Sensors

30 pages, 7004 KB

Open AccessEditor’s ChoiceArticle

A Deep Learning-Based Sensing System for Identifying Salmon and Rainbow Trout Meat and Grading Freshness for Consumer Protection

by Hong-Dar Lin, Jun-Liang Chen and Chou-Hsien Lin

Sensors 2025, 25(20), 6299; https://doi.org/10.3390/s25206299 - 11 Oct 2025

Cited by 1 | Viewed by 1517

Abstract

Seafood fraud, such as mislabeling low-cost rainbow trout as premium salmon, poses serious food safety risks and damages consumer rights. To address this growing concern, this study develops a deep learning-based, smartphone-compatible sensing system for fish meat identification and salmon freshness grading. By [...] Read more.

Seafood fraud, such as mislabeling low-cost rainbow trout as premium salmon, poses serious food safety risks and damages consumer rights. To address this growing concern, this study develops a deep learning-based, smartphone-compatible sensing system for fish meat identification and salmon freshness grading. By providing consumers with real-time, image-based verification tools, the system supports informed purchasing decisions and enhances food safety. The system adopts a two-stage design: first classifying fish meat types, then grading salmon freshness into three levels based on visual cues. An improved DenseNet121 architecture, enhanced with global average pooling, dropout layers, and a customized output layer, improves accuracy and reduces overfitting, while transfer learning with partial layer freezing enhances efficiency by reducing training time without significant accuracy loss. Experimental results show that the two-stage method outperforms the one-stage approach and several baseline models, achieving robust accuracy in both classification and grading tasks. Sensitivity analysis demonstrates resilience to blur and camera tilt, though real-world adaptability under diverse lighting and packaging conditions remains a challenge. Overall, the proposed system represents a practical, consumer-oriented tool for seafood authentication and freshness evaluation, with potential to enhance food safety and consumer protection. Full article

(This article belongs to the Special Issue Intelligent Sensing Technologies for Monitoring Food Quality and Safety)

► Show Figures

Figure 1

39 pages, 445 KB

Open AccessEditor’s ChoiceArticle

A Study on IoT Device Authentication Using Artificial Intelligence

by Shahram Miri Kelaniki and Nikos Komninos

Sensors 2025, 25(18), 5809; https://doi.org/10.3390/s25185809 - 17 Sep 2025

Cited by 1 | Viewed by 2379

Abstract

Designing reliable authentication mechanisms for IoT devices is increasingly necessary to protect citizens’ private information and data. One of the most significant issues in today’s digital age is authentication. As IoT device technology advances and data grow rapidly, machine learning techniques improve the [...] Read more.

Designing reliable authentication mechanisms for IoT devices is increasingly necessary to protect citizens’ private information and data. One of the most significant issues in today’s digital age is authentication. As IoT device technology advances and data grow rapidly, machine learning techniques improve the accuracy and efficiency of authentication and offer advantages over traditional methods, making them valuable in both academia and industry. Device authentication aims to verify legitimate computing devices and identify impostors based on their behavioral data. This paper explores research that applies artificial intelligence algorithms to enhance device authentication mechanisms. We discuss AI authentication models, including deep learning algorithms, convolutional neural networks, and reinforcement learning. We also highlight research challenges and provide recommendations for future studies to support innovation in this field. Full article

(This article belongs to the Section Intelligent Sensors)

24 pages, 11665 KB

Open AccessEditor’s ChoiceArticle

Response of Nearby Sensors to Variable Doses of Nitrogen Fertilization in Winter Fodder Crops Under Mediterranean Climate

by Luís Silva, Caroline Brunelli, Raphael Moreira, Sofia Barbosa, Manuela Fernandes, Andreia Miguel, Benvindo Maçãs, Constantino Valero, Manuel Patanita, Fernando Cebola Lidon and Luís Alcino Conceição

Sensors 2025, 25(18), 5811; https://doi.org/10.3390/s25185811 - 17 Sep 2025

Viewed by 1192

Abstract

The sustainable intensification of forage production in Mediterranean climates requires technological solutions that optimize the use of agricultural inputs. This study aimed to evaluate the performance of proximal optical sensors in recommending and monitoring variable rate nitrogen fertilization in winter forage crops cultivated [...] Read more.

The sustainable intensification of forage production in Mediterranean climates requires technological solutions that optimize the use of agricultural inputs. This study aimed to evaluate the performance of proximal optical sensors in recommending and monitoring variable rate nitrogen fertilization in winter forage crops cultivated under Mediterranean conditions. A handheld multispectral active sensor (HMA), a multispectral camera on an unmanned aircraft vehicle (UAV), and one passive on-the-go sensor (OTG) were used to generate real-time nitrogen (N) application prescriptions. The sensors were assessed for their correlation with agronomic parameters such as plant fresh matter (PFM), plant dry matter (PDM), plant N content (PNC), crude protein (CP) in%, crude protein yield (CPyield) per unit of area, and N uptake (NUp). The real-time N fertilization stood out by promoting a 15.23% reduction in the total N fertilizer applied compared to a usual farmer-fixed dose of 150 kg ha⁻¹, saving 22.90 kg ha⁻¹ without compromising crop productivity. Additionally, NDVI_OTG showed moderate simple linear correlation with PFM (R² = 0.52), confirming its effectiveness in prescription based on vegetative vigor. UAV_II (NDVI after fertilization) showed even stronger correlations with CP (R² = 0.58), CPyield (R² = 0.53), and NUp (R² = 0.53), highlighting its sensitivity to physiological responses induced by N fertilization. Although the HMA sensor operates via point readings, it also proved effective, with significant correlations to NUp (R² = 0.55) and CPyield (R² = 0.53). It is concluded that integrating sensors enables both precise input prescription and efficient monitoring of plant physiological responses, fostering cost-effectiveness, sustainability, and improved agronomic efficiency. Full article

(This article belongs to the Section Smart Agriculture)

► Show Figures

Figure 1

23 pages, 979 KB

Open AccessEditor’s ChoiceArticle

Transparent EEG Analysis: Leveraging Autoencoders, Bi-LSTMs, and SHAP for Improved Neurodegenerative Diseases Detection

by Badr Mouazen, Ahmed Bendaouia, Omaima Bellakhdar, Khaoula Laghdaf, Aya Ennair, El Hassan Abdelwahed and Giovanni De Marco

Sensors 2025, 25(18), 5690; https://doi.org/10.3390/s25185690 - 12 Sep 2025

Cited by 1 | Viewed by 1591

Abstract

This study explores the use of deep learning techniques for classifying EEG signals in the context of Alzheimer’s disease (AD) and frontotemporal dementia (FTD). We propose a novel classification pipeline that combines autoencoders for feature extraction and bidirectional long short-term memory (Bi-LSTM) networks [...] Read more.

This study explores the use of deep learning techniques for classifying EEG signals in the context of Alzheimer’s disease (AD) and frontotemporal dementia (FTD). We propose a novel classification pipeline that combines autoencoders for feature extraction and bidirectional long short-term memory (Bi-LSTM) networks for analyzing patterns over time in EEG data. Given the complexity and high dimensionality of EEG signals, we employed an autoencoder to reduce data dimensionality while preserving key diagnostic features. The Bi-LSTM model effectively identified subtle temporal patterns in brain activity that are indicative of AD and FTD. To enhance interpretability, we applied SHapley Additive exPlanations (SHAP), providing insights into how individual features contribute to the model’s predictions. We evaluated our approach on a publicly available EEG dataset from OpenNeuro, which includes resting-state EEG recordings from 88 elderly participants—36 with AD, 23 with FTD, and 29 cognitively normal controls. EEG provides a non-invasive, cost-effective tool for brain monitoring, but presents challenges such as noise sensitivity and inter-subject variability. Despite these challenges, our approach achieved 98% accuracy while maintaining transparency, making it a promising tool for clinical applications in the diagnosis of neurodegenerative diseases. Full article

(This article belongs to the Special Issue Advanced Sensors in Brain–Computer Interfaces)

► Show Figures

Figure 1

17 pages, 1120 KB

Open AccessEditor’s ChoiceArticle

Effects of Induced Physical Fatigue on Heart Rate Variability in Healthy Young Adults

by Pei-Chun Kao and David J. Cornell

Sensors 2025, 25(17), 5572; https://doi.org/10.3390/s25175572 - 6 Sep 2025

Viewed by 4353

Abstract

Detecting physical fatigue can help prevent overexertion. While typically defined at the muscle level, systemic fatigue remains less clear. Heart rate variability (HRV) reflects autonomic adaptability to physical stressors and may provide insight into fatigue-related responses. This study investigated the impact of physical [...] Read more.

Detecting physical fatigue can help prevent overexertion. While typically defined at the muscle level, systemic fatigue remains less clear. Heart rate variability (HRV) reflects autonomic adaptability to physical stressors and may provide insight into fatigue-related responses. This study investigated the impact of physical fatigue on HRV and its correlation with endurance performance. Twenty participants (9 F, 11 M; 23.4 ± 5.0 y) walked on the treadmill at 1.25 m/s with progressively increased incline. HRV metrics were derived from baseline standing (STAND), pre-fatigued (PRE) and post-fatigued walking (POST). Time-domain HRV measures (lnTRI and lnTINN) were significantly reduced at POST compared to PRE or STAND (p < 0.05). Non-linear measures (DFA-α1, lnApEn, and lnSampEn) decreased at POST, while lnPoincaré SD2/SD1 increased. Normalized frequency-domain measures showed no condition effects. Baseline non-linear measures (lnApEn, lnSampEn, lnPoincaré SD2/SD1), normalized frequency measures and Total Power were significantly correlated with total fatiguing duration. Significant reductions in HRV and irregularity were observed post-fatigue. Greater baseline variability, irregularity, and high-frequency band power, reflecting parasympathetic activity, were associated with better endurance performance. Time-domain and non-linear measures were more sensitive to fatigue, whereas frequency-domain measures remain useful for identifying associations with endurance. The findings highlight HRV features that could enhance wearable sensing for fatigue and performance. Full article

(This article belongs to the Special Issue Smart Sensing Technology for Industry and Environmental Applications)

► Show Figures

Figure 1

24 pages, 4680 KB

Open AccessEditor’s ChoiceArticle

Indoor Pedestrian Location via Factor Graph Optimization Based on Sliding Windows

by Yu Cheng, Haifeng Li, Xixiang Liu, Shuai Chen and Shouzheng Zhu

Sensors 2025, 25(17), 5545; https://doi.org/10.3390/s25175545 - 5 Sep 2025

Viewed by 4336

Abstract

Global navigation satellite systems (GNSS) can provide high-quality location information in outdoor environments. In indoor environments, GNSS cannot achieve accurate and stable location information due to the obstruction and attenuation of buildings together with the influence of multipath effects. Due to the rapid [...] Read more.

Global navigation satellite systems (GNSS) can provide high-quality location information in outdoor environments. In indoor environments, GNSS cannot achieve accurate and stable location information due to the obstruction and attenuation of buildings together with the influence of multipath effects. Due to the rapid development of micro-electro-mechanical system (MEMS) sensors, today’s smartphones are equipped with various low-cost and small-volume MEMS sensors. Therefore, it is of great significance to study indoor pedestrian positioning technology based on smartphones. In order to provide pedestrians with high-precision and reliable location information in indoor environments, we propose a pedestrian dead reckoning (PDR) method based on Transformer+TCN (temporal convolutional network). Firstly, we use IMU (inertial measurement unit)/PDR pre-integration to suppress the inertial navigation divergence. Secondly, we propose a step length estimation algorithm based on Transformer+TCN. The Transformer and TCN networks are superimposed to improve the ability to capture complex dependencies and improve the generalization and reliability of step length estimation. Finally, we propose factor graph optimization (FGO) models based on sliding windows (SW-FGO) to provide accurate posture, which use accelerometer (ACC)/gyroscope/magnetometer (MAG) data to establish factors. We designed a fusion positioning estimation test and a comparison test on step length estimation algorithm. The results show that the fusion method based on SW-FGO proposed by us improves the positioning accuracy by 29.68% compared with the traditional FGO algorithm, and the absolute position error of the step length estimation algorithm based on Transformer+TCN in pocket mode is mitigated by 42.15% compared with the LSTM algorithm. The step length estimation model error of Transformer+TCN is 1.61%, and the step length estimation accuracy is improved by 24.41%. Full article

(This article belongs to the Section Navigation and Positioning)

► Show Figures

Figure 1

25 pages, 1716 KB

Open AccessEditor’s ChoiceArticle

Comparison of Wearable and Depth-Sensing Technologies with Electronic Walkway for Comprehensive Gait Analysis

by Marjan Nassajpour, Mahmoud Seifallahi, Amie Rosenfeld, Magdalena I. Tolea, James E. Galvin and Behnaz Ghoraani

Sensors 2025, 25(17), 5501; https://doi.org/10.3390/s25175501 - 4 Sep 2025

Cited by 3 | Viewed by 2615

Abstract

Accurate and scalable gait assessment is essential for clinical and research applications, including fall risk evaluation, rehabilitation monitoring, and early detection of neurodegenerative diseases. While electronic walkways remain the clinical gold standard, their high cost and limited portability restrict widespread use. Wearable inertial [...] Read more.

Accurate and scalable gait assessment is essential for clinical and research applications, including fall risk evaluation, rehabilitation monitoring, and early detection of neurodegenerative diseases. While electronic walkways remain the clinical gold standard, their high cost and limited portability restrict widespread use. Wearable inertial measurement units (IMUs) and markerless depth cameras have emerged as promising alternatives; however, prior studies have typically assessed these systems under tightly controlled conditions, with single participants in view, limited marker sets, and without direct cross-technology comparisons. This study addresses these gaps by simultaneously evaluating three sensing technologies—APDM wearable IMUs (tested in two separate configurations: foot-mounted and lumbar-mounted) and the Azure Kinect depth camera—against ProtoKinetics Zeno™ Walkway Gait Analysis System in a realistic clinical environment where multiple individuals were present in the camera’s field of view. Gait data from 20 older adults (mean age

70.06 \pm 9.45

years) performing Single-Task and Dual-Task walking trials were synchronously captured using custom hardware for precise temporal alignment. Eleven gait markers spanning macro, micro-temporal, micro-spatial, and spatiotemporal domains were compared using mean absolute error (MAE), Pearson correlation (r), and Bland–Altman analysis. Foot-mounted IMUs demonstrated the highest accuracy (MAE

= 0.00

–

6.12

,

r = 0.92

–

1.00

), followed closely by the Azure Kinect (MAE

= 0.01

–

6.07

,

r = 0.68

–0.98). Lumbar-mounted IMUs showed consistently lower agreement with the reference system. These findings provide the first comprehensive comparison of wearable and depth-sensing technologies with a clinical gold standard under real-world conditions and across an extensive set of gait markers. The results establish a foundation for deploying scalable, low-cost gait assessment systems in diverse healthcare contexts, supporting early detection, mobility monitoring, and rehabilitation outcomes across multiple patient populations. Full article

(This article belongs to the Special Issue The Sensing Technologies and Computational Methods for Biomedical Engineering: Next-Generation Perspectives)

► Show Figures

Figure 1

26 pages, 11096 KB

Open AccessEditor’s ChoiceArticle

A Novel ML-Powered Nanomembrane Sensor for Smart Monitoring of Pollutants in Industrial Wastewater

by Gabriele Cavaliere, Luca Tari, Francesco Siconolfi, Hamza Rehman, Polina Kuzhir, Antonio Maffucci and Luigi Ferrigno

Sensors 2025, 25(17), 5390; https://doi.org/10.3390/s25175390 - 1 Sep 2025

Cited by 1 | Viewed by 1266

Abstract

This study presents a comprehensive analysis aimed at validating the use of an innovative nanosensor based on graphitic nanomembranes for the smart monitoring of industrial wastewater. The validation of the potential of the nanosensor was carried out through the development of advanced analytical [...] Read more.

This study presents a comprehensive analysis aimed at validating the use of an innovative nanosensor based on graphitic nanomembranes for the smart monitoring of industrial wastewater. The validation of the potential of the nanosensor was carried out through the development of advanced analytical methodologies, a direct experimental comparison with commercially available electrode sensors commonly used for the detection of chemical species, and the evaluation of performance under conditions very similar to real-world field applications. The investigation involved a series of controlled experiments using an organic pollutant—benzoquinone—at varying concentrations. Initially, data analysis was performed using classical linear regression models, representing a conventional approach in chemical analysis. Subsequently, a more advanced methodology was implemented, incorporating machine-learning techniques to train a classifier capable of detecting the presence of pollutants in water samples. The study builds upon an experimental protocol previously developed by the authors for the nanomembranes, based on electrochemical impedance spectroscopy. The results clearly demonstrate that integrating the nanosensor with machine-learning algorithms yields significant performance. The intrinsic properties of the nanosensor make it well-suited for potential integration into field-deployable platforms, offering a real-time, cost-effective, and high-performance solution for the detection and quantification of contaminants in wastewater. These features position the nanomembrane-based sensor as a promising alternative to overcome current technological limitations in this domain. Full article

(This article belongs to the Special Issue Sensors for Water Quality Monitoring and Assessment)

► Show Figures

Figure 1

24 pages, 6086 KB

Open AccessEditor’s ChoiceArticle

Design of a Mobile and Electromagnetic Emissions-Compliant Brain Positron Emission Tomography (PET) Scanner

by Cristian Fuentes, Marina Béguin, Volker Commichau, Judith Flock, Anthony J. Lomax, Shubhangi Makkar, Keegan McNamara, John O. Prior, Christian Ritzer, Carla Winterhalter and Günther Dissertori

Sensors 2025, 25(17), 5344; https://doi.org/10.3390/s25175344 - 28 Aug 2025

Cited by 1 | Viewed by 1327

Abstract

This paper presents the development of two mobile brain Positron Emission Tomography (PET) scanners under the PETITION project, designed for Intensive Care Units (ICUs) and Proton Beam Therapy (PBT) applications. The ICU scanner facilitates bedside imaging for critically ill patients, while the PBT [...] Read more.

This paper presents the development of two mobile brain Positron Emission Tomography (PET) scanners under the PETITION project, designed for Intensive Care Units (ICUs) and Proton Beam Therapy (PBT) applications. The ICU scanner facilitates bedside imaging for critically ill patients, while the PBT scanner enables undisturbed proton beam irradiation during imaging. Key aspects of the hardware design, including modular detectors and electromagnetic interference considerations, are discussed along with preliminary performance evaluations. Operational testing, employing a ²²Na source and a hot-rod phantom, was conducted to determine the timing resolution (548

p

s

), energy resolution (11.4%) and a qualitative spatial resolution (around 2.2

m

m

). Our study presents findings on the ICU PET scanner’s electromagnetic emissions measured in a controlled EMC testing facility, where all the emissions tests performed comply with the standard EN 60601-1-2 (radiated emissions 15 dB below regulatory limits in the frequency range of 30

M

Hz

to 1 G

Hz

). Full article

(This article belongs to the Collection Biomedical Imaging & Instrumentation)

► Show Figures

Figure 1

15 pages, 968 KB

Open AccessEditor’s ChoiceArticle

Validity of AI-Driven Markerless Motion Capture for Spatiotemporal Gait Analysis in Stroke Survivors

by Balsam J. Alammari, Brandon Schoenwether, Zachary Ripic, Neva Kirk-Sanchez, Moataz Eltoukhy and Lauri Bishop

Sensors 2025, 25(17), 5315; https://doi.org/10.3390/s25175315 - 27 Aug 2025

Cited by 1 | Viewed by 1932

Abstract

Gait recovery after stroke is a primary goal of rehabilitation, therefore it is imperative to develop technologies that accurately identify gait impairments after stroke. Markerless motion capture (MMC) is an emerging technology that has been validated in healthy individuals. Our study aims to [...] Read more.

Gait recovery after stroke is a primary goal of rehabilitation, therefore it is imperative to develop technologies that accurately identify gait impairments after stroke. Markerless motion capture (MMC) is an emerging technology that has been validated in healthy individuals. Our study aims to evaluate the validity of MMC against an instrumented walkway system (IWS) commonly used to evaluate gait in stroke survivors. Nineteen participants performed three comfortable speed (CS) and three fastest speed (FS) walking trials simultaneously recorded with IWS and MMC system, KinaTrax (HumanVersion 8.2, KinaTrax Inc., Boca Raton, FL, USA). Pearson’s correlation coefficient and intraclass correlation coefficient (ICC (3,1), 95%CI) were used to evaluate the agreement and consistency between systems. Furthermore, Bland–Altman plots were used to estimate bias and Limits of Agreement (LoA). For both CS and FS, agreements between MMC and IWS were good to excellent in all parameters except for non-paretic single-limb support time (SLS), which revealed moderate agreement during CS. Additionally, stride width and paretic SLS showed poor agreement in both conditions. Biases eliminated systematic errors, with variable LoAs in all parameters during both conditions. Findings indicated high validity of MMC in measuring spatiotemporal gait parameters in stroke survivors. Further validity work is warranted. Full article

(This article belongs to the Special Issue Human Activity Recognition Based on Sensors: Challenges and Perspectives)

► Show Figures

Figure 1

20 pages, 1235 KB

Open AccessEditor’s ChoiceArticle

Variable-Speed UAV Path Optimization Based on the CRLB Criterion for Passive Target Localization

by Lijia Chen, Chengfeng You, Yixin Wang and Xueting Li

Sensors 2025, 25(17), 5297; https://doi.org/10.3390/s25175297 - 26 Aug 2025

Cited by 2 | Viewed by 1248

Abstract

The performance of passive target localization is significantly influenced by the positions of unmanned aerial vehicle swarms (UAVs). In this paper, we investigate the problem of UAV path optimization to enhance the localization accuracy. Firstly, a passive target localization signal model based on [...] Read more.

The performance of passive target localization is significantly influenced by the positions of unmanned aerial vehicle swarms (UAVs). In this paper, we investigate the problem of UAV path optimization to enhance the localization accuracy. Firstly, a passive target localization signal model based on the time difference of arrival (TDOA) algorithm, which is then improved by the Chan method and Taylor series expansion, is established. Secondly, the Cramer–Rao lower bound (CRLB) of the modified TDOA algorithm is derived and adopted as the evaluation criterion to optimize the UAVs’ positions at each time step. Different from the existing works, in this paper, we consider the UAVs to have variable speed; therefore, the feasible region of the UAVs’ positions is changed from a circle into an annular region, which will extend the feasible region, enhancing the localization accuracy while increasing the computation complexity. Thirdly, to improve the efficiency of the UAV path optimization algorithm, the particle swarm optimization (PSO) algorithm is applied to search for the optimal positions of the UAVs for the next time step. Finally, numerical simulations are conducted to verify the validity and effectiveness of the proposals in this paper. Full article

(This article belongs to the Special Issue Radar Target Detection, Imaging and Recognition)

► Show Figures

Figure 1

35 pages, 6244 KB

Open AccessEditor’s ChoiceReview

Comprehensive Analysis of FBG and Distributed Rayleigh, Brillouin, and Raman Optical Sensor-Based Solutions for Road Infrastructure Monitoring Applications

by Ugis Senkans, Nauris Silkans, Sandis Spolitis and Janis Braunfelds

Sensors 2025, 25(17), 5283; https://doi.org/10.3390/s25175283 - 25 Aug 2025

Cited by 2 | Viewed by 2203

Abstract

This study focuses on a comprehensive analysis of the common methods for road infrastructure monitoring, as well as the perspective of various fiber-optic sensor (FOS) realization solutions in road monitoring applications. Fiber-optic sensors are a topical technology that ensures multiple advantages such as [...] Read more.

This study focuses on a comprehensive analysis of the common methods for road infrastructure monitoring, as well as the perspective of various fiber-optic sensor (FOS) realization solutions in road monitoring applications. Fiber-optic sensors are a topical technology that ensures multiple advantages such as passive nature, immunity to electromagnetic interference, multiplexing capabilities, high sensitivity, and spatial resolution, as well as remote operation and multiple physical parameter monitoring, hence offering embedment potential within the road pavement structure for needed smart road solutions. The main key factors that affect FOS-based road monitoring scenarios and configurations are analyzed within this review. One such factor is technology used for optical sensing—fiber Bragg grating (FBG), Brillouin, Rayleigh, or Raman-based sensing. A descriptive comparison is made comparing typical sensitivity, spatial resolution, measurement distance, and applications. Technological approaches for monitoring physical parameters, such as strain, temperature, vibration, humidity, and pressure, as a means of assessing road infrastructure integrity and smart application integration, are also evaluated. Another critical aspect concerns spatial positioning, focusing on the point, quasi-distributed, and distributed methodologies. Lastly, the main topical FOS-based application areas are discussed, analyzed, and evaluated. Full article

(This article belongs to the Special Issue Advances in Fiber Bragg Grating (FBG) and Fiber Optic Sensor Technology)

► Show Figures

Figure 1

34 pages, 1151 KB

Open AccessEditor’s ChoiceArticle

Innovative Technologies to Improve Occupational Safety in Mining and Construction Industries—Part I

by Paweł Bęś, Paweł Strzałkowski, Justyna Górniak-Zimroz, Mariusz Szóstak and Mateusz Janiszewski

Sensors 2025, 25(16), 5201; https://doi.org/10.3390/s25165201 - 21 Aug 2025

Cited by 3 | Viewed by 5295

Abstract

Innovative technologies have been helping to improve comfort and safety at work in high-risk sectors for years. The study analysed the impact, along with an assessment of potential implementations (opportunities and limitations) of innovative technological solutions for improving occupational safety in two selected [...] Read more.

Innovative technologies have been helping to improve comfort and safety at work in high-risk sectors for years. The study analysed the impact, along with an assessment of potential implementations (opportunities and limitations) of innovative technological solutions for improving occupational safety in two selected sectors of the economy: mining and construction. The technologies evaluated included unmanned aerial vehicles and inspection robots, the Internet of Things and sensors, artificial intelligence, virtual and augmented reality, innovative individual and collective protective equipment, and exoskeletons. Due to the extensive nature of the obtained materials, the research description has been divided into two articles (Part I and Part II). This article presents the first three technologies. After the scientific literature from the Scopus database was analysed, some research gaps that need to be filled were identified. In addition to the obvious benefits of increased occupational safety for workers, innovative technological solutions also offer employers several economic advantages that affect the industry’s sustainability. Innovative technologies are playing an increasingly important role in improving safety in mining and construction. However, further integration and overcoming implementation barriers, such as the need for changes in education, are needed to realise their full potential. Full article

(This article belongs to the Section Industrial Sensors)

► Show Figures

Figure 1

36 pages, 2830 KB

Open AccessEditor’s ChoiceReview

Real-Time, Continuous Monitoring of Tissue Chips as an Emerging Opportunity for Biosensing

by John S. Cognetti and Benjamin L. Miller

Sensors 2025, 25(16), 5153; https://doi.org/10.3390/s25165153 - 19 Aug 2025

Cited by 1 | Viewed by 3242

Abstract

Tissue chips (TCs), otherwise known as organs-on-a-chip (OoC), organ chips (OCs), or microphysiological systems (MPS), are rapidly gaining prominence as an extension of or even replacement for traditional animal models of disease physiology. They also have recognized utility in the context of drug [...] Read more.

Tissue chips (TCs), otherwise known as organs-on-a-chip (OoC), organ chips (OCs), or microphysiological systems (MPS), are rapidly gaining prominence as an extension of or even replacement for traditional animal models of disease physiology. They also have recognized utility in the context of drug development: for example, data from TCs can now be submitted in place of some animal testing to the FDA. In principle, TCs are structured to allow measurement of any number of outputs that yield information about the tissue. However, to date, measurements made during experiments with TCs have been largely restricted to immunofluorescence microscopy and benchtop assays performed on media extracted from the cell culture within the device. With the development of biosensors that are sensitive and have an ever-shrinking footprint, on-board biosensing is now in the early stages of exploration. This review discusses the importance of tissue chips and the advances in sensing that will aid the complexity and utility of tissue chip research moving forward. We cover several sensing modalities, including electrical and optical sensing modes. Finally, challenges and opportunities for the future are discussed. Full article

(This article belongs to the Special Issue Feature Review Papers in Optical Sensors)

► Show Figures

Figure 1

29 pages, 1052 KB

Open AccessEditor’s ChoiceReview

Prediction of Soil Properties Using Vis-NIR Spectroscopy Combined with Machine Learning: A Review

by Su Kyeong Shin, Seung Jun Lee and Jin Hee Park

Sensors 2025, 25(16), 5045; https://doi.org/10.3390/s25165045 - 14 Aug 2025

Cited by 9 | Viewed by 5854

Abstract

Stable crop yields require an appropriate supply of essential soil nutrients such as nitrogen (N), phosphorus (P), and potassium (K) based on the accurate diagnosis of soil nutrient status. Traditional laboratory analysis of soil nutrients is often complicated and time-consuming and does not [...] Read more.

Stable crop yields require an appropriate supply of essential soil nutrients such as nitrogen (N), phosphorus (P), and potassium (K) based on the accurate diagnosis of soil nutrient status. Traditional laboratory analysis of soil nutrients is often complicated and time-consuming and does not provide real-time nutrient status. Visible–near-infrared (Vis-NIR) spectroscopy has emerged as a non-destructive and rapid method for estimating soil nutrient levels. Vis-NIR spectra reflect sample characteristics as the peak intensities; however, they are often affected by various artifacts and complex variables. Since Vis-NIR spectroscopy does not directly measure nutrient levels in soil, improving estimation accuracy is essential. For spectral preprocessing, the most important aspect is to develop an appropriate preprocessing strategy based on the characteristics of the data and identify artifacts such as noise, baseline drift, and scatter in the spectral data. Machine learning-based modeling techniques such as partial least-squares regression (PLSR) and support vector machine regression (SVMR) enhance estimation accuracy by capturing complex patterns of spectral data. Therefore, this review focuses on the use of Vis-NIR spectroscopy for evaluating soil properties including soil water content, organic carbon (C), and nutrients and explores its potential for real-time field application through spectral preprocessing and machine learning algorithms. Vis-NIR spectroscopy combined with machine learning is expected to enable more efficient and site-specific nutrient management, thereby contributing to sustainable agricultural practices. Full article

(This article belongs to the Special Issue Advancing Land Monitoring through Synergistic Harmonization of Optical, Radar and Lidar Satellite Technologies)

► Show Figures

Figure 1

23 pages, 3199 KB

Open AccessEditor’s ChoiceArticle

A Motion Segmentation Dynamic SLAM for Indoor GNSS-Denied Environments

by Yunhao Wu, Ziyao Zhang, Haifeng Chen and Jian Li

Sensors 2025, 25(16), 4952; https://doi.org/10.3390/s25164952 - 10 Aug 2025

Cited by 2 | Viewed by 1580

Abstract

In GNSS-deprived settings, such as indoor and underground environments, research on simultaneous localization and mapping (SLAM) technology remains a focal point. Addressing the influence of dynamic variables on positional precision and constructing a persistent map comprising solely static elements are pivotal objectives in [...] Read more.

In GNSS-deprived settings, such as indoor and underground environments, research on simultaneous localization and mapping (SLAM) technology remains a focal point. Addressing the influence of dynamic variables on positional precision and constructing a persistent map comprising solely static elements are pivotal objectives in visual SLAM for dynamic scenes. This paper introduces optical flow motion segmentation-based SLAM(OS-SLAM), a dynamic environment SLAM system that incorporates optical flow motion segmentation for enhanced robustness. Initially, a lightweight multi-scale optical flow network is developed and optimized using multi-scale feature extraction and update modules to enhance motion segmentation accuracy with rigid masks while maintaining real-time performance. Subsequently, a novel fusion approach combining the YOLO-fastest method and Rigidmask fusion is proposed to mitigate mis-segmentation errors of static backgrounds caused by non-rigid moving objects. Finally, a static dense point cloud map is generated by filtering out abnormal point clouds. OS-SLAM integrates optical flow estimation with motion segmentation to effectively reduce the impact of dynamic objects. Experimental findings from the Technical University of Munich (TUM) dataset demonstrate that the proposed method significantly outperforms ORB-SLAM3 in handling high dynamic sequences, achieving a reduction of 91.2% in absolute position error (APE) and 45.1% in relative position error (RPE) on average. Full article

(This article belongs to the Collection Navigation Systems and Sensors)

► Show Figures

Figure 1

20 pages, 10724 KB

Open AccessEditor’s ChoiceArticle

Leakage Detection Using Distributed Acoustic Sensing in Gas Pipelines

by Mouna-Keltoum Benabid, Peyton Baumgartner, Ge Jin and Yilin Fan

Sensors 2025, 25(16), 4937; https://doi.org/10.3390/s25164937 - 10 Aug 2025

Cited by 2 | Viewed by 5809

Abstract

This study investigates the performance of Distributed Acoustic Sensing (DAS) for detecting gas pipeline leaks under controlled experimental conditions, using multiple fiber cable types deployed both internally and externally. A 21 m steel pipeline with a 1 m test section was configured to [...] Read more.

This study investigates the performance of Distributed Acoustic Sensing (DAS) for detecting gas pipeline leaks under controlled experimental conditions, using multiple fiber cable types deployed both internally and externally. A 21 m steel pipeline with a 1 m test section was configured to simulate leakage scenarios with varying leak sizes (¼”, ½”, ¾”, and 1”), orientations (top, side, bottom), and flow velocities (2–18 m/s). Experiments were conducted under two installation conditions: a supported pipeline mounted on tripods, and a buried pipeline laid on the ground and covered with sand. Four fiber deployment methods were tested: three internal cables of varying geometries and one externally mounted straight cable. DAS data were analyzed using both time-domain vibration intensity and frequency-domain spectral methods. The results demonstrate that leak detectability is influenced by multiple interacting factors, including flow rate, leak size and orientation, pipeline installation method, and fiber cable type and deployment approach. Internally deployed black and flat cables exhibited higher sensitivity to leak-induced vibrations, particularly at higher flow velocities, larger leak sizes, and for bottom-positioned leaks. The thick internal cable showed limited response due to its wireline-like construction. In contrast, the external straight cable responded selectively, with performance dependent on mechanical coupling. Overall, leakage detectability was reduced in the buried configuration due to damping effects. The novelty of this work lies in the successful detection of gas leaks using internally deployed fiber optic cables, which has not been demonstrated in previous studies. This deployment approach is practical for field applications, particularly for pipelines that cannot be inspected using conventional methods, such as unpiggable pipelines. Full article

(This article belongs to the Special Issue Optical Sensors for Industrial Applications)

► Show Figures

Figure 1

23 pages, 1519 KB

Open AccessEditor’s ChoiceArticle

Machine Learning-Based Assessment of Parkinson’s Disease Symptoms Using Wearable and Smartphone Sensors

by Tomasz Gutowski, Olga Stodulska, Aleksandra Ćwiklińska, Katarzyna Gutowska, Kamila Kopeć, Marta Betka, Ryszard Antkiewicz, Dariusz Koziorowski and Stanisław Szlufik

Sensors 2025, 25(16), 4924; https://doi.org/10.3390/s25164924 - 9 Aug 2025

Cited by 3 | Viewed by 2117

Abstract

This study explores the use of machine learning models to assess the severity of Parkinson’s disease symptoms based on data from wearable and smartphone sensors. It presents models to predict the severities of individual symptoms—tremor, bradykinesia, stiffness, and dyskinesia—as well as the overall [...] Read more.

This study explores the use of machine learning models to assess the severity of Parkinson’s disease symptoms based on data from wearable and smartphone sensors. It presents models to predict the severities of individual symptoms—tremor, bradykinesia, stiffness, and dyskinesia—as well as the overall state of patients, using both clinician and patient self-assessments as labels. The dataset, although limited and imbalanced, enabled the identification of key trends. The best performance was achieved when combining data from both the MYO armband and smartphone, and when using patient self-assessments as targets. Tremor was the most predictable symptom, while others proved more challenging—especially at higher severity levels, which were poorly represented in the dataset. These results highlight the value of multimodal data and the importance of patient input in symptom monitoring. However, they also point to the need for more balanced and extensive datasets to improve prediction accuracy across all severity levels and symptoms. Full article

(This article belongs to the Special Issue Sensors and Artificial Intelligence Technologies in Neurodegenerative Disease Diagnosis)

► Show Figures

Figure 1

16 pages, 1140 KB

Open AccessEditor’s ChoiceReview

Future Designs of Clinical Trials in Nephrology: Integrating Methodological Innovation and Computational Power

by Camillo Tancredi Strizzi and Francesco Pesce

Sensors 2025, 25(16), 4909; https://doi.org/10.3390/s25164909 - 8 Aug 2025

Cited by 1 | Viewed by 2596

Abstract

Clinical trials in nephrology have historically been hindered by significant challenges, including slow disease progression, patient heterogeneity, and recruitment difficulties. While recent therapeutic breakthroughs have transformed care, they have also created a ‘paradox of success’ by lowering baseline event rates, further complicating traditional [...] Read more.

Clinical trials in nephrology have historically been hindered by significant challenges, including slow disease progression, patient heterogeneity, and recruitment difficulties. While recent therapeutic breakthroughs have transformed care, they have also created a ‘paradox of success’ by lowering baseline event rates, further complicating traditional trial designs. We hypothesize that integrating innovative trial methodologies with advanced computational tools is essential for overcoming these hurdles and accelerating therapeutic development in kidney disease. This narrative review synthesizes the literature on persistent challenges in nephrology trials and explores methodological innovations. It investigates the transformative impact of computational tools, specifically Artificial Intelligence (AI), techniques like Augmented Reality (AR) and Conditional Tabular Generative Adversarial Networks (CTGANs), in silico clinical trials (ISCTs) and Digital Health Technologies across the research lifecycle. Key methodological innovations include adaptive designs, pragmatic trials, real-world evidence, and validated surrogate endpoints. AI offers transformative potential in optimizing trial design, accelerating patient stratification, and enabling complex data analysis, while AR can improve procedural accuracy, and CTGANs can augment scarce datasets. ISCTs provide complementary capabilities for simulating drug effects and optimizing designs using virtual patient cohorts. The future of clinical research in nephrology lies in the synergistic convergence of methodological and computational innovation. This integrated approach offers a pathway for conducting more efficient, precise, and patient-centric trials, provided that critical barriers related to data quality, model validation, regulatory acceptance, and ethical implementation are addressed. Full article

(This article belongs to the Section Biomedical Sensors)

► Show Figures

Figure 1

21 pages, 4772 KB

Open AccessEditor’s ChoiceArticle

Integrating Environmental Sensing into Cargo Bikes for Pollution-Aware Logistics in Last-Mile Deliveries

by Leonardo Cameli, Margherita Pazzini, Riccardo Ceriani, Valeria Vignali, Andrea Simone and Claudio Lantieri

Sensors 2025, 25(15), 4874; https://doi.org/10.3390/s25154874 - 7 Aug 2025

Cited by 1 | Viewed by 1386

Abstract

Cycling represents a significant share of urban transportation, especially in terms of last-mile delivery. It has clear benefits for delivery times, as well as for environmental issues related to freight distribution. Furthermore, the increasing accessibility of low-cost environmental sensors (LCSs) provides an opportunity [...] Read more.

Cycling represents a significant share of urban transportation, especially in terms of last-mile delivery. It has clear benefits for delivery times, as well as for environmental issues related to freight distribution. Furthermore, the increasing accessibility of low-cost environmental sensors (LCSs) provides an opportunity for urban monitoring in any situation. Moving in this direction, this research aims to investigate the use of LCSs to monitor particulate PM2.5 and PM10 levels and map them over delivery ride paths. The calibration process took 49 days of measurements into account, spanning different seasonal conditions (from May 2024 to November 2024). The employment of multiple linear regression and robust regression revealed a strong correlation between pollutant levels from two sources and other factors such as temperature and humidity. Subsequently, a one-month trial was carried out in the city of Faenza (Italy). In this study, a commercially available LCS was mounted on a cargo bike for measurement during delivery processes. This approach was adopted to reveal biker exposure to air pollutants. In this way, the operator’s route would be designed to select the best route in terms of delivery timing and polluting factors in the future. Furthermore, the integration of environmental monitoring to map urban environments has the potential to enhance the accuracy of local pollution mapping, thereby supporting municipal efforts to inform citizens and develop targeted air quality strategies. Full article

(This article belongs to the Section Environmental Sensing)

► Show Figures

Figure 1

11 pages, 3181 KB

Open AccessEditor’s ChoiceArticle

Development of a Three-Dimensional Nanostructure SnO₂-Based Gas Sensor for Room-Temperature Hydrogen Detection

by Zhilong Song, Yi Tian, Yue Kang and Jia Yan

Sensors 2025, 25(15), 4784; https://doi.org/10.3390/s25154784 - 3 Aug 2025

Cited by 2 | Viewed by 2200

Abstract

The development of gas sensors with high sensitivity and low operating temperatures is essential for practical applications in environmental monitoring and industrial safety. SnO₂-based gas sensors, despite their widespread use, often suffer from high working temperatures and limited sensitivity to H [...] Read more.

The development of gas sensors with high sensitivity and low operating temperatures is essential for practical applications in environmental monitoring and industrial safety. SnO₂-based gas sensors, despite their widespread use, often suffer from high working temperatures and limited sensitivity to H₂ gas, which presents significant challenges for their performance and application. This study addresses these issues by introducing a novel SnO₂-based sensor featuring a three-dimensional (3D) nanostructure, designed to enhance sensitivity and allow for room-temperature operation. This work lies in the use of a 3D anodic aluminum oxide (AAO) template to deposit SnO₂ nanoparticles through ultrasonic spray pyrolysis, followed by modification with platinum (Pt) nanoparticles to further enhance the sensor’s response. The as-prepared sensors were extensively characterized, and their H₂ sensing performance was evaluated. The results show that the 3D nanostructure provides a uniform and dense distribution of SnO₂ nanoparticles, which significantly improves the sensor’s sensitivity and repeatability, especially in H₂ detection at room temperature. This work demonstrates the potential of utilizing 3D nanostructures to overcome the traditional limitations of SnO₂-based sensors. Full article

(This article belongs to the Special Issue Bio- and Chemical Sensors: Young Scientists in the Asia-Pacific Region)

► Show Figures

Figure 1

21 pages, 1928 KB

Open AccessEditor’s ChoiceArticle

A CNN-Transformer Hybrid Framework for Multi-Label Predator–Prey Detection in Agricultural Fields

by Yifan Lyu, Feiyu Lu, Xuaner Wang, Yakui Wang, Zihuan Wang, Yawen Zhu, Zhewei Wang and Min Dong

Sensors 2025, 25(15), 4719; https://doi.org/10.3390/s25154719 - 31 Jul 2025

Cited by 3 | Viewed by 1524

Abstract

Accurate identification of predator–pest relationships is essential for implementing effective and sustainable biological control in agriculture. However, existing image-based methods struggle to recognize insect co-occurrence under complex field conditions, limiting their ecological applicability. To address this challenge, we propose a hybrid deep learning [...] Read more.

Accurate identification of predator–pest relationships is essential for implementing effective and sustainable biological control in agriculture. However, existing image-based methods struggle to recognize insect co-occurrence under complex field conditions, limiting their ecological applicability. To address this challenge, we propose a hybrid deep learning framework that integrates convolutional neural networks (CNNs) and Transformer architectures for multi-label recognition of predator–pest combinations. The model leverages a novel co-occurrence attention mechanism to capture semantic relationships between insect categories and employs a pairwise label matching loss to enhance ecological pairing accuracy. Evaluated on a field-constructed dataset of 5,037 images across eight categories, the model achieved an F1-score of 86.5%, mAP50 of 85.1%, and demonstrated strong generalization to unseen predator–pest pairs with an average F1-score of 79.6%. These results outperform several strong baselines, including ResNet-50, YOLOv8, and Vision Transformer. This work contributes a robust, interpretable approach for multi-object ecological detection and offers practical potential for deployment in smart farming systems, UAV-based monitoring, and precision pest management. Full article

(This article belongs to the Special Issue Sensor and AI Technologies in Intelligent Agriculture: 2nd Edition)

► Show Figures

Figure 1

20 pages, 5696 KB

Open AccessEditor’s ChoiceArticle

Classification of User Behavior Patterns for Indoor Navigation Problem

by Aleksandra Borsuk, Andrzej Chybicki and Michał Zieliński

Sensors 2025, 25(15), 4673; https://doi.org/10.3390/s25154673 - 29 Jul 2025

Cited by 2 | Viewed by 1253

Abstract

Indoor navigation poses persistent challenges due to the limitations of traditional positioning systems within buildings. In this study, we propose a novel approach to address this issue—not by continuously tracking the user’s location, but by estimating their position based on how closely their [...] Read more.

Indoor navigation poses persistent challenges due to the limitations of traditional positioning systems within buildings. In this study, we propose a novel approach to address this issue—not by continuously tracking the user’s location, but by estimating their position based on how closely their observed behavior matches the expected progression along a predefined route. This concept, while not universally applicable, is well-suited for specific indoor navigation scenarios, such as guiding couriers or delivery personnel through complex residential buildings. We explore this idea in detail in our paper. To implement this behavior-based localization, we introduce an LSTM-based method for classifying user behavior patterns, including standing, walking, and using stairs or elevators, by analyzing velocity sequences derived from smartphone sensors’ data. The developed model achieved 75% accuracy for individual activity type classification within one-second time windows, and 98.6% for full-sequence classification through majority voting. These results confirm the viability of real-time activity recognition as the foundation for a navigation system that aligns live user behavior with pre-recorded patterns, offering a cost-effective alternative to infrastructure-heavy indoor positioning systems. Full article

(This article belongs to the Special Issue Indoor Navigation: Indoor Positioning System Using Sensing Technologies)

► Show Figures

Figure 1

17 pages, 4137 KB

Open AccessEditor’s ChoiceArticle

Satellite Positioning Accuracy Improvement in Urban Canyons Through a New Weight Model Utilizing GPS Signal Strength Variability

by Hye-In Kim and Kwan-Dong Park

Sensors 2025, 25(15), 4678; https://doi.org/10.3390/s25154678 - 29 Jul 2025

Cited by 3 | Viewed by 3462

Abstract

Urban environments present substantial obstacles to GPS positioning accuracy, primarily due to multipath interference and limited satellite visibility. To address these challenges, we propose a novel weighting approach, referred to as the HK model, that enhances real-time GPS positioning performance by leveraging the [...] Read more.

Urban environments present substantial obstacles to GPS positioning accuracy, primarily due to multipath interference and limited satellite visibility. To address these challenges, we propose a novel weighting approach, referred to as the HK model, that enhances real-time GPS positioning performance by leveraging the variability of the signal-to-noise ratio (SNR), without requiring auxiliary sensors. Analysis of 24 h observational datasets collected across diverse environments, including open-sky (OS), city streets (CS), and urban canyons (UC), demonstrates that multipath-affected non-line-of-sight (NLOS) signals exhibit significantly greater SNR variability than direct line-of-sight (LOS) signals. The HK model classifies received signals based on the standard deviation of their SNR and assigns corresponding weights during position estimation. Comparative performance evaluation indicates that relative to existing weighting models, the HK model improves 3D positioning accuracy by up to 22.4 m in urban canyon scenarios, reducing horizontal RMSE from 13.0 m to 4.7 m and vertical RMSE from 19.5 m to 6.9 m. In city street environments, horizontal RMSE is reduced from 11.6 m to 3.8 m. Furthermore, a time-sequential analysis at the TEHE site confirms consistent improvements in vertical positioning accuracy across all 24-hourly datasets, and in terms of horizontal accuracy, in 22 out of 24 cases. These results demonstrate that the HK model substantially surpasses conventional SNR- or elevation-based weighting techniques, particularly under severe multipath conditions frequently encountered in dense urban settings. Full article

(This article belongs to the Section Navigation and Positioning)

► Show Figures

Figure 1

24 pages, 74760 KB

Open AccessEditor’s ChoiceArticle

The Application of Mobile Devices for Measuring Accelerations in Rail Vehicles: Methodology and Field Research Outcomes in Tramway Transport

by Michał Urbaniak, Jakub Myrcik, Martyna Juda and Jan Mandrysz

Sensors 2025, 25(15), 4635; https://doi.org/10.3390/s25154635 - 26 Jul 2025

Cited by 1 | Viewed by 3669

Abstract

Unbalanced accelerations occurring during tram travel have a significant impact on passenger comfort and safety, as well as on the rate of wear and tear on infrastructure and rolling stock. Ideally, these dynamic forces should be monitored continuously in real-time; however, traditional systems [...] Read more.

Unbalanced accelerations occurring during tram travel have a significant impact on passenger comfort and safety, as well as on the rate of wear and tear on infrastructure and rolling stock. Ideally, these dynamic forces should be monitored continuously in real-time; however, traditional systems require high-precision accelerometers and proprietary software—investments often beyond the reach of municipally funded tram operators. To this end, as part of the research project “Accelerometer Measurements in Rail Passenger Transport Vehicles”, pilot measurement campaigns were conducted in Poland on tram lines in Gdańsk, Toruń, Bydgoszcz, and Olsztyn. Off-the-shelf smartphones equipped with MEMS accelerometers and GPS modules, running the Physics Toolbox Sensor Suite Pro app, were used. Although the research employs widely known methods, this paper addresses part of the gap in affordable real-time monitoring by demonstrating that, in the future, equipment equipped solely with consumer-grade MEMS accelerometers can deliver sufficiently accurate data in applications where high precision is not critical. This paper presents an analysis of a subset of results from the Gdańsk tram network. Lateral (x) and vertical (z) accelerations were recorded at three fixed points inside two tram models (Pesa 128NG Jazz Duo and Düwag N8C), while longitudinal accelerations were deliberately omitted at this stage due to their strong dependence on driver behavior. Raw data were exported as CSV files, processed and analyzed in R version 4.2.2, and then mapped spatially using ArcGIS cartograms. Vehicle speed was calculated both via the haversine formula—accounting for Earth’s curvature—and via a Cartesian approximation. Over the ~7 km route, both methods yielded virtually identical results, validating the simpler approach for short distances. Acceleration histograms approximated Gaussian distributions, with most values between 0.05 and 0.15 m/s², and extreme values approaching 1 m/s². The results demonstrate that low-cost mobile devices, after future calibration against certified accelerometers, can provide sufficiently rich data for ride-comfort assessment and show promise for cost-effective condition monitoring of both track and rolling stock. Future work will focus on optimizing the app’s data collection pipeline, refining standard-based analysis algorithms, and validating smartphone measurements against benchmark sensors. Full article

(This article belongs to the Collection Sensors and Actuators for Intelligent Vehicles)

► Show Figures

Figure 1

24 pages, 5256 KB

Open AccessEditor’s ChoiceArticle

In-Wheel Motor Fault Diagnosis Method Based on Two-Stream 2DCNNs with DCBA Module

by Junwei Zhu, Xupeng Ouyang, Zongkang Jiang, Yanlong Xu, Hongtao Xue, Huiyu Yue and Huayuan Feng

Sensors 2025, 25(15), 4617; https://doi.org/10.3390/s25154617 - 25 Jul 2025

Cited by 5 | Viewed by 1029

Abstract

To address the challenge of fault diagnosis for in-wheel motors in four-wheel independent driving systems under variable driving conditions and harsh environments, this paper proposes a novel method based on two-stream 2DCNNs (two-dimensional convolutional neural networks) with a DCBA (depthwise convolution block attention) [...] Read more.

To address the challenge of fault diagnosis for in-wheel motors in four-wheel independent driving systems under variable driving conditions and harsh environments, this paper proposes a novel method based on two-stream 2DCNNs (two-dimensional convolutional neural networks) with a DCBA (depthwise convolution block attention) module. The main contributions are twofold: (1) A DCBA module is introduced to extract multi-scale features—including prominent, local, and average information—from grayscale images reconstructed from vibration signals across different domains; and (2) a two-stream network architecture is designed to learn complementary feature representations from time-domain and time–frequency-domain signals, which are fused through fully connected layers to improve diagnostic accuracy. Experimental results demonstrate that the proposed method achieves high recognition accuracy under various working speeds, loads, and road surfaces. Comparative studies with SENet, ECANet, CBAM, and single-stream 2DCNN models confirm its superior performance and robustness. The integration of DCBA with dual-domain feature learning effectively enhances fault feature extraction under complex operating conditions. Full article

(This article belongs to the Special Issue Intelligent Maintenance and Fault Diagnosis of Mobility Equipment)

► Show Figures

Figure 1

17 pages, 13125 KB

Open AccessEditor’s ChoiceArticle

Evaluating the Accuracy and Repeatability of Mobile 3D Imaging Applications for Breast Phantom Reconstruction

by Elena Botti, Bart Jansen, Felipe Ballen-Moreno, Ayush Kapila and Redona Brahimetaj

Sensors 2025, 25(15), 4596; https://doi.org/10.3390/s25154596 - 24 Jul 2025

Viewed by 3027

Abstract

Three-dimensional imaging technologies are increasingly used in breast reconstructive and plastic surgery due to their potential for efficient and accurate preoperative assessment and planning. This study systematically evaluates the accuracy and consistency of six commercially available 3D scanning applications (apps)—Structure Sensor, 3D Scanner [...] Read more.

Three-dimensional imaging technologies are increasingly used in breast reconstructive and plastic surgery due to their potential for efficient and accurate preoperative assessment and planning. This study systematically evaluates the accuracy and consistency of six commercially available 3D scanning applications (apps)—Structure Sensor, 3D Scanner App, Heges, Polycam, SureScan, and Kiri—in reconstructing the female torso. To avoid variability introduced by human subjects, a silicone breast mannequin model was scanned, with fiducial markers placed at known anatomical landmarks. Manual distance measurements were obtained using calipers by two independent evaluators and compared to digital measurements extracted from 3D reconstructions in Blender software. Each scan was repeated six times per application to ensure reliability. SureScan demonstrated the lowest mean error (2.9 mm), followed by Structure Sensor (3.0 mm), Heges (3.6 mm), 3D Scanner App (4.4 mm), Kiri (5.0 mm), and Polycam (21.4 mm), which showed the highest error and variability. Even the app using an external depth sensor (Structure Sensor) showed no statistically significant accuracy advantage over those using only the iPad’s built-in camera (except for Polycam), underscoring that software is the primary driver of performance, not hardware (alone). This work provides practical insights for selecting mobile 3D scanning tools in clinical workflows and highlights key limitations, such as scaling errors and alignment artifacts. Future work should include patient-based validation and explore deep learning to enhance reconstruction quality. Ultimately, this study lays the foundation for more accessible and cost-effective 3D imaging in surgical practice, showing that smartphone-based tools can produce clinically useful scans. Full article

(This article belongs to the Special Issue Biomedical Imaging, Sensing and Signal Processing)

► Show Figures

Figure 1

30 pages, 4239 KB

Open AccessEditor’s ChoiceArticle

Real-Time Object Detection for Edge Computing-Based Agricultural Automation: A Case Study Comparing the YOLOX and YOLOv12 Architectures and Their Performance in Potato Harvesting Systems

by Joonam Kim, Giryeon Kim, Rena Yoshitoshi and Kenichi Tokuda

Sensors 2025, 25(15), 4586; https://doi.org/10.3390/s25154586 - 24 Jul 2025

Cited by 9 | Viewed by 2736

Abstract

In this paper, we presents a case study involving the implementation experience and a methodological framework through a comprehensive comparative analysis of the YOLOX and YOLOv12 object detection models for agricultural automation systems deployed in the Jetson AGX Orin edge computing platform. We [...] Read more.

In this paper, we presents a case study involving the implementation experience and a methodological framework through a comprehensive comparative analysis of the YOLOX and YOLOv12 object detection models for agricultural automation systems deployed in the Jetson AGX Orin edge computing platform. We examined the architectural differences between the models and their impact on detection capabilities in data-imbalanced potato-harvesting environments. Both models were trained on identical datasets with images capturing potatoes, soil clods, and stones, and their performances were evaluated through 30 independent trials under controlled conditions. Statistical analysis confirmed that YOLOX achieved a significantly higher throughput (107 vs. 45 FPS, p < 0.01) and superior energy efficiency (0.58 vs. 0.75 J/frame) than YOLOv12, meeting real-time processing requirements for agricultural automation. Although both models achieved an equivalent overall detection accuracy (F1-score, 0.97), YOLOv12 demonstrated specialized capabilities for challenging classes, achieving 42% higher recall for underrepresented soil clod objects (0.725 vs. 0.512, p < 0.01) and superior precision for small objects (0–3000 pixels). Architectural analysis identified a YOLOv12 residual efficient layer aggregation network backbone and area attention mechanism as key enablers of balanced precision–recall characteristics, which were particularly valuable for addressing agricultural data imbalance. However, NVIDIA Nsight profiling revealed implementation inefficiencies in the YOLOv12 multiprocess architecture, which prevented the theoretical advantages from being fully realized in edge computing environments. These findings provide empirically grounded guidelines for model selection in agricultural automation systems, highlighting the critical interplay between architectural design, implementation efficiency, and application-specific requirements. Full article

(This article belongs to the Section Smart Agriculture)

► Show Figures

Figure 1

19 pages, 3666 KB

Open AccessEditor’s ChoiceArticle

Rapid and Accurate Shape-Sensing Method Using a Multi-Core Fiber Bragg Grating-Based Optical Fiber

by Georgios Violakis, Nikolaos Vardakis, Zhenyu Zhang, Martin Angelmahr and Panagiotis Polygerinos

Sensors 2025, 25(14), 4494; https://doi.org/10.3390/s25144494 - 19 Jul 2025

Cited by 3 | Viewed by 3188

Abstract

Shape-sensing optical fibers have become increasingly important in applications requiring flexible navigation, spatial awareness, and deformation monitoring. Fiber Bragg Grating (FBG) sensors inscribed in multi-core optical fibers have been democratized over the years and nowadays offer a compact and robust platform for shape [...] Read more.

Shape-sensing optical fibers have become increasingly important in applications requiring flexible navigation, spatial awareness, and deformation monitoring. Fiber Bragg Grating (FBG) sensors inscribed in multi-core optical fibers have been democratized over the years and nowadays offer a compact and robust platform for shape reconstruction. In this work, we propose a novel, computationally efficient method for determining the 3D tip position of a bent multi-core FBG-based optical fiber using a second-order polynomial approximation of the fiber’s shape. The method begins with a calibration procedure, where polynomial coefficients are fitted for known bend configurations and subsequently modeled as a function of curvature using exponential decay functions. This allows for real-time estimation of the fiber tip position from curvature measurements alone, with no need for iterative numerical solutions or high processing power. The method was validated using miniaturized test structures and achieved sub-millimeter accuracy (<0.1 mm) over a 4.5 mm displacement range. Its simplicity and accuracy make it suitable for embedded or edge-computing applications in confined navigation, structural inspection, and medical robotics. Full article

(This article belongs to the Special Issue New Prospects in Fiber Optic Sensors and Applications)

► Show Figures

Graphical abstract

20 pages, 18517 KB

Open AccessEditor’s ChoiceArticle

A Highly Sensitive Low-Temperature N-Butanol Gas Sensor Based on a Co-Doped MOF-ZnO Nanomaterial Under UV Excitation

by Yinzhong Liu, Xiaoshun Wei, Yun Guo, Lingchao Wang, Hui Guo, Qingjie Wang, Yiyu Qiao, Xiaotao Zhu, Xuechun Yang, Lingli Cheng and Zheng Jiao

Sensors 2025, 25(14), 4480; https://doi.org/10.3390/s25144480 - 18 Jul 2025

Cited by 3 | Viewed by 1222

Abstract

Volatile organic compounds (VOCs) are presently posing a rather considerable threat to both human health and environmental sustainability. Among these, n-butanol is commonly identified as bringing potential hazards to environmental integrity and individual health. This study presents the creation of a highly sensitive [...] Read more.

Volatile organic compounds (VOCs) are presently posing a rather considerable threat to both human health and environmental sustainability. Among these, n-butanol is commonly identified as bringing potential hazards to environmental integrity and individual health. This study presents the creation of a highly sensitive n-butanol gas sensor utilizing cobalt-doped zinc oxide (ZnO) derived from a metal–organic framework (MOF). A series of x-Co/MOF-ZnO (x = 1, 3, 5, 7 wt%) nanomaterials with varying Co ratios were generated using the homogeneous co-precipitation method and assessed for their gas-sensing performances under a low operating temperature (191 °C) and UV excitation (220 mW/cm²). These findings demonstrated that the 5-Co/MOF-ZnO sensor presented the highest oxygen vacancy (O_v) concentration and the largest specific surface area (SSA), representing the optimal reactivity, selectivity, and durability for n-butanol detection. Regarding the sensor’s response to 100 ppm n-butanol under UV excitation, it achieved a value of 1259.06, 9.80 times greater than that of pure MOF-ZnO (128.56) and 2.07 times higher than that in darkness (608.38). Additionally, under UV illumination, the sensor achieved a rapid response time (11 s) and recovery rate (23 s). As a strategy to transform the functionality of ZnO-based sensors for n-butanol gas detection, this study also investigated potential possible redox reactions occurring during the detection process. Full article

(This article belongs to the Special Issue New Sensors Based on Inorganic Material)

► Show Figures

Figure 1

25 pages, 6123 KB

Open AccessEditor’s ChoiceArticle

SDA-YOLO: An Object Detection Method for Peach Fruits in Complex Orchard Environments

by Xudong Lin, Dehao Liao, Zhiguo Du, Bin Wen, Zhihui Wu and Xianzhi Tu

Sensors 2025, 25(14), 4457; https://doi.org/10.3390/s25144457 - 17 Jul 2025

Cited by 5 | Viewed by 1713

Abstract

To address the challenges of leaf–branch occlusion, fruit mutual occlusion, complex background interference, and scale variations in peach detection within complex orchard environments, this study proposes an improved YOLOv11n-based peach detection method named SDA-YOLO. First, in the backbone network, the LSKA module is [...] Read more.

To address the challenges of leaf–branch occlusion, fruit mutual occlusion, complex background interference, and scale variations in peach detection within complex orchard environments, this study proposes an improved YOLOv11n-based peach detection method named SDA-YOLO. First, in the backbone network, the LSKA module is embedded into the SPPF module to construct an SPPF-LSKA fusion module, enhancing multi-scale feature representation for peach targets. Second, an MPDIoU-based bounding box regression loss function replaces CIoU to improve localization accuracy for overlapping and occluded peaches. The DyHead Block is integrated into the detection head to form a DMDetect module, strengthening feature discrimination for small and occluded targets in complex backgrounds. To address insufficient feature fusion flexibility caused by scale variations from occlusion and illumination differences in multi-scale peach detection, a novel Adaptive Multi-Scale Fusion Pyramid (AMFP) module is proposed to enhance the neck network, improving flexibility in processing complex features. Experimental results demonstrate that SDA-YOLO achieves precision (P), recall (R), mAP@0.95, and mAP@0.5:0.95 of 90.8%, 85.4%, 90%, and 62.7%, respectively, surpassing YOLOv11n by 2.7%, 4.8%, 2.7%, and 7.2%. This verifies the method’s robustness in complex orchard environments and provides effective technical support for intelligent fruit harvesting and yield estimation. Full article

(This article belongs to the Special Issue Sensing Technology and Computer Vision for Precision Agriculture and Smart Farming)

► Show Figures

Figure 1

23 pages, 6991 KB

Open AccessEditor’s ChoiceArticle

Comparing the Accuracy of Soil Moisture Estimates Derived from Bulk and Energy-Resolved Gamma Radiation Measurements

by Sonia Akter, Johan Alexander Huisman and Heye Reemt Bogena

Sensors 2025, 25(14), 4453; https://doi.org/10.3390/s25144453 - 17 Jul 2025

Viewed by 1662

Abstract

Monitoring soil moisture (SM) using permanently installed gamma radiation (GR) detectors is a promising non-invasive method based on the inverse relationship between SM and soil-emitted GR. In a previous study, we successfully estimated SM from environmental gamma radiation (EGR) measured by a low-cost [...] Read more.

Monitoring soil moisture (SM) using permanently installed gamma radiation (GR) detectors is a promising non-invasive method based on the inverse relationship between SM and soil-emitted GR. In a previous study, we successfully estimated SM from environmental gamma radiation (EGR) measured by a low-cost counter-tube detector. Since this detector type provides a bulk GR response across a wide energy range, EGR signals are influenced by several confounding factors, e.g., soil radon emanation, biomass. To what extent these confounding factors deteriorate the accuracy of SM estimates obtained from EGR is not fully understood. Therefore, the aim of this study was to compare the accuracy of SM estimates from EGR with those from reference ⁴⁰K GR (1460 keV) measurements which are much less influenced by these factors. For this, a Geiger–Mueller counter (G–M), which is commonly used for EGR monitoring, and a gamma spectrometer were installed side by side in an agricultural field equipped with in situ sensors to measure reference SM and a meteorological station. The EGR_G–M and spectrometry-based ⁴⁰K measurements were related to reference SM using a functional relationship derived from theory. We found that daily SM can be predicted with an RMSE of 3.39 vol. % from ⁴⁰K using the theoretical value of α = 1.11 obtained from the effective ratio of GR mass attenuation coefficients for the water and solid phase. A lower accuracy was achieved for the EGR_G–M measurements (RMSE = 6.90 vol. %). Wavelet coherence analysis revealed that the EGR_G–M measurements were influenced by radon-induced noise in winter. Additionally, biomass shielding had a stronger impact on EGR_G–M than on ⁴⁰K GR estimates of SM during summer. In summary, our study provides a better understanding on the lower prediction accuracy of EGR_G–M and suggests that correcting for biomass can improve SM estimation from the bulk EGR data of operational radioactivity monitoring networks. Full article

(This article belongs to the Special Issue Sensors in Smart Irrigation Systems)

► Show Figures

Figure 1

19 pages, 5627 KB

Open AccessEditor’s ChoiceArticle

Reliability Modeling of Wind Turbine Gearbox System Considering Failure Correlation Under Shock–Degradation

by Xiaojun Liu, Ziwen Wu, Yiping Yuan, Wenlei Sun and Jianxiong Gao

Sensors 2025, 25(14), 4425; https://doi.org/10.3390/s25144425 - 16 Jul 2025

Cited by 4 | Viewed by 1418

Abstract

To address traditional methods’ limitations in neglecting the interaction between random shock loads and progressive degradation, as well as failure correlations, this study proposes a dynamic reliability framework integrating Gamma processes, homogeneous Poisson processes (HPP), and mixed Copula functions. The framework develops a [...] Read more.

To address traditional methods’ limitations in neglecting the interaction between random shock loads and progressive degradation, as well as failure correlations, this study proposes a dynamic reliability framework integrating Gamma processes, homogeneous Poisson processes (HPP), and mixed Copula functions. The framework develops a wind turbine gearbox reliability model under shock–degradation coupling while quantifying failure correlations. Gamma processes characterize continuous degradation, with parameters estimated from P-S-N curves. Based on stress–strength interference theory, random shocks within damage thresholds are integrated to form a coupled reliability model. A Gumbel–Clayton–Frank mixed Copula with a multi-layer nested algorithm quantifies failure correlations, with correlation parameters estimated via the RSS principle and genetic algorithms. Validation using a 2 MW gearbox’s planetary gear-stage system covers four scenarios: natural degradation, shock–degradation coupling, and both scenarios with failure correlations. The results show that compared to independent assumptions, the model accelerates reliability decline, increasing failure rates by >37%. Relative to natural degradation-only models, failure rates rise by >60%, validating the model’s effectiveness and alignment with real-world operational conditions. Full article

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

► Show Figures

Figure 1

25 pages, 315 KB

Open AccessEditor’s ChoiceReview

Motion Capture Technologies for Athletic Performance Enhancement and Injury Risk Assessment: A Review for Multi-Sport Organizations

by Bahman Adlou, Christopher Wilburn and Wendi Weimar

Sensors 2025, 25(14), 4384; https://doi.org/10.3390/s25144384 - 13 Jul 2025

Cited by 4 | Viewed by 6981

Abstract

Background: Motion capture (MoCap) technologies have transformed athlete monitoring, yet athletic departments face complex decisions when selecting systems for multiple sports. Methods: We conducted a narrative review of peer-reviewed studies (2015–2025) examining optical marker-based, inertial measurement unit (IMU) systems, including Global Navigation Satellite [...] Read more.

Background: Motion capture (MoCap) technologies have transformed athlete monitoring, yet athletic departments face complex decisions when selecting systems for multiple sports. Methods: We conducted a narrative review of peer-reviewed studies (2015–2025) examining optical marker-based, inertial measurement unit (IMU) systems, including Global Navigation Satellite System (GNSS)-integrated systems, and markerless computer vision systems. Studies were evaluated for validated accuracy metrics across indoor court, aquatic, and outdoor field environments. Results: Optical systems maintain sub-millimeter accuracy in controlled environments but face field limitations. IMU systems demonstrate an angular accuracy of 2–8° depending on movement complexity. Markerless systems show variable accuracy (sagittal: 3–15°, transverse: 3–57°). Environmental factors substantially impact system performance, with aquatic settings introducing an additional orientation error of 2° versus terrestrial applications. Outdoor environments challenge GNSS-based tracking (±0.3–3 m positional accuracy). Critical gaps include limited gender-specific validation and insufficient long-term reliability data. Conclusions: This review proposes a tiered implementation framework combining foundation-level team monitoring with specialized assessment tools. This evidence-based approach guides the selection of technology aligned with organizational priorities, sport-specific requirements, and resource constraints. Full article

(This article belongs to the Special Issue Sensors Technology for Sports Biomechanics Applications)

15 pages, 4804 KB

Open AccessEditor’s ChoiceArticle

Improving Cell Detection and Tracking in Microscopy Images Using YOLO and an Enhanced DeepSORT Algorithm

by Mokhaled N. A. Al-Hamadani, Richard Poroszlay, Gabor Szeman-Nagy, Andras Hajdu, Stathis Hadjidemetriou, Luca Ferrarini and Balazs Harangi

Sensors 2025, 25(14), 4361; https://doi.org/10.3390/s25144361 - 12 Jul 2025

Cited by 3 | Viewed by 2923

Abstract

Accurate and automated detection and tracking of cells in microscopy images is a persistent challenge in biotechnology and biomedical research. Effective detection and tracking are crucial for understanding biological processes and extracting meaningful data for subsequent simulations. In this study, we present an [...] Read more.

Accurate and automated detection and tracking of cells in microscopy images is a persistent challenge in biotechnology and biomedical research. Effective detection and tracking are crucial for understanding biological processes and extracting meaningful data for subsequent simulations. In this study, we present an integrated pipeline that leverages a fine-tuned YOLOv8x model for detecting cells and cell divisions across microscopy image series. While YOLOv8x exhibits strong detection capabilities, it occasionally misses certain cells, leading to gaps in data. To mitigate this, we incorporate the DeepSORT tracking algorithm, which enhances data association and reduces the cells’ identity (ID) switches by utilizing a pre-trained convolutional network for robust multi-object tracking. This combination ensures continuous detection and compensates for missed detections, thereby improving overall recall. Our approach achieves a recall of 93.21% with the enhanced DeepSORT algorithm, compared to the 53.47% recall obtained by the original YOLOv8x model. The proposed pipeline effectively extracts detailed information from structured image datasets, providing a reliable approximation of cellular processes in culture environments. Full article

(This article belongs to the Section Intelligent Sensors)

► Show Figures

Graphical abstract

39 pages, 1305 KB

Open AccessEditor’s ChoiceReview

AI Trustworthiness in Manufacturing: Challenges, Toolkits, and the Path to Industry 5.0

by M. Nadeem Ahangar, Z. A. Farhat and Aparajithan Sivanathan

Sensors 2025, 25(14), 4357; https://doi.org/10.3390/s25144357 - 11 Jul 2025

Cited by 7 | Viewed by 7749

Abstract

The integration of Artificial Intelligence (AI) into manufacturing is transforming the industry by advancing predictive maintenance, quality control, and supply chain optimisation, while also driving the shift from Industry 4.0 towards a more human-centric and sustainable vision. This emerging paradigm, known as Industry [...] Read more.

The integration of Artificial Intelligence (AI) into manufacturing is transforming the industry by advancing predictive maintenance, quality control, and supply chain optimisation, while also driving the shift from Industry 4.0 towards a more human-centric and sustainable vision. This emerging paradigm, known as Industry 5.0, emphasises resilience, ethical innovation, and the symbiosis between humans and intelligent systems, with AI playing a central enabling role. However, challenges such as the “black box” nature of AI models, data biases, ethical concerns, and the lack of robust frameworks for trustworthiness hinder its widespread adoption. This paper provides a comprehensive survey of AI trustworthiness in the manufacturing industry, examining the evolution of industrial paradigms, identifying key barriers to AI adoption, and examining principles such as transparency, fairness, robustness, and accountability. It offers a detailed summary of existing toolkits and methodologies for explainability, bias mitigation, and robustness, which are essential for fostering trust in AI systems. Additionally, this paper examines challenges throughout the AI pipeline, from data collection to model deployment, and concludes with recommendations and research questions aimed at addressing these issues. By offering actionable insights, this study aims to guide researchers, practitioners, and policymakers in developing ethical and reliable AI systems that align with the principles of Industry 5.0, ensuring both technological advancement and societal value. Full article

(This article belongs to the Section Industrial Sensors)

► Show Figures

Figure 1

25 pages, 9813 KB

Open AccessEditor’s ChoiceArticle

Digital Twin Approach for Fault Diagnosis in Photovoltaic Plant DC–DC Converters

by Pablo José Hueros-Barrios, Francisco Javier Rodríguez Sánchez, Pedro Martín Sánchez, Carlos Santos-Pérez, Ariya Sangwongwanich, Mateja Novak and Frede Blaabjerg

Sensors 2025, 25(14), 4323; https://doi.org/10.3390/s25144323 - 10 Jul 2025

Cited by 6 | Viewed by 3654

Abstract

This article presents a hybrid fault diagnosis framework for DC–DC converters in photovoltaic (PV) systems, combining digital twin (DT) modelling and detection with machine learning anomaly classification. The proposed method addresses both hardware faults such as open and short circuits in insulated-gate bipolar [...] Read more.

This article presents a hybrid fault diagnosis framework for DC–DC converters in photovoltaic (PV) systems, combining digital twin (DT) modelling and detection with machine learning anomaly classification. The proposed method addresses both hardware faults such as open and short circuits in insulated-gate bipolar transistors (IGBTs) and diodes and sensor-level false data injection attacks (FDIAs). A five-dimensional DT architecture is employed, where a virtual entity implemented using FMI-compliant FMUs interacts with a real-time emulated physical plant. Fault detection is performed by comparing the real-time system behaviour with DT predictions, using dynamic thresholds based on power, voltage, and current sensors errors. Once a discrepancy is flagged, a second step classifier processes normalized time-series windows to identify the specific fault type. Synthetic training data are generated using emulation models under normal and faulty conditions, and feature vectors are constructed using a compact, interpretable set of statistical and spectral descriptors. The model was validated using OPAL-RT Hardware in the Loop emulations. The results show high classification accuracy, robustness to environmental fluctuations, and transferability across system configurations. The framework also demonstrates compatibility with low-cost deployment hardware, confirming its practical applicability for fault diagnosis in real-world PV systems. Full article

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

► Show Figures

Figure 1

21 pages, 10356 KB

Open AccessEditor’s ChoiceArticle

Autonomous Greenhouse Cultivation of Dwarf Tomato: Performance Evaluation of Intelligent Algorithms for Multiple-Sensor Feedback

by Stef C. Maree, Pinglin Zhang, Bart M. van Marrewijk, Feije de Zwart, Monique Bijlaard and Silke Hemming

Sensors 2025, 25(14), 4321; https://doi.org/10.3390/s25144321 - 10 Jul 2025

Cited by 1 | Viewed by 4510

Abstract

Greenhouse horticulture plays an important role globally by producing nutritious fruits and vegetables with high resource use efficiency. Modern greenhouses are large-scale high-tech production factories that are increasingly data-driven, and where climate and irrigation control are gradually becoming more autonomous. This is enabled [...] Read more.

Greenhouse horticulture plays an important role globally by producing nutritious fruits and vegetables with high resource use efficiency. Modern greenhouses are large-scale high-tech production factories that are increasingly data-driven, and where climate and irrigation control are gradually becoming more autonomous. This is enabled by technological developments and driven by shortages in skilled labor and the demand for improved resource use efficiency. In the Autonomous Greenhouse Challenge, it has been shown that controlling greenhouse cultivation can be done efficiently with intelligent algorithms. For an optimal strategy, however, it is essential that control algorithms properly account for crop responses, which requires appropriate sensors, reliable data, and accurate models. This paper presents the results of the 4th Autonomous Greenhouse Challenge, in which international teams developed six intelligent algorithms that fully controlled a dwarf tomato cultivation, a crop that is well-suited for robotic harvesting, but for which little prior cultivation data exists. Nevertheless, the analysis of the experiment showed that all teams managed to obtain a profitable strategy, and the best algorithm resulted a production equivalent to 45 kg/m²/year, higher than in the commercial practice of high-wire cherry tomato growing. The predominant factor was found to be the much higher plant density that can be achieved in the applied growing system. More difficult challenges were found to be related to measuring crop status to determine the harvest moment. Finally, this experiment shows the potential for novel greenhouse cultivation systems that are inherently well-suited for autonomous control, and results in a unique and rich dataset to support future research. Full article

(This article belongs to the Special Issue AI, IoT and Smart Sensors for Precision Agriculture: 2nd Edition)

► Show Figures

Figure 1

15 pages, 3836 KB

Open AccessEditor’s ChoiceArticle

Porous-Cladding Polydimethylsiloxane Optical Waveguide for Biomedical Pressure Sensing Applications

by Koffi Novignon Amouzou, Alberto Alonso Romero, Dipankar Sengupta, Camila Aparecida Zimmermann, Aashutosh Kumar, Normand Gravel, Jean-Marc Lina, Xavier Daxhelet and Bora Ung

Sensors 2025, 25(14), 4311; https://doi.org/10.3390/s25144311 - 10 Jul 2025

Cited by 1 | Viewed by 1836

Abstract

We report a new concept of a pressure sensor fully made from polydimethylsiloxane with a solid core and porous cladding that operates through (frustrated) total internal reflection. A flexible and sensitive rectangular cross-section waveguide was fabricated via the casting and molding method. The [...] Read more.

We report a new concept of a pressure sensor fully made from polydimethylsiloxane with a solid core and porous cladding that operates through (frustrated) total internal reflection. A flexible and sensitive rectangular cross-section waveguide was fabricated via the casting and molding method. The waveguide’s optical losses can be temperature-controlled during the fabrication process by controlling the quantity of microbubbles incorporated (2% approximately for samples made at 70 °C). By controlling the precuring temperature, the microbubbles are incorporated into the waveguides during the simple and cost-effective fabrication process through the casting and molding method. For these samples, we measured good optical loss tradeoff of the order of 1.85 dB/cm, which means that it is possible to fabricate a solid-core/clad waveguide with porous cladding able to guide light properly. We demonstrated the microbubble concentration control in the waveguide, and we measured an average diameter of 239 ± 16 µm. A sensitivity to pressure of 0.1035 dB/kPa optical power loss was measured. The results show that in a biomedical dynamic pressure range (0 to 13.3 kPa), this new device indicates the critical pressure threshold level, which constitutes a crucial asset for potential applications such as pressure injury prevention. Full article

(This article belongs to the Special Issue Optical-Based Sensors and Sensing Technologies for Biomedical Applications)

► Show Figures

Graphical abstract

16 pages, 3101 KB

Open AccessEditor’s ChoiceArticle

Enhanced High-Resolution and Long-Range FMCW LiDAR with Directly Modulated Semiconductor Lasers

by Luís C. P. Pinto and Maria C. R. Medeiros

Sensors 2025, 25(13), 4131; https://doi.org/10.3390/s25134131 - 2 Jul 2025

Cited by 2 | Viewed by 3586

Abstract

Light detection and ranging (LiDAR) sensors are essential for applications where high-resolution distance and velocity measurements are required. In particular, frequency-modulated continuous wave (FMCW) LiDAR, compared with other LiDAR implementations, provides superior receiver sensitivity, enhanced range resolution, and the capability to measure velocity. [...] Read more.

Light detection and ranging (LiDAR) sensors are essential for applications where high-resolution distance and velocity measurements are required. In particular, frequency-modulated continuous wave (FMCW) LiDAR, compared with other LiDAR implementations, provides superior receiver sensitivity, enhanced range resolution, and the capability to measure velocity. Integrating LiDARs into electronic and photonic semiconductor chips can lower their cost, size, and power consumption, making them affordable for cost-sensitive applications. Additionally, simple designs are required, such as FMCW signal generation by the direct modulation of the current of a semiconductor laser. However, semiconductor lasers are inherently nonlinear, and the driving waveform needs to be optimized to generate linear FMCW signals. In this paper, we employ pre-distortion techniques to compensate for chirp nonlinearity, achieving frequency nonlinearities of 0.0029% for the down-ramp and the up-ramp at 55 kHz. Experimental results demonstrate a highly accurate LiDAR system with a resolution of under 5 cm, operating over a 210-m range through single-mode fiber, which corresponds to approximately 308 m in free space, towards meeting the requirements for long-range autonomous driving. Full article

(This article belongs to the Special Issue Feature Papers in Optical Sensors 2025)

► Show Figures

Figure 1

18 pages, 2887 KB

Open AccessEditor’s ChoiceArticle

Polymer-Based Chemicapacitive Hybrid Sensor Array for Improved Selectivity in e-Nose Systems

by Pavithra Munirathinam, Mohd Farhan Arshi, Haleh Nazemi, Gian Carlo Antony Raj and Arezoo Emadi

Sensors 2025, 25(13), 4130; https://doi.org/10.3390/s25134130 - 2 Jul 2025

Viewed by 3515

Abstract

Detecting volatile organic compounds (VOCs) is essential for health, environmental protection, and industrial safety. VOCs contribute to air pollution, pose health risks, and can indicate leaks or contamination in industries. Applications include air quality monitoring, disease diagnosis, and food safety. This paper focuses [...] Read more.

Detecting volatile organic compounds (VOCs) is essential for health, environmental protection, and industrial safety. VOCs contribute to air pollution, pose health risks, and can indicate leaks or contamination in industries. Applications include air quality monitoring, disease diagnosis, and food safety. This paper focuses on polymer-based hybrid sensor arrays (HSAs) utilizing interdigitated electrode (IDE) geometries for VOC detection. Achieving high selectivity and sensitivity in gas sensing remains a challenge, particularly in complex environments. To address this, we propose HSAs as an innovative solution to enhance sensor performance. IDE-based sensors are designed and fabricated using the Polysilicon Multi-User MEMS process (PolyMUMPs). Experimental evaluations are performed by exposing sensors to VOCs under controlled conditions. Traditional multi-sensor arrays (MSAs) achieve 82% prediction accuracy, while virtual sensor arrays (VSAs) leveraging frequency dependence improve performance: PMMA-VSA and PVP-VSA predict compounds with 100% and 98% accuracy, respectively. The proposed HSA, integrating these VSAs, consistently achieves 100% accuracy in compound identification and concentration estimation, surpassing MSA and VSA performance. These findings demonstrate that proposed polymer-based HSAs and VSAs, particularly with advanced IDE geometries, significantly enhance selectivity and sensitivity, advancing e-Nose technology for more accurate and reliable VOC detection across diverse applications. Full article

(This article belongs to the Special Issue Advanced Sensors for Gas Monitoring)

► Show Figures

Figure 1

27 pages, 2935 KB

Open AccessEditor’s ChoiceArticle

A Pilot Study on Emotional Equivalence Between VR and Real Spaces Using EEG and Heart Rate Variability

by Takato Kobayashi, Narumon Jadram, Shukuka Ninomiya, Kazuhiro Suzuki and Midori Sugaya

Sensors 2025, 25(13), 4097; https://doi.org/10.3390/s25134097 - 30 Jun 2025

Cited by 1 | Viewed by 3784

Abstract

In recent years, the application of virtual reality (VR) for spatial evaluation has gained traction in the fields of architecture and interior design. However, for VR to serve as a viable substitute for real-world environments, it is essential that experiences within VR elicit [...] Read more.

In recent years, the application of virtual reality (VR) for spatial evaluation has gained traction in the fields of architecture and interior design. However, for VR to serve as a viable substitute for real-world environments, it is essential that experiences within VR elicit emotional responses comparable to those evoked by actual spaces. Despite this prerequisite, there remains a paucity of studies that objectively compare and evaluate the emotional responses elicited by VR and real-world environments. Consequently, it is not yet fully understood whether VR can reliably replicate the emotional experiences induced by physical spaces. This study aims to investigate the influence of presentation modality on emotional responses by comparing a VR space and a real-world space with identical designs. The comparison was conducted using both subjective evaluations (Semantic Differential method) and physiological indices (electroencephalography and heart rate variability). The results indicated that the real-world environment was associated with impressions of comfort and preference, whereas the VR environment evoked impressions characterized by heightened arousal. Additionally, elevated beta wave activity and increased beta/alpha ratios were observed in the VR condition, suggesting a state of high arousal, as further supported by positioning on the Emotion Map. Moreover, analysis of pNN50 revealed a transient increase in parasympathetic nervous activity during the VR experience. This study is positioned as a pilot investigation to explore physiological and emotional differences between VR and real spaces. Full article

(This article belongs to the Special Issue Neuroarchitecture: Innovations in the Human-Environment Relationship Through Sensors and Sensing Technologies)

► Show Figures

Figure 1

18 pages, 7709 KB

Open AccessEditor’s ChoiceArticle

Orientation Controllable RCS Enhancement Electromagnetic Surface to Improve the Road Barriers Detectability for Autonomous Driving Radar

by Yanbin Chen, Tong Wang, Qi Liu, Haochen Wang and Cheng Jin

Sensors 2025, 25(13), 4048; https://doi.org/10.3390/s25134048 - 29 Jun 2025

Cited by 1 | Viewed by 976

Abstract

An orientation controllable radar cross section (RCS) enhancement surface is presented in this paper, which can be used to improve the road pile detectability of on-board microwave radar for autonomous driving system. In addition, the RCS enhancement orientation can be controlled in a [...] Read more.

An orientation controllable radar cross section (RCS) enhancement surface is presented in this paper, which can be used to improve the road pile detectability of on-board microwave radar for autonomous driving system. In addition, the RCS enhancement orientation can be controlled in a specified direction without interfering with other microwave systems. We first designed a modified one-dimensional VanAtta array with adjustable phase for retrodirective backtracking the incoming electromagnetic waves, which can achieve wide-angle RCS enhancement. Then, we arranged the one-dimensional VanAtta array in another dimension forming a two-dimensional array, enabling adjustable orientation RCS enhancement due to the controllable phase of the reflected electromagnetic waves. We designed, manufactured, and tested a 4 × 8 array to validate the theory and assess the design’s feasibility. Finally, six orientation controllable VanAtta arrays were mounted on the outside surface of a cylinder road barrier, and measurements demonstrated that RCS enhancement of over 10 dB have been achieved compared to the same pile with perfect electric conductor surface. Full article

(This article belongs to the Section Radar Sensors)

► Show Figures

Figure 1

28 pages, 1634 KB

Open AccessEditor’s ChoiceReview

AI-Powered Vocalization Analysis in Poultry: Systematic Review of Health, Behavior, and Welfare Monitoring

by Venkatraman Manikandan and Suresh Neethirajan

Sensors 2025, 25(13), 4058; https://doi.org/10.3390/s25134058 - 29 Jun 2025

Cited by 3 | Viewed by 4946

Abstract

Artificial intelligence and bioacoustics represent a paradigm shift in non-invasive poultry welfare monitoring through advanced vocalization analysis. This comprehensive systematic review critically examines the transformative evolution from traditional acoustic feature extraction—including Mel-Frequency Cepstral Coefficients (MFCCs), spectral entropy, and spectrograms—to cutting-edge deep learning architectures [...] Read more.

Artificial intelligence and bioacoustics represent a paradigm shift in non-invasive poultry welfare monitoring through advanced vocalization analysis. This comprehensive systematic review critically examines the transformative evolution from traditional acoustic feature extraction—including Mel-Frequency Cepstral Coefficients (MFCCs), spectral entropy, and spectrograms—to cutting-edge deep learning architectures encompassing Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, attention mechanisms, and groundbreaking self-supervised models such as wav2vec2 and Whisper. The investigation reveals compelling evidence for edge computing deployment via TinyML frameworks, addressing critical scalability challenges in commercial poultry environments characterized by acoustic complexity and computational constraints. Advanced applications spanning emotion recognition, disease detection, and behavioral phenotyping demonstrate unprecedented potential for real-time welfare assessment. Through rigorous bibliometric co-occurrence mapping and thematic clustering analysis, this review exposes persistent methodological bottlenecks: dataset standardization deficits, evaluation protocol inconsistencies, and algorithmic interpretability limitations. Critical knowledge gaps emerge in cross-species domain generalization and contextual acoustic adaptation, demanding urgent research prioritization. The findings underscore explainable AI integration as essential for establishing stakeholder trust and regulatory compliance in automated welfare monitoring systems. This synthesis positions acoustic AI as a cornerstone technology enabling ethical, transparent, and scientifically robust precision livestock farming, bridging computational innovation with biological relevance for sustainable poultry production systems. Future research directions emphasize multi-modal sensor integration, standardized evaluation frameworks, and domain-adaptive models capable of generalizing across diverse poultry breeds, housing conditions, and environmental contexts while maintaining interpretability for practical farm deployment. Full article

(This article belongs to the Special Issue Feature Papers in Smart Agriculture 2025)

► Show Figures

Figure 1

27 pages, 10314 KB

Open AccessEditor’s ChoiceArticle

Immersive Teleoperation via Collaborative Device-Agnostic Interfaces for Smart Haptics: A Study on Operational Efficiency and Cognitive Overflow for Industrial Assistive Applications

by Fernando Hernandez-Gobertti, Ivan D. Kudyk, Raul Lozano, Giang T. Nguyen and David Gomez-Barquero

Sensors 2025, 25(13), 3993; https://doi.org/10.3390/s25133993 - 26 Jun 2025

Cited by 1 | Viewed by 4033

Abstract

This study presents a novel investigation into immersive teleoperation systems using collaborative, device-agnostic interfaces for advancing smart haptics in industrial assistive applications. The research focuses on evaluating the quality of experience (QoE) of users interacting with a teleoperation system comprising a local robotic [...] Read more.

This study presents a novel investigation into immersive teleoperation systems using collaborative, device-agnostic interfaces for advancing smart haptics in industrial assistive applications. The research focuses on evaluating the quality of experience (QoE) of users interacting with a teleoperation system comprising a local robotic arm, a robot gripper, and heterogeneous remote tracking and haptic feedback devices. By employing a modular device-agnostic framework, the system supports flexible configurations, including one-user-one-equipment (1U-1E), one-user-multiple-equipment (1U-ME), and multiple-users-multiple-equipment (MU-ME) scenarios. The experimental set-up involves participants manipulating predefined objects and placing them into designated baskets by following specified 3D trajectories. Performance is measured using objective QoE metrics, including temporal efficiency (time required to complete the task) and spatial accuracy (trajectory similarity to the predefined path). In addition, subjective QoE metrics are assessed through detailed surveys, capturing user perceptions of presence, engagement, control, sensory integration, and cognitive load. To ensure flexibility and scalability, the system integrates various haptic configurations, including (1) a Touch kinaesthetic device for precision tracking and grounded haptic feedback, (2) a DualSense tactile joystick as both a tracker and mobile haptic device, (3) a bHaptics DK2 vibrotactile glove with a camera tracker, and (4) a SenseGlove Nova force-feedback glove with VIVE trackers. The modular approach enables comparative analysis of how different device configurations influence user performance and experience. The results indicate that the objective QoE metrics varied significantly across device configurations, with the Touch and SenseGlove Nova set-ups providing the highest trajectory similarity and temporal efficiency. Subjective assessments revealed a strong correlation between presence and sensory integration, with users reporting higher engagement and control in scenarios utilizing force feedback mechanisms. Cognitive load varied across the set-ups, with more complex configurations (e.g., 1U-ME) requiring longer adaptation periods. This study contributes to the field by demonstrating the feasibility of a device-agnostic teleoperation framework for immersive industrial applications. It underscores the critical interplay between objective task performance and subjective user experience, providing actionable insights into the design of next-generation teleoperation systems. Full article

(This article belongs to the Special Issue Recent Development of Flexible Tactile Sensors and Their Applications)

► Show Figures

Figure 1

18 pages, 2462 KB

Open AccessEditor’s ChoiceArticle

Autonomous Drilling and the Idea of Next-Generation Deep Mineral Exploration

by George Nikolakopoulos, Anton Koval, Matteo Fumagalli, Martyna Konieczna-Fuławka, Laura Santas Moreu, Victor Vigara-Puche, Kashish Verma, Bob de Waard and René Deutsch

Sensors 2025, 25(13), 3953; https://doi.org/10.3390/s25133953 - 25 Jun 2025

Cited by 1 | Viewed by 2990

Abstract

Remote drilling technologies play a crucial role in automating both underground and open-pit hard rock mining operations. These technologies enhance efficiency and, most importantly, improve safety in the mining sector. Autonomous drilling rigs can navigate to pre-determined positions and utilize the appropriate parameters [...] Read more.

Remote drilling technologies play a crucial role in automating both underground and open-pit hard rock mining operations. These technologies enhance efficiency and, most importantly, improve safety in the mining sector. Autonomous drilling rigs can navigate to pre-determined positions and utilize the appropriate parameters to drill boreholes effectively. This article explores various aspects of automation, including the integration of advanced data collection methods that monitor the drilling parameters and facilitate the creation of 3D models of rock hardness. The shift toward machine automation involves transitioning from human-operated machines to systems powered by artificial intelligence, which are capable of making real-time decisions. Navigating underground environments presents unique challenges, as traditional RF-based localization systems often fail in these settings. New solutions, such as constant localization and mapping techniques like SLAM (simultaneous localization and mapping), provide innovative methods for navigating mines, particularly in uncharted territories. The development of robotic exploration rigs equipped with modules that can operate autonomously in hazardous areas has the potential to revolutionize mineral exploration in underground mines. This article also discusses solutions aimed at validating and improving existing methods by optimizing drilling strategies to ensure accuracy, enhance efficiency, and ensure safety. These topics are explored in the context of the Horizon Europe-funded PERSEPHONE project, which seeks to deliver fully autonomous, sensor-integrated robotic systems for deep mineral exploration in challenging underground environments. Full article

(This article belongs to the Section Sensors and Robotics)

► Show Figures

Figure 1

16 pages, 2985 KB

Open AccessFeature PaperEditor’s ChoiceArticle

Fault Identification Model Using Convolutional Neural Networks with Transformer Architecture

by Yongxin Fan, Yiming Dang and Yangming Guo

Sensors 2025, 25(13), 3897; https://doi.org/10.3390/s25133897 - 23 Jun 2025

Cited by 3 | Viewed by 1973

Abstract

With the advancement of industrial manufacturing and the shift toward high automation, machines have increasingly taken over many production tasks, greatly improving efficiency and reducing human labor. However, this also introduces new challenges, particularly the inability of machines to autonomously detect and diagnose [...] Read more.

With the advancement of industrial manufacturing and the shift toward high automation, machines have increasingly taken over many production tasks, greatly improving efficiency and reducing human labor. However, this also introduces new challenges, particularly the inability of machines to autonomously detect and diagnose faults. Such undetected issues may cause unexpected breakdowns, interrupting critical operations, leading to economic losses and potential safety hazards. To address this, the present study proposes a novel hybrid deep learning framework that integrates Convolutional Neural Networks (CNN) for feature extraction with Transformer architecture for temporal modeling. The model is validated using NASA’s CMAPSS dataset, a widely used benchmark that includes multi-sensor data and Remaining Useful Life (RUL) labels from aeroengines. By learning from time-series sensor data, the framework achieves accurate RUL predictions and early fault detection. Experimental results show that the model attains over 97% accuracy under both single and multiple operating conditions, highlighting its robustness and adaptability. These findings suggest the framework’s potential in developing intelligent maintenance systems and contribute to the field of Prognostics and Health Management (PHM), enabling more reliable, efficient, and self-monitoring industrial systems. Full article

(This article belongs to the Special Issue Communications and Networking Based on Artificial Intelligence)

► Show Figures

Figure 1

21 pages, 21726 KB

Open AccessEditor’s ChoiceArticle

Evaluation of Positioning Accuracy Using Smartphone RGB and LiDAR Sensors with the viDoc RTK Rover

by Sara Zollini and Laura Marconi

Sensors 2025, 25(13), 3867; https://doi.org/10.3390/s25133867 - 21 Jun 2025

Cited by 3 | Viewed by 5475

Abstract

Modern surveying is increasingly focused on fast data acquisition and processing using lightweight, low-cost equipment, particularly for the continuous monitoring of structures and infrastructures. This study investigates the use of LiDAR and RGB sensors embedded in Apple and Android smartphones, paired with an [...] Read more.

Modern surveying is increasingly focused on fast data acquisition and processing using lightweight, low-cost equipment, particularly for the continuous monitoring of structures and infrastructures. This study investigates the use of LiDAR and RGB sensors embedded in Apple and Android smartphones, paired with an innovative device, the viDoc RTK Rover, for centimeter-level surveying. Three case studies were selected, each characterized by different materials, functional uses, and environmental contexts. The methodology centers on evaluating final accuracy during both the data acquisition and processing phases. Coordinates of target points were obtained directly via the viDoc device and indirectly through dense point clouds. Validation was conducted using a geodetic GNSS receiver. Results demonstrate that, in most cases, the system achieves accuracy comparable to traditional surveying methods. The findings confirm that these emerging tools offer a reliable and efficient solution for rapid 3D surveys with centimeter-level accuracy. Full article

(This article belongs to the Section Navigation and Positioning)

► Show Figures

Figure 1

28 pages, 3303 KB

Open AccessEditor’s ChoiceReview

Structural Fault Detection and Diagnosis for Combine Harvesters: A Critical Review

by Haiyang Wang, Liyun Lao, Honglei Zhang, Zhong Tang, Pengfei Qian and Qi He

Sensors 2025, 25(13), 3851; https://doi.org/10.3390/s25133851 - 20 Jun 2025

Cited by 4 | Viewed by 1992

Abstract

Combine harvesters, as essential equipment in agricultural engineering, frequently experience structural faults due to their complex structure and harsh working conditions, which severely affect their reliability and operational efficiency, leading to significant downtime and reduced agricultural productivity during critical harvesting periods. Therefore, developing [...] Read more.

Combine harvesters, as essential equipment in agricultural engineering, frequently experience structural faults due to their complex structure and harsh working conditions, which severely affect their reliability and operational efficiency, leading to significant downtime and reduced agricultural productivity during critical harvesting periods. Therefore, developing accurate and timely Fault Detection and Diagnosis (FDD) techniques is crucial for ensuring food security. This paper provides a systematic and critical review and analysis of the latest advancements in research on data-driven FDD methods for structural faults in combine harvesters. First, it outlines the typical structural sections of combine harvesters and their common structural fault types. Subsequently, it details the core steps of data-driven methods, including the acquisition of operational data from various sensors (e.g., vibration, acoustic, strain), signal preprocessing methods, signal processing and feature extraction techniques covering time-domain, frequency-domain, time–frequency domain combination, and modal analysis among others, and the use of machine learning and artificial intelligence models for fault pattern learning and diagnosis. Furthermore, it explores the required system and technical support for implementing such data-driven FDD methods, such as the applications of on-board diagnostic units, remote monitoring platforms, and simulation modeling. It provides an in-depth analysis of the key challenges currently encountered in this field, including difficulties in data acquisition, signal complexity, and insufficient model robustness, and consequently proposes future research directions, aiming to provide insights for the development of intelligent maintenance and efficient and reliable operation of combine harvesters and other complex agricultural machinery. Full article

(This article belongs to the Special Issue Feature Review Papers in Fault Diagnosis & Sensors)

► Show Figures

Figure 1

18 pages, 1032 KB

Open AccessEditor’s ChoiceArticle

AI for Sustainable Recycling: Efficient Model Optimization for Waste Classification Systems

by Oriol Chacón-Albero, Mario Campos-Mocholí, Cédric Marco-Detchart, Vicente Julian, Jaime Andrés Rincon and Vicent Botti

Sensors 2025, 25(12), 3807; https://doi.org/10.3390/s25123807 - 18 Jun 2025

Cited by 4 | Viewed by 3737

Abstract

The increasing volume of global waste presents a critical environmental and societal challenge, demanding innovative solutions to support sustainable practices such as recycling. Advances in Computer Vision (CV) have enabled automated waste recognition systems that guide users in correctly sorting their waste, with [...] Read more.

The increasing volume of global waste presents a critical environmental and societal challenge, demanding innovative solutions to support sustainable practices such as recycling. Advances in Computer Vision (CV) have enabled automated waste recognition systems that guide users in correctly sorting their waste, with state-of-the-art architectures achieving high accuracy. More recently, attention has shifted toward lightweight and efficient models suitable for mobile and edge deployment. These systems process data from integrated camera sensors in Internet of Things (IoT) devices, operating in real time to classify waste at the point of disposal, whether embedded in smart bins, mobile applications, or assistive tools for household use. In this work, we extend our previous research by improving both dataset diversity and model efficiency. We introduce an expanded dataset that includes an organic waste class and more heterogeneous images, and evaluate a range of quantized CNN models to reduce inference time and resource usage. Additionally, we explore ensemble strategies using aggregation functions to boost classification performance, and validate selected models on real embedded hardware and under simulated lighting variations. Our results support the development of robust, real-time recycling assistants for resource-constrained devices. We also propose architectural deployment scenarios for smart containers, and cloud-assisted solutions. By improving waste sorting accuracy, these systems can help reduce landfill use, support citizen engagement through real-time feedback, increase material recovery, support data-informed environmental decision making, and ease operational challenges for recycling facilities caused by misclassified materials. Ultimately, this contributes to circular economy objectives and advances the broader field of environmental intelligence. Full article

(This article belongs to the Special Issue Internet of Things (IoT) and Cloud Computing Technologies for Smart Cities and Rural Areas: Sensing and Information Management Technologies)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Editor’s Choice Articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI