mora Murillo, Marco Alejandro Hinojosa, Santiago Bustamante Sanchez, Javier Oswaldo Obregón Gutiérrez, Nelson Gutiérrez and Miguel Sánchez

Sensors 2025, 25(11), 3553; https://doi.org/10.3390/s25113553 - 5 Jun 2025

Viewed by 281

Abstract

Precise modeling of differential drive robots is crucial for effective control and trajectory planning in autonomous systems. A comparative analysis of two modeling approaches for a four-wheel differential drive robot is presented in this paper. The first approach, named Motor-Based Model (MBM), identifies [...] Read more.

Precise modeling of differential drive robots is crucial for effective control and trajectory planning in autonomous systems. A comparative analysis of two modeling approaches for a four-wheel differential drive robot is presented in this paper. The first approach, named Motor-Based Model (MBM), identifies four transfer functions, one for each motor, while the second approach, named Simplified Model (SM), uses only two transfer functions, one for linear velocity and another for angular velocity. Both models were validated by comparing their predicted trajectories against real odometry data obtained from a SLAM system implemented on a differential-drive robot. This provided a practical assessment of each model’s accuracy and underscored the importance of model selection in control design and navigation tasks. The results showed that the Motor-Based Model (MBM) consistently outperformed the Simplified Model (SM) in terms of odometry accuracy, both in position and orientation. Across all trajectories, the average RMSE for position using MBM was 0.309 m, while the SM recorded a higher average RMSE of 0.414 m. Similarly, the maximum position error averaged 0.522 m for MBM and 0.710 m for SM, confirming that MBM is more accurate and consistent in position tracking. Regarding the results of orientation estimation, when averaged across all experiments, the MBM maintained a lower angular RMSE of 0.170 rad in contrast to SM, which achieves an RMSE of 0.239 rad. The maximum angular error was also higher for the MBM at 0.316 rad, compared to 0.447 rad for the SM. Moreover, the computational performance evaluation indicated that the SM consistently outperformed MBM, achieving a 30% reduction in simulation time and substantially lower memory usage. These results demonstrate the relationship between model complexity and accuracy and suggest that the motor-specific model is more appropriate for applications requiring precise mapping or localization, such as SLAM, while the simplified model may be suitable for simpler use cases with lower computational requirements, such as embedded systems with limited resources. This paper provides a practical evaluation of the accuracy and computational performance of two modeling approaches, highlighting the implications of model selection for the design of navigation tasks. Full article

(This article belongs to the Section Sensors and Robotics)

► Show Figures

Figure 1

12 pages, 1646 KiB

Open AccessArticle

Estimation of the Relative Chlorophyll Content of Pear Leaves Based on Field Spectrometry in Alaer, Xinjiang

by Yufen Huang, Zhenqi Fan, Hongxin Wu, Ximeng Zhang and Yanlong Liu

Sensors 2025, 25(11), 3552; https://doi.org/10.3390/s25113552 - 5 Jun 2025

Viewed by 196

Abstract

Leaf chlorophyll content is an important indicator of the health status of pear trees. This study used Korla fragrant pears, a Xinjiang regional product, to investigate methods for estimating the relative chlorophyll content of pear leaves. Samples were collected from pear trees in [...] Read more.

Leaf chlorophyll content is an important indicator of the health status of pear trees. This study used Korla fragrant pears, a Xinjiang regional product, to investigate methods for estimating the relative chlorophyll content of pear leaves. Samples were collected from pear trees in the east, south, west, and north positions of peripheral canopy leaves. The leaf soil plant analysis development (SPAD) method was implemented using a SPAD-502 laser chlorophyll meter. The instrument measures the relative chlorophyll content as the SPAD value. Leaf spectra were acquired using a portable field spectrometer, ASD FieldSpec4. ViewSpecPro 6.2 software was employed to smooth the ground spectral data. Traditional mathematical transformations and the discrete wavelet transform were used to process the spectral data, then correlation analysis was employed to extract the sensitive bands, and partial least squares regression (PLS) was used to establish a model for estimating the chlorophyll content of pear tree leaves. The findings indicate that (1) the models developed using the discrete wavelet transform had coefficients of determination (R²) exceeding 0.65, and their predictive performance surpassed that of other models employing various mathematical transformations, and (2) the model constructed using the L1 scale for the discrete wavelet transform had greater estimation accuracy and stability than models established through traditional mathematical transformations or the high-frequency scale for discrete wavelet transform, with an R² value of 0.742 and a root mean square error (RMSE) of 0.936. The prediction model for relative chlorophyll content established in this study was more accurate for chlorophyll monitoring in pear trees, and thus, it provided a new method for rapid estimation. Moreover, the model provides an important theoretical basis for the efficient management of pear trees. Full article

(This article belongs to the Section Sensing and Imaging)

► Show Figures

Figure 1

18 pages, 5973 KiB

Open AccessArticle

Power Line Segmentation Algorithm Based on Lightweight Network and Residue-like Cross-Layer Feature Fusion

by Wenqiang Zhu, Huarong Ding, Gujing Han, Wei Wang, Minlong Li and Liang Qin

Sensors 2025, 25(11), 3551; https://doi.org/10.3390/s25113551 - 4 Jun 2025

Viewed by 271

Abstract

Power line segmentation plays a critical role in ensuring the safety of transmission line UAV inspection flights. To address the challenges of small target scale, complex backgrounds, and excessive model parameters in existing deep learning-based power line segmentation algorithms, this paper introduces RGS-UNet, [...] Read more.

Power line segmentation plays a critical role in ensuring the safety of transmission line UAV inspection flights. To address the challenges of small target scale, complex backgrounds, and excessive model parameters in existing deep learning-based power line segmentation algorithms, this paper introduces RGS-UNet, a lightweight segmentation model integrating a residual-like cross-layer feature fusion module. First, ResNet18 is adopted to reconstruct a UNet backbone network as an encoder module to enhance the network’s feature extraction capability for small targets. Second, ordinary convolution in the residual block of ResNet18 is optimized by introducing the Ghost Module, which significantly reduces the computational load of the model’s backbone network. Third, a residual-like addition method is designed to embed the SIMAM attention mechanism module into both encoder and decoder stages, which improves the model’s ability to extract power lines from complex backgrounds. Finally, the Mish activation function is applied in deep convolutional layers to maintain feature extraction accuracy and mitigate overfitting. Experimental results demonstrate that compared with classical UNet, the optimized network achieves 2.05% and 2.58% improvements in F1-Score and IoU, respectively, while reducing the parameter count to 57.25% of the original model. The algorithm achieves better performance improvements in both accuracy and lightweighting, making it suitable for edge-side deployment. Full article

(This article belongs to the Section Fault Diagnosis & Sensors)

► Show Figures

Figure 1

14 pages, 3919 KiB

Open AccessArticle

PCB Electronic Component Soldering Defect Detection Using YOLO11 Improved by Retention Block and Neck Structure

by Youzhi Xu, Hao Wu, Yulong Liu and Xing Zhang

Sensors 2025, 25(11), 3550; https://doi.org/10.3390/s25113550 - 4 Jun 2025

Viewed by 257

Abstract

Printed circuit board (PCB) assembly, on the basis of surface mount electronic component welding, is one of the most important electronic assembly processes, and its defect detection is also an important part of industrial generation. The traditional two-stage target detection algorithm model has [...] Read more.

Printed circuit board (PCB) assembly, on the basis of surface mount electronic component welding, is one of the most important electronic assembly processes, and its defect detection is also an important part of industrial generation. The traditional two-stage target detection algorithm model has a large number of parameters and the runtime is too long. The single-stage target detection algorithm has a faster running time, but the detection accuracy needs to be improved. To solve this problem, we innovated and modified the YOLO11n model. Firstly, we used the Retention Block (RetBlock) to improve the C3K2 module in the backbone, creating the RetC3K2 module, which makes up for the limitation of the original module’s limited, purely convolutional local receptive field. Secondly, the neck structure of the original model network is fused with a Multi-Branch Auxiliary Feature Pyramid Network (MAFPN) structure and turned into a multi-branch auxiliary neck network, which enhances the model’s ability to fuse multiple scaled characteristics and conveys diverse information about the gradient for the output layer. The improved YOLO11n model improves its mAP50 by 0.023 (2.5%) and mAP75 by 0.026 (2.8%) in comparison with the primitive model network, and detection precision is significantly improved, proving the superiority of our proposed approach. Full article

(This article belongs to the Section Electronic Sensors)

► Show Figures

Figure 1

26 pages, 2931 KiB

Open AccessArticle

CB-MTE: Social Bot Detection via Multi-Source Heterogeneous Feature Fusion

by Meng Cheng, Yuzhi Xiao, Tao Huang, Chao Lei and Chuang Zhang

Sensors 2025, 25(11), 3549; https://doi.org/10.3390/s25113549 - 4 Jun 2025

Viewed by 214

Abstract

Social bots increasingly mimic real users and collaborate in large-scale influence campaigns, distorting public perception and making their detection both critical and challenging. Traditional bot detection methods, constrained by single-source features, often fail to capture the complete behavioral and contextual characteristics of social [...] Read more.

Social bots increasingly mimic real users and collaborate in large-scale influence campaigns, distorting public perception and making their detection both critical and challenging. Traditional bot detection methods, constrained by single-source features, often fail to capture the complete behavioral and contextual characteristics of social bots, especially their dynamic behavioral evolution and group coordination tactics, resulting in feature incompleteness and reduced detection performance. To address this challenge, we propose CB-MTE, a social bot detection framework based on multi-source heterogeneous feature fusion. CB-MTE adopts a hierarchical architecture: user metadata is used to construct behavioral portraits, deep semantic representations are extracted from textual content via DistilBERT, and community-aware graph embeddings are learned through a combination of random walk and Skip-gram modeling. To mitigate feature redundancy and preserve structural consistency, manifold learning is applied for nonlinear dimensionality reduction, ensuring both local and global topology are maintained. Finally, a CatBoost-based collaborative reasoning mechanism enhances model robustness through ordered target encoding and symmetric tree structures. Experiments on the TwiBot-22 benchmark dataset demonstrate that CB-MTE significantly outperforms mainstream detection models in recognizing dynamic behavioral traits and detecting collaborative bot activities. These results confirm the framework’s capability to capture the complete behavioral and contextual characteristics of social bots through multi-source feature integration. Full article

(This article belongs to the Section Sensors and Robotics)

► Show Figures

Figure 1

18 pages, 4069 KiB

Open AccessArticle

Linking Neurocardiovascular Responses in the Active Stand Test to Adverse Outcomes: Insights from the Irish Longitudinal Study on Ageing (TILDA)

by Feng Xue and Roman Romero-Ortuno

Sensors 2025, 25(11), 3548; https://doi.org/10.3390/s25113548 - 4 Jun 2025

Viewed by 293

Abstract

Background: This study aimed to investigate the neurocardiovascular responses during an Active Stand (AS) test, utilizing both pre-processed and raw signals, to predict adverse health outcomes including orthostatic intolerance (OI) during the AS, and future falls and mortality. Methods: A total of 2794 [...] Read more.

Background: This study aimed to investigate the neurocardiovascular responses during an Active Stand (AS) test, utilizing both pre-processed and raw signals, to predict adverse health outcomes including orthostatic intolerance (OI) during the AS, and future falls and mortality. Methods: A total of 2794 participants from The Irish Longitudinal Study on Ageing (TILDA) were included. Continuous cardiovascular (heart rate (HR), systolic (sBP), and diastolic (dBP) blood pressure) and near infra-red spectroscopy-based neurovascular (tissue saturation index (TSI), oxygenated hemoglobin (O₂Hb), and deoxygenated hemoglobin (HHb)) signals were analyzed using Statistical Parametric Mapping (SPM) to identify significant group differences across health outcomes. Results: The results demonstrated that raw (unprocessed) signals, particularly O₂Hb and sBP/dBP, were more effective in capturing significant physiological differences associated with mortality and OI compared to pre-processed signals. Specifically, for OI, raw sBP and dBP captured significant changes across the entire test, whereas pre-processed signals showed intermittent significance. TSI captured OI only in its pre-processed form, at approximately 10 s post-stand. For mortality, raw O₂Hb was effective throughout the AS test. No significant differences were observed in either pre-processed or raw signals related to falls, suggesting that fall risk may require a multifactorial assessment beyond neurocardiovascular responses. Conclusions: These findings highlight the potential utility of raw signal analysis in improving risk stratification for OI and mortality, with further studies needed to validate these findings and refine predictive models for clinical applications. This study underscores the importance of retaining raw data for certain physiological assessments and provides a foundation for future work in developing machine-learning models for early health outcome detection. Full article

(This article belongs to the Special Issue (Bio)sensors for Physiological Monitoring)

► Show Figures

Figure 1

17 pages, 2495 KiB

Open AccessArticle

Developing a Low-Cost Device for Estimating Air–Water ΔpCO₂ in Coastal Environments

by Elizabeth B. Farquhar, Philip J. Bresnahan, Michael Tydings, Jessie C. Jarvis, Robert F. Whitehead and Dan Portelli

Sensors 2025, 25(11), 3547; https://doi.org/10.3390/s25113547 - 4 Jun 2025

Viewed by 407

Abstract

The ocean is one of the world’s largest anthropogenic carbon dioxide (CO₂) sinks, but closing the carbon budget is logistically difficult and expensive, and uncertainties in carbon fluxes and reservoirs remain. One specific challenge is that measuring the CO₂ flux [...] Read more.

The ocean is one of the world’s largest anthropogenic carbon dioxide (CO₂) sinks, but closing the carbon budget is logistically difficult and expensive, and uncertainties in carbon fluxes and reservoirs remain. One specific challenge is that measuring the CO₂ flux at the air–sea interface usually requires costly sensors or analyzers (USD > 30,000), which can limit observational capacity. Our group has developed and validated a low-cost ΔpCO₂ system, able to measure both pCO₂^water and pCO₂^air, for USD ~1400 to combat this limitation. The device is equipped with Internet of Things (IoT) capabilities and built around a USD ~100 pCO₂ K30 sensor at its core. Our Sensor for the Exchange of Atmospheric CO₂ with Water (SEACOW) may be placed in an observational network with traditional pCO₂ sensors or ∆pCO₂ sensors to extend the spatial coverage and resolution of monitoring systems. After calibration, the SEACOW reports atmospheric pCO₂ measurements that are within 2–3% of the measurements made with a calibrated LI-COR LI-850. We also demonstrate the SEACOW’s ability to capture diel pCO₂ cycling in seagrass, provide recommendations for SEACOW field deployments, and provide additional technical specifications for the SEACOW and for the K30 itself (e.g., air- and water-side 99.3% response time; 5.7 and 29.6 min, respectively). Full article

(This article belongs to the Section Environmental Sensing)

► Show Figures

Figure 1

18 pages, 2325 KiB

Open AccessArticle

Enhanced Rail Surface Defect Segmentation Using Polarization Imaging and Dual-Stream Feature Fusion

by Yucheng Pan, Jiasi Chen, Peiwen Wu, Hongsheng Zhong, Zihao Deng and Daozong Sun

Sensors 2025, 25(11), 3546; https://doi.org/10.3390/s25113546 - 4 Jun 2025

Viewed by 270

Abstract

Rail surface defects pose significant risks to the operational efficiency and safety of industrial equipment. Traditional visual defect detection methods typically rely on high-quality RGB images; however, they struggle in low-light conditions due to small, low-contrast defects that blend into complex backgrounds. Therefore, [...] Read more.

Rail surface defects pose significant risks to the operational efficiency and safety of industrial equipment. Traditional visual defect detection methods typically rely on high-quality RGB images; however, they struggle in low-light conditions due to small, low-contrast defects that blend into complex backgrounds. Therefore, this paper proposes a novel defect segmentation method leveraging a dual-stream feature fusion network that combines polarization images with DeepLabV3+. The approach utilizes the pruned MobileNetV3 as the backbone network, incorporating a coordinate attention mechanism for feature extraction. This reduces the number of model parameters and enhances computational efficiency. The dual-stream module implements cascade and addition strategies to effectively merge shallow and deep features from both the original and polarization images. This enhances the detection of low-contrast defects in complex backgrounds. Furthermore, the CBAM is integrated into the decoding area to refine feature fusion and mitigate the issue of missing small-target defects. Experimental results demonstrate that the enhanced DeepLabV3+ model outperforms existing models such as U-Net, PSPNet, and the original DeepLabV3+ in terms of MIoU and MPA metrics, achieving 73.00% and 80.59%, respectively. The comprehensive detection accuracy reaches 97.82%, meeting the demanding requirements for effective rail surface defect detection. Full article

(This article belongs to the Section Industrial Sensors)

► Show Figures

Figure 1

13 pages, 4369 KiB

Open AccessArticle

A Ka-Band Omnidirectional Metamaterial-Inspired Antenna for Sensing Applications

by Khan Md. Zobayer Hassan, Nantakan Wongkasem and Heinrich Foltz

Sensors 2025, 25(11), 3545; https://doi.org/10.3390/s25113545 - 4 Jun 2025

Viewed by 191

Abstract

A Ka-Band, 26.5–40 GHz, omnidirectional metamaterial-inspired antenna was designed, built, and tested to develop a simple printed compact (10.3 mm × 10.3 mm × 0.0787 mm) multiple-point sensor for air pollution monitoring. This Ka-band antenna generated a dual band at 27.49–29.74 GHz and [...] Read more.

A Ka-Band, 26.5–40 GHz, omnidirectional metamaterial-inspired antenna was designed, built, and tested to develop a simple printed compact (10.3 mm × 10.3 mm × 0.0787 mm) multiple-point sensor for air pollution monitoring. This Ka-band antenna generated a dual band at 27.49–29.74 GHz and 33.0–34.34 GHz. The VSWR values within the two bands are less than 1.5. The radiation and total efficiency are 97% and 92% in the first band and they are both 96% in the second band. The maximum gain is between 3.26 and 5.50 dBi and between 5.09 and 6.52 dBi in the first and second bands, respectively. The dual band is the key to enhancing the sensor’s detection accuracy. This Omni MTM-inspired antenna/sensor can effectively detect toxic and neurotoxic metal particles, i.e., lead, zinc, copper, and nickel, in evidently polluted living environments, such as factory/industrial environments, with different particle/mass concentrations. This sensor can be adapted to detect metal pollutants in different environments, such as water or other fluid-based matrices, and can also be applied to long-range communication repeaters and 5G harvesting energy devices, to name a few. Full article

(This article belongs to the Special Issue Electromagnetic Waves, Antennas and Sensor Technologies)

► Show Figures

Figure 1

24 pages, 4659 KiB

Open AccessArticle

Optimizing Autonomous Taxi Deployment for Safety at Skewed Intersections: A Simulation Study

by Zi Yang, Yaojie Yao and Liyan Zhang

Sensors 2025, 25(11), 3544; https://doi.org/10.3390/s25113544 - 4 Jun 2025

Viewed by 239

Abstract

This study optimizes the deployment of autonomous taxis for safety at skewed intersections through a simulation-based approach, identifying an optimal penetration rate and control strategies. Here, we investigate the safety impacts of autonomous taxis (ATs) at such intersections using a simulation-based approach, leveraging [...] Read more.

This study optimizes the deployment of autonomous taxis for safety at skewed intersections through a simulation-based approach, identifying an optimal penetration rate and control strategies. Here, we investigate the safety impacts of autonomous taxis (ATs) at such intersections using a simulation-based approach, leveraging the VISSIM traffic simulation tool and the Surrogate Safety Assessment Model (SSAM). Our study identifies an optimal AT penetration rate of approximately 0.5–0.7, as exceeding this range may lead to a decline in safety metrics such as TTC and PET. We find that connectivity among ATs does not linearly correlate with safety improvements, suggesting a nuanced approach to AT deployment is necessary. The “Normal” control strategy, which mimics human driving, shows a direct proportionality between AT penetration and TTC, indicating that not all levels of automation enhance safety. Our conflict analysis reveals distinct patterns for crossing, lane-change, and rear-end conflicts under various control strategies, underscoring the need for tailored approaches at skewed intersections. This research contributes to the discourse on AT safety and informs the development of traffic management strategies and policy frameworks that prioritize safety and efficiency in the context of skewed intersections. Full article

(This article belongs to the Special Issue Modeling and Control of Mixed Traffic Flow with Connected Automated Vehicle)

► Show Figures

Figure 1

19 pages, 4740 KiB

Open AccessArticle

Digital Twin Network-Based 6G Self-Evolution

by Yuhong Huang, Mancong Kang, Yanhong Zhu, Na Li, Guangyi Liu and Qixing Wang

Sensors 2025, 25(11), 3543; https://doi.org/10.3390/s25113543 - 4 Jun 2025

Viewed by 132

Abstract

Digital twins (DTs) will revolutionize network autonomy. Recent studies have promoted the idea of a DT-native 6G network, deeply integrating DTs into mobile network architectures to improve the timeliness of physical–digital synchronization and network optimizations. However, DTs have mainly acted just as a [...] Read more.

Digital twins (DTs) will revolutionize network autonomy. Recent studies have promoted the idea of a DT-native 6G network, deeply integrating DTs into mobile network architectures to improve the timeliness of physical–digital synchronization and network optimizations. However, DTs have mainly acted just as a tool for network autonomy, leading to a gap regarding the ultimate goal of network self-evolution. This paper analyzes future directions concerning DT-native networks. Specifically, the proposed architecture introduces a key concept called “future shots” that gives accurate network predictions under different time scales of self-evolution strategies for various network elements. To realize the future shots, we propose a long-term hierarchical convolutional graph attention model for cost-effective network predictions, a conditional hierarchical graph neural network for strategy generation, and methods for efficient small-to-large-scale interactions. The architecture is expected to facilitate high-level network autonomy for 6G networks. Full article

(This article belongs to the Special Issue Future Horizons in Networking: Exploring the Potential of 6G)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Sensors, Volume 25, Issue 11 (June-1 2025) – 322 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI