Sensors

21 pages, 11172 KiB

Open AccessArticle

Detection and Pattern Recognition of Chemical Warfare Agents by MOS-Based MEMS Gas Sensor Array

by Mengxue Xu, Xiaochun Hu, Hongpeng Zhang, Ting Miao, Lan Ma, Jing Liang, Yuefeng Zhu, Haiyan Zhu, Zhenxing Cheng and Xuhui Sun

Sensors 2025, 25(8), 2633; https://doi.org/10.3390/s25082633 - 21 Apr 2025

Abstract

Chemical warfare agents (CWAs), including hydrogen cyanide (AC), 2-[fluoro(methyl)phosphoryl]oxypropane (GB), 3-[fluoro(methyl)phosphoryl]oxy-2,2-dimethylbutane (GD), ethyl S-(2-diisopropylaminoethyl) methylphosphonothioate (VX), and di-2-chloroethyl sulfide (HD), pose a great threat to public safety; therefore, it is important to develop sensing technology for CWAs. Herein, a sensor array consisting of [...] Read more.

Chemical warfare agents (CWAs), including hydrogen cyanide (AC), 2-[fluoro(methyl)phosphoryl]oxypropane (GB), 3-[fluoro(methyl)phosphoryl]oxy-2,2-dimethylbutane (GD), ethyl S-(2-diisopropylaminoethyl) methylphosphonothioate (VX), and di-2-chloroethyl sulfide (HD), pose a great threat to public safety; therefore, it is important to develop sensing technology for CWAs. Herein, a sensor array consisting of 24 metal oxide semiconductor (MOS)-based MEMS sensors with good gas sensing performance, a simple device structure (0.9 mm × 0.9 mm), and low power consumption (<10 mW on average) was developed. The experimental results show that there are always several sensors among the 24 sensors that show good sensing performance in relation to each CWA, such as a relatively significant response, a broad detection range (AC: 5.8–89 ppm; GB: 0.04–0.47 ppm; GD: 0.06–4.7 ppm; VX: 9.978 × 10⁻⁴–1.101 × 10⁻³; HD: 0.61–4.9 ppm), and a low detection limit that is lower than the immediately dangerous for life and health (IDLH) level of the five CWAs. This indicates that these sensors can meet the needs for qualitative detection and can provide an early warning regarding low concentrations of CWAs. In addition, features were extracted from the initial kinetic characteristics and dynamic change characteristics of the sensing response. Finally, principal component analysis (PCA) and machine learning algorithms were applied for CWA classification. The obtained PCA plots showed significant differences between groups, and the narrow neural network among the machine learning algorithms achieves a prediction accuracy of nearly 100.0%. In summary, the proposed MOS-based MEMS sensor array driven by pattern recognition algorithms can be integrated into portable devices, showing great potential and practical applications in the rapid, in situ, and on-site detection and identification of CWAs. Full article

(This article belongs to the Section Chemical Sensors)

► Show Figures

Figure 1

15 pages, 5551 KiB

Open AccessArticle

Online Nodal Demand Estimation in Branched Water Distribution Systems Using an Array of Extended Kalman Filters

by Francisco-Ronay López-Estrada, Leonardo Gómez-Coronel, Lizeth Torres, Guillermo Valencia-Palomo, Ildeberto Santos-Ruiz and Arlette Cano

Sensors 2025, 25(8), 2632; https://doi.org/10.3390/s25082632 - 21 Apr 2025

Abstract

This paper proposes a model-based methodology to estimate multiple nodal demands by using only pressure and flow rate measurements, which should be recorded at the inlet of the distribution system. The algorithm is based on an array of multiple extended Kalman filters (EKFs) [...] Read more.

This paper proposes a model-based methodology to estimate multiple nodal demands by using only pressure and flow rate measurements, which should be recorded at the inlet of the distribution system. The algorithm is based on an array of multiple extended Kalman filters (EKFs) in a cascade configuration. Each EKF functions as an unknown input observer and focuses on a section of the pipeline. The pipeline model used to design the filters is an adaptation of the well-known rigid water column model. Simulation and experimental tests on standardized pipeline systems are presented to demonstrate the proposed method’s effectiveness. Finally, for the case of the experimental validation, both steady-state and variable input conditions were considered. Full article

(This article belongs to the Special Issue Sensor-Based State Estimation and Fault Diagnosis in Automatic Control)

► Show Figures

Figure 1

14 pages, 2577 KiB

Open AccessArticle

Dual-Branch Cross-Fusion Normalizing Flow for RGB-D Track Anomaly Detection

by Xiaorong Gao, Pengxu Wen, Jinlong Li and Lin Luo

Sensors 2025, 25(8), 2631; https://doi.org/10.3390/s25082631 - 21 Apr 2025

Abstract

With the ease of acquiring RGB-D images from line-scan 3D cameras and the development of computer vision, anomaly detection is now widely applied to railway inspection. As 2D anomaly detection is susceptible to capturing condition, a combination of depth maps is now being [...] Read more.

With the ease of acquiring RGB-D images from line-scan 3D cameras and the development of computer vision, anomaly detection is now widely applied to railway inspection. As 2D anomaly detection is susceptible to capturing condition, a combination of depth maps is now being explored in industrial inspection to reduce these interferences. In this case, this paper proposes a novel approach for RGB-D anomaly detection called Dual-Branch Cross-Fusion Normalizing Flow (DCNF). In this work, we aim to exploit the fusion strategy for dual-branch normalizing flow with multi-modal inputs to be applied in the field of track detection. On the one hand, we introduce the mutual perception module to acquire cross-complementary prior knowledge in the early stage. On the other hand, we exploit the effectiveness of the fusion flow to fuse the dual-branch of RGB-D inputs. We experiment on the real-world Track Anomaly (TA) dataset. The performance evaluation of DCNF on TA dataset achieves an impressive AUROC score of 98.49%, which is 3.74% higher than the second-best method. Full article

(This article belongs to the Section Sensing and Imaging)

► Show Figures

Figure 1

21 pages, 21463 KiB

Open AccessArticle

A Study of Corrosion-Grade Recognition on Metal Surfaces Based on Improved YOLOv8 Model

by Hao Chen, Ying Cao, Shengxian Cao and Heng Piao

Sensors 2025, 25(8), 2630; https://doi.org/10.3390/s25082630 - 21 Apr 2025

Abstract

Typical metal equipment in substations is exposed to high-temperature, high-humidity, and high-salt environments for a long time, and surface corrosion is a serious threat to operational safety. Traditional manual inspection is limited by the complexity of the environment and subjective assessment errors, and [...] Read more.

Typical metal equipment in substations is exposed to high-temperature, high-humidity, and high-salt environments for a long time, and surface corrosion is a serious threat to operational safety. Traditional manual inspection is limited by the complexity of the environment and subjective assessment errors, and there is an urgent need for a method that can quickly and accurately locate the corrosion area and assess the degree of corrosion. In this paper, based on YOLOv8, the feature extraction ability is improved by introducing the attention mechanism; a mixed-mixed-sample data augmentation algorithm is designed to increase the diversity of data; and a cosine annealing learning rate adjustment is adopted to improve the training efficiency. The corrosion process of metal materials is accelerated by a neutral salt spray test in order to collect corrosion samples at different stages and establish a dataset, and a model of a corrosion-state recognition algorithm for typical equipment in substations based on an improved YOLOv8 model is established. Finally, based on ablation experiments and comparison experiments, performance analyses of multiple algorithmic models are conducted for horizontal and vertical comparisons in order to verify the effectiveness of the improved method and the superiority of the models in this paper. The experiments verify that the improved model is comprehensively leading in multi-dimensional indicators: the mAP reaches 96.3% and the F1 score reaches 93.6%, which is significantly better than mainstream models such as Faster R-CNN, and provides a reliable technical solution for the intelligent inspection of substation equipment. Full article

(This article belongs to the Section Physical Sensors)

► Show Figures

Figure 1

15 pages, 6517 KiB

Open AccessArticle

A Fading Suppression Method for Φ-OTDR Systems Based on Multi-Domain Multiplexing

by Shuai Tong, Shaoxiong Tang, Yifan Lu, Nuo Yuan, Chi Zhang, Huanhuan Liu, Dao Zhang, Ningmu Zou, Xuping Zhang and Yixin Zhang

Sensors 2025, 25(8), 2629; https://doi.org/10.3390/s25082629 - 21 Apr 2025

Abstract

The phase-sensitive optical time domain reflectometry (Φ-OTDR) has been widely applied in various fields. However, due to fading noise, false alarms often occur; this limits its engineering applications. In this paper, a fading suppression method for Φ-OTDR systems based on multi-domain multiplexing (MDM) [...] Read more.

The phase-sensitive optical time domain reflectometry (Φ-OTDR) has been widely applied in various fields. However, due to fading noise, false alarms often occur; this limits its engineering applications. In this paper, a fading suppression method for Φ-OTDR systems based on multi-domain multiplexing (MDM) is proposed. The principles and limitations of existing suppression methods such as spatial-domain multiplexing (SDM), polarization-domain multiplexing (PDM), and frequency-domain multiplexing (FDM) are analyzed. The principle of MDM is explained in detail, and an experimental system is established to test the fading noise suppression capabilities of different parameter combinations of the PDM, FDM, and SDM methods. Experimental results show that it is difficult to comprehensively suppress fading noise with single-domain multiplexing. Through optimizations of different parameter combinations, MDM can comprehensively suppress fading noise by appropriately selecting the number of FDM frequencies, the spatial weighting intervals, and using PDM, thus obtaining the optimal anti-fading solution between performance and hardware complexity. Through MDM, the fade-free measurement is achieved, providing a promising technical solution for the practical application of the Φ-OTDR technology. Full article

(This article belongs to the Section Optical Sensors)

► Show Figures

Figure 1

15 pages, 3682 KiB

Open AccessArticle

Multi-Sensor Information Fusion Positioning of AUKF Maglev Trains Based on Self-Corrected Weighting

by Qian Hu, Hong Tang, Kuangang Fan and Wenlong Cai

Sensors 2025, 25(8), 2628; https://doi.org/10.3390/s25082628 - 21 Apr 2025

Abstract

Achieving accurate positioning of maglev trains is one of the key technologies for the safe operation of maglev trains and train schedules. Aiming at magnetic levitation train positioning, there are problems such as being easily interfered with by external noise, the single positioning [...] Read more.

Achieving accurate positioning of maglev trains is one of the key technologies for the safe operation of maglev trains and train schedules. Aiming at magnetic levitation train positioning, there are problems such as being easily interfered with by external noise, the single positioning method, and traditional weighting affected by historical data, which lead to the deviation of positioning fusion results. Therefore, this paper adopts self-corrected weighting and Sage–Husa noise estimation algorithms to improve them and proposes a research method of multi-sensor information fusion and positioning of an AUKF magnetic levitation train based on self-correcting weighting. Multi-sensor information fusion technology is applied to the positioning of maglev trains, which does not rely on a single sensor. It combines the Sage–Husa algorithm with the unscented Kalman filter (UKF) to form the AUKF algorithm using the data collected by the cross-sensor lines, INS, Doppler radar, and GNSS, which adaptively updates the statistical feature estimation of the measurement noise and eliminates the single-function and low-integration shortcomings of the various modules to achieve the precise positioning of maglev trains. The experimental results point out that the self-correction-based AUKF filter trajectories are closer to the real values, and their ME and RMSE errors are smaller, indicating that the self-correction-weighted AUKF algorithm proposed in this paper has significant advantages in terms of stability, accuracy, and simplicity. Full article

(This article belongs to the Section Navigation and Positioning)

► Show Figures

Figure 1

16 pages, 3900 KiB

Open AccessArticle

Classifying Storage Temperature for Mandarin (Citrus reticulata L.) Using Bioimpedance and Diameter Measurements with Machine Learning

by Daesik Son, Siun Lee, Sehyeon Jeon, Jae Joon Kim and Soo Chung

Sensors 2025, 25(8), 2627; https://doi.org/10.3390/s25082627 - 21 Apr 2025

Abstract

Mandarin (Citrus reticulata L.) is consumed worldwide. Improper storage temperatures cause flavor loss and shorten shelf lives, reducing marketability. Mandarins’ quality is difficult to assess visually, as they show no apparent changes during storage. Therefore, a simple, non-destructive method is needed to [...] Read more.

Mandarin (Citrus reticulata L.) is consumed worldwide. Improper storage temperatures cause flavor loss and shorten shelf lives, reducing marketability. Mandarins’ quality is difficult to assess visually, as they show no apparent changes during storage. Therefore, a simple, non-destructive method is needed to assess their freshness as affected by temperature. This work utilized non-invasive bioimpedance spectroscopy (BIS) on mandarins stored at different temperatures. Eight machine learning (ML) models were trained with bioimpedance data to classify storage temperature. Also, we confirmed whether integrating diameter and time-series changes into the bioimpedance could improve the ML models’ accuracies by minimizing sample variations. Additionally, we evaluated the effectiveness of equivalent circuit (EC) parameters derived from bioimpedance data for ML training. Although slightly less accurate than using raw bioimpedance data, EC parameters can efficiently reduce data dimensionality. Among all models, the SVM model trained with changes in bioimpedance integrated with diameter data achieved the highest accuracy of 0.92. It was a significant improvement compared to the accuracy of 0.76 achieved when using only the raw bioimpedance data. Thus, this study suggests a novel method of integrating diameter and bioimpedance changes to assess the storage temperature of mandarins. This approach can also be applied to other fruits when utilizing BIS. Full article

(This article belongs to the Special Issue Bioimpedance Measurements and Microelectrodes)

► Show Figures

Graphical abstract

18 pages, 7212 KiB

Open AccessArticle

Integrating Complex Permittivity Measurements with Histological Analysis for Advanced Tissue Characterization

by Sandra Lopez-Prades, Mónica Torrecilla-Vall-llossera, Mercedes Rus, Miriam Cuatrecasas and Joan M. O’Callaghan

Sensors 2025, 25(8), 2626; https://doi.org/10.3390/s25082626 - 21 Apr 2025

Abstract

We developed a measurement setup and protocol reliably relating complex permittivity measurements with tissue characterization and specific histological features. We measured 148 fresh human tissue samples across 14 tissue types at 51 frequencies ranging from 200 MHz to 20 GHz, using an open-ended [...] Read more.

We developed a measurement setup and protocol reliably relating complex permittivity measurements with tissue characterization and specific histological features. We measured 148 fresh human tissue samples across 14 tissue types at 51 frequencies ranging from 200 MHz to 20 GHz, using an open-ended coaxial slim probe. Tissue samples were collected using a punch biopsy, ensuring that the sampled area encompassed the region where complex permittivity measurements were performed. This approach minimized experimental uncertainty related to potential position-dependent variations in permittivity. Once measured, the samples were then formalin-fixed and paraffin-embedded (FFPE) to obtain histological slides for microscopic analysis of tissue features. We observed that complex permittivity values are strongly associated with key histological features, including fat content, necrosis, and fibrosis. Most tissue samples exhibiting these features could be differentiated from nominal values for that tissue type, even accounting for statistical variability and instrumental uncertainties. These findings demonstrate the potential of incorporating fast in situ complex permittivity for fresh tissue characterization in pathology workflows. Furthermore, our work lays the groundwork for enhancing databases where complex permittivity values are measured under histological control, enabling precise correlations between permittivity values, tissue characterization, and histological features. Full article

(This article belongs to the Special Issue Advanced Sensing and Signal Processing Technologies for Medical Applications)

► Show Figures

Figure 1

34 pages, 2603 KiB

Open AccessReview

Research and Prospects of Digital Twin-Based Fault Diagnosis of Electric Machines

by Jiaqi Hu, Han Xiao, Zhihao Ye, Ningzhao Luo and Minhao Zhou

Sensors 2025, 25(8), 2625; https://doi.org/10.3390/s25082625 - 21 Apr 2025

Abstract

This paper focuses on the application of digital twins in the field of electric motor fault diagnosis. Firstly, it explains the origin, concept, key technology and application areas of digital twins, compares and analyzes the advantages and disadvantages of digital twin technology and [...] Read more.

This paper focuses on the application of digital twins in the field of electric motor fault diagnosis. Firstly, it explains the origin, concept, key technology and application areas of digital twins, compares and analyzes the advantages and disadvantages of digital twin technology and traditional methods in the application of electric motor fault diagnosis, discusses in depth the key technology of digital twins in electric motor fault diagnosis, including data acquisition and processing, digital modeling, data analysis and mining, visualization technology, etc., and enumerates digital twin application examples in the fields of induction motors, permanent magnet synchronous motors, wind turbines and other motor fields. A concept of multi-phase synchronous generator fault diagnosis based on digital twins is given, and challenges and future development directions are discussed. Full article

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

► Show Figures

Figure 1

20 pages, 4322 KiB

Open AccessArticle

A Wearable Silent Text Input System Using EMG and Piezoelectric Sensors

by John S. Kang, Kee S. Moon, Sung Q. Lee, Nicholas Satterlee and Xiaowei Zuo

Sensors 2025, 25(8), 2624; https://doi.org/10.3390/s25082624 - 21 Apr 2025

Abstract

This paper introduces a wearable silent text input system designed to capture text input through silent speech, without generating audible sound. The system integrates Electromyography (EMG) and piezoelectric lead zirconate titanate (PZT) sensors in a miniaturized form that can be comfortably attached to [...] Read more.

This paper introduces a wearable silent text input system designed to capture text input through silent speech, without generating audible sound. The system integrates Electromyography (EMG) and piezoelectric lead zirconate titanate (PZT) sensors in a miniaturized form that can be comfortably attached to the chin, making it both comfortable to wear and esthetically pleasing. The EMG sensor records muscle activity linked to specific tongue and jaw movements, while the PZT sensor measures the minute vibrations and pressure changes in the chin skin caused by silent speech. Data from both sensors are analyzed to capture the timing and intensity of the silent speech signals, allowing the extraction of key features in both time and frequency domain. Several machine learning (ML) models, including both feature-based and non-feature-based approaches commonly used for classification tasks, are employed and compared to detect and classify subtle variations in sensor signals associated with individual alphabet letters. To evaluate and compare the ML models, EMG and PZT signals for the eight most frequently used English letters are collected across one hundred trials each. Results showed that non-feature-based models, particularly the Fea-Shot Learning with fused EMG and PZT signals, achieved the highest accuracy (95.63%) and F1-score (95.62%). The proposed system’s accuracy and real-time performance make it promising for silent text input and assistive communication applications. Full article

(This article belongs to the Special Issue Biosignal Sensing Analysis (EEG, EMG, ECG, PPG) (2nd Edition))

► Show Figures

Figure 1

37 pages, 11936 KiB

Open AccessArticle

A Vision-Based Method for Detecting the Position of Stacked Goods in Automated Storage and Retrieval Systems

by Chuanjun Chen, Junjie Liu, Haonan Yin and Biqing Huang

Sensors 2025, 25(8), 2623; https://doi.org/10.3390/s25082623 - 21 Apr 2025

Abstract

Automated storage and retrieval systems (AS/RS) play a crucial role in modern logistics, yet effectively monitoring cargo stacking patterns remains challenging. While computer vision and deep learning offer promising solutions, existing methods struggle to balance detection accuracy, computational efficiency, and environmental adaptability. This [...] Read more.

Automated storage and retrieval systems (AS/RS) play a crucial role in modern logistics, yet effectively monitoring cargo stacking patterns remains challenging. While computer vision and deep learning offer promising solutions, existing methods struggle to balance detection accuracy, computational efficiency, and environmental adaptability. This paper proposes a novel machine vision-based detection algorithm that integrates a pallet surface object detection network (STEGNet) with a box edge detection algorithm. STEGNet’s core innovation is the Efficient Gated Pyramid Feature Network (EG-FPN), which integrates a Gated Feature Fusion module and a Lightweight Attention Mechanism to optimize feature extraction and fusion. In addition, we introduce a geometric constraint method for box edge detection and employ a Perspective-n-Point (PnP)-based 2D-to-3D transformation approach for precise pose estimation. Experimental results show that STEGNet achieves 93.49% mAP on our proposed GY Warehouse Box View 4-Dimension (GY-WSBW-4D) dataset and 83.2% mAP on the WSGID-B dataset, surpassing existing benchmarks. The lightweight variant maintains competitive accuracy while reducing the model size by 34% and increasing the inference speed by 68%. In practical applications, the system achieves pose estimation with a Mean Absolute Error within 4 cm and a Rotation Angle Error below 2°, demonstrating robust performance in complex warehouse environments. This research provides a reliable solution for automated cargo stack monitoring in modern logistics systems. Full article

(This article belongs to the Section Sensing and Imaging)

► Show Figures

Figure 1

22 pages, 2956 KiB

Open AccessArticle

Novel Dual-Constraint-Based Semi-Supervised Deep Clustering Approach

by Mona Suliman AlZuhair, Mohamed Maher Ben Ismail and Ouiem Bchir

Sensors 2025, 25(8), 2622; https://doi.org/10.3390/s25082622 - 21 Apr 2025

Abstract

Semi-supervised clustering can be viewed as a clustering paradigm that exploits both labeled and unlabeled data to steer learning accurate data clusters and avoid local minimum solutions. Nonetheless, the attempts to refine existing semi-supervised clustering methods are relatively limited when compared to the [...] Read more.

Semi-supervised clustering can be viewed as a clustering paradigm that exploits both labeled and unlabeled data to steer learning accurate data clusters and avoid local minimum solutions. Nonetheless, the attempts to refine existing semi-supervised clustering methods are relatively limited when compared to the advancements witnessed in the current benchmark methods in fully unsupervised clustering. This research introduces a novel semi-supervised method for deep clustering that leverages deep neural networks and fuzzy memberships to better capture the data partitions. In particular, the proposed Dual-Constraint-based Semi-Supervised Deep Clustering (DC-SSDEC) method utilizes two sets of pairwise soft constraints; “should-link” and “shouldNot-link”, to guide the clustering process. The intended clustering task is expressed as an optimization of a newly designed objective function. Additionally, DC-SSDEC performance was evaluated through comprehensive experiments using three real-world and benchmark datasets. Moreover, a comparison with related state-of-the-art clustering techniques was conducted to showcase the DC-SSDEC outperformance. In particular, DC-SSDEC significance consists of the proposed dual-constraint formulation and its integration into a novel objective function. This contribution yielded an improvement in the resulting clustering performance compared to relevant state-of-the-art approaches. In addition, the assessment of the proposed model using real-world datasets represents another contribution of this research. In fact, increases of 3.25%, 1.44%, and 1.82% in the clustering accuracy were gained by DC-SSDEC over the best performing single-constraint-based approach, using MNIST, STL-10, and USPS datasets, respectively. Full article

(This article belongs to the Section Intelligent Sensors)

► Show Figures

Figure 1

15 pages, 1261 KiB

Open AccessArticle

Validation of the Use of a Smart Band in Recording Spatiotemporal Gait Parameters in the 6-Minute Walk Test

by Rosa María Ortiz-Gutiérrez, José Javier López-Marcos, José Luis Maté-Muñoz, Paloma Moreta-de-Esteban and Patricia Martín-Casas

Sensors 2025, 25(8), 2621; https://doi.org/10.3390/s25082621 - 21 Apr 2025

Abstract

Wearable monitoring devices, such as smart bands, have emerged as accessible and non-invasive tools for assessing physiological and spatiotemporal gait parameters in various clinical tests. This study aimed to validate the use of the Xiaomi Mi Band 6 for recording gait parameters during [...] Read more.

Wearable monitoring devices, such as smart bands, have emerged as accessible and non-invasive tools for assessing physiological and spatiotemporal gait parameters in various clinical tests. This study aimed to validate the use of the Xiaomi Mi Band 6 for recording gait parameters during the six-minute walk test (6MWT). Seventy participants without gait impairments were recruited, and the measurements obtained with the smart band were compared to reference methods (evaluator, pedometer, and pulse oximeter). The physiological parameter results showed that the smart band demonstrated good accuracy in heart rate monitoring but lower agreement in oxygen saturation measurements. Gait parameters indicated excellent agreement in step count (ICC > 0.9) and step frequency (ICC > 0.75), whereas step length and distance estimations showed higher variability. These findings suggest that the Xiaomi Mi Band 6 is a viable alternative for measuring specific gait parameters, though with limitations in certain aspects of accuracy. Full article

(This article belongs to the Special Issue Sensors and Artificial Intelligence for Analyzing Human Behavior in Sports and Physical Activity)

► Show Figures

Figure 1

18 pages, 5455 KiB

Open AccessArticle

Three-Dimensional Focusing Measurement Method for Confocal Microscopy Based on Liquid Crystal Spatial Light Modulator

by Yupeng Li and Yifan Li

Sensors 2025, 25(8), 2620; https://doi.org/10.3390/s25082620 - 21 Apr 2025

Abstract

Micro-nano measurement represents a critical engineering focus in the advancement of micro-nano fabrication technologies. Exploring advanced micro-nano measurement methods is a key direction for driving progress in micro-nano manufacturing. This study proposes a confocal measurement method utilizing a liquid crystal spatial light modulator [...] Read more.

Micro-nano measurement represents a critical engineering focus in the advancement of micro-nano fabrication technologies. Exploring advanced micro-nano measurement methods is a key direction for driving progress in micro-nano manufacturing. This study proposes a confocal measurement method utilizing a liquid crystal spatial light modulator (LC-SLM) to simulate a binary Fresnel lens for 3D focusing, enabling the non-mechanical measurement of spatial positions on sample surfaces. Specifically, it introduces a 3D focusing method based on LC-SLM, constructs a confocal microscopy 3D focusing system, and conducts lateral focusing experiments and axial focusing experiments. Experimental results demonstrate that the system can freely adjust lateral focusing positions. Within an axial focusing range of 900 μm, it achieves axial measurement accuracy exceeding 1 μm, with a maximum resolution capability of approximately 16.667 nm. Compared to similar confocal microscopy systems, this method allows rapid adjustment of lateral focusing positions without regenerating phase grayscale maps, achieves comparable axial measurement accuracy, and enhances measurement speed. Full article

(This article belongs to the Section Optical Sensors)

► Show Figures

Figure 1

23 pages, 6882 KiB

Open AccessArticle

Research on Displacement Monitoring of Key Points in Caverns Based on Distributed Fiber Optic Sensing Technology

by Jiangdong Wang, Ziming Xiong, Sheng Li, Hao Lu, Minqian Sun, Zhizhong Li and Hao Chen

Sensors 2025, 25(8), 2619; https://doi.org/10.3390/s25082619 - 21 Apr 2025

Abstract

The accurate and real-time monitoring of key-point displacements in cavern structures is crucial for assessing structural safety and stability. However, traditional monitoring methods often fail to meet the high-precision requirements in complex environments. This study explored the potential application of fiber optic sensors [...] Read more.

The accurate and real-time monitoring of key-point displacements in cavern structures is crucial for assessing structural safety and stability. However, traditional monitoring methods often fail to meet the high-precision requirements in complex environments. This study explored the potential application of fiber optic sensors in monitoring key-point displacements by leveraging their sensitivity to optical parameters and spectral changes. Through theoretical analysis, a linear relationship model between key-point displacements and circumferential strain was derived and validated via uniaxial compression tests. Further numerical simulations revealed that different material properties and structural characteristics significantly affect the slope and intercept of the fitting curve, establishing a correlation between these factors and the model parameters. The results demonstrated that fiber optic sensors could accurately measure circumferential strain within the elastic range and reliably reflect key-point displacement trends through the linear relationship model. This paper provides a new theoretical basis for the application of fiber optic sensors in structural health monitoring and expands their potential in civil and geotechnical engineering fields, offering scientific support for engineering design optimization and disaster prevention. Full article

(This article belongs to the Topic Recent Advances in Structural Health Monitoring, 2nd Volume)

► Show Figures

Figure 1

11 pages, 800 KiB

Open AccessArticle

Combined Proxies for Heart Rate Variability as a Global Tool to Assess and Monitor Autonomic Dysregulation in Fibromyalgia and Disease-Related Impairments

by Emanuella Ladisa, Chiara Abbatantuono, Elena Ammendola, Giusy Tancredi, Marianna Delussi, Giulia Paparella, Livio Clemente, Annalisa Di Dio, Antonio Federici and Marina de Tommaso

Sensors 2025, 25(8), 2618; https://doi.org/10.3390/s25082618 - 21 Apr 2025

Abstract

Background: Heart rate variability (HRV) provides both linear and nonlinear autonomic proxies that can be informative of health status in fibromyalgia (FM), where sympatho-vagal abnormalities are common. This retrospective observational study aims to: 1. detect differences in correlation dimension (D2) between FM patients [...] Read more.

Background: Heart rate variability (HRV) provides both linear and nonlinear autonomic proxies that can be informative of health status in fibromyalgia (FM), where sympatho-vagal abnormalities are common. This retrospective observational study aims to: 1. detect differences in correlation dimension (D2) between FM patients and healthy controls (HCs); 2. correlate D2 with standard HRV parameters; 3. correlate the degree of HRV changes using a global composite parameter called HRV grade, derived from three linear indices (SDNN = intervals between normal sinus beats; RMSSD = mean square of successive differences; total power), with FM clinical outcomes; 4. correlate all linear and nonlinear HRV parameters with clinical variables in patients. Methods: N = 85 patients were considered for the analysis and compared to 35 healthy subjects. According to standard diagnostic protocol, they underwent a systematic HRV protocol with a 5-min paced breathing task. Disease duration, pain intensity, mood, sleep, fatigue, and quality of life were assessed. Non-parametric tests for independent samples and pairwise correlations were performed using JMP (all p < 0.001). Results: Mann-Whitney U found a significant difference in D2 values between FM patients and HCs (p < 0.001). In patients, D2 was associated with all HRV standard indices (all p < 0.001) and FM impairment (FIQ = −0.4567; p < 0.001). HRV grade was also associated with FM impairment (FIQ = 0.5058; p < 0.001). Conclusion: Combining different HRV measurements may help understand the correlates of autonomic dysregulation in FM. Specifically, clinical protocols could benefit from the inclusion and validation of D2 and HRV parameters to target FM severity and related dysautonomia. Full article

(This article belongs to the Special Issue Intelligent Medical Sensors and Applications)

► Show Figures

Figure 1

28 pages, 14146 KiB

Open AccessArticle

Virtual MOS Sensor Array Design for Ammonia Monitoring in Pig Barns

by Raphael Parsiegel, Miguel Budag Becker, Pieter Try and Marion Gebhard

Sensors 2025, 25(8), 2617; https://doi.org/10.3390/s25082617 - 20 Apr 2025

Abstract

Animal welfare in barns is strongly influenced by air quality, with gaseous emissions like ammonia posing significant respiratory health risks. However, current state-of-the-art ammonia monitoring systems are labor-intensive and expensive. Metal Oxide Semiconductor (MOS) sensors offer a promising alternative due to their compatibility [...] Read more.

Animal welfare in barns is strongly influenced by air quality, with gaseous emissions like ammonia posing significant respiratory health risks. However, current state-of-the-art ammonia monitoring systems are labor-intensive and expensive. Metal Oxide Semiconductor (MOS) sensors offer a promising alternative due to their compatibility with sensor networks, enabling high-resolution ammonia monitoring across spatial and temporal scales. While MOS sensors exhibit high sensitivity to various volatile compounds, temperature-cycled operation is commonly employed to enhance selectivity, effectively creating virtual sensor arrays. This study aims to improve ammonia detection by designing a virtual sensor array through a cyclic data-driven approach, integrating machine learning with solid-state sensor modeling. The results of a two-week dataset with measurements of four different pig barns demonstrate ammonia sensing with a sampling rate of about 2/min and a range of 1–30 ppm. The method is robust and exhibits a 10 increase in normalized RMSE when comparing testing results of an unseen sensor module with results of the training dataset. A filter membrane boosts accuracy and prevents data loss due to contamination, such as flyspecks. Overall, the used MOS sensor BME688 is effective and economical for widespread continuous ammonia monitoring and localization of ammonia sources in pig barns. Full article

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

24 pages, 4034 KiB

Open AccessArticle

Dual-Layer Fusion Model Using Bayesian Optimization for Asphalt Pavement Condition Index Prediction

by Jun Hao, Zhaoyun Sun, Zhenzhen Xing, Lili Pei and Xin Feng

Sensors 2025, 25(8), 2616; https://doi.org/10.3390/s25082616 - 20 Apr 2025

Abstract

To address the technical limitations of traditional pavement performance prediction models in capturing temporal features and analyzing multi-factor coupling, this study proposes a Bayesian Optimization Dual-Layer Feature Fusion Model (BO-DLFF). The framework integrates heterogeneous data streams from embedded strain sensors, temperature/humidity monitoring nodes, [...] Read more.

To address the technical limitations of traditional pavement performance prediction models in capturing temporal features and analyzing multi-factor coupling, this study proposes a Bayesian Optimization Dual-Layer Feature Fusion Model (BO-DLFF). The framework integrates heterogeneous data streams from embedded strain sensors, temperature/humidity monitoring nodes, and weigh-in-motion (WIM) systems, combined with pavement distress detection and historical maintenance records. A dual-stage feature selection mechanism (BP-MIV/RF-RFECV) is developed to identify 12 critical predictors from multi-modal sensor measurements, effectively resolving dimensional conflicts between static structural parameters and dynamic operational data. The model architecture adopts a dual-layer fusion design: the lower layer captures statistical patterns and temporal–spatial dependencies from asynchronous sensor time-series through Local Cascade Ensemble (LCE) ensemble learning and improved TCN-Transformer networks; the upper layer implements feature fusion using a Stacking framework with logistic regression as the meta-learner. BO is introduced to simultaneously optimize network hyperparameters and feature fusion coefficients. The experimental results demonstrate that the model achieves a prediction accuracy of R² = 0.9292 on an 8-year observation dataset, effectively revealing the non-linear mapping relationship between the Pavement Condition Index (PCI) and multi-source heterogeneous features. The framework demonstrates particular efficacy in correlating high-frequency strain gauge responses with long-term performance degradation, providing mechanistic insights into pavement deterioration processes. This methodology advances infrastructure monitoring through the intelligent synthesis of IoT-enabled sensing systems and empirical inspection data. Full article

(This article belongs to the Section Intelligent Sensors)

► Show Figures

Figure 1

30 pages, 6367 KiB

Open AccessReview

Overview of Research on Digital Image Denoising Methods

by Jing Mao, Lianming Sun, Jie Chen and Shunyuan Yu

Sensors 2025, 25(8), 2615; https://doi.org/10.3390/s25082615 - 20 Apr 2025

Abstract

During image collection, images are often polluted by noise because of imaging conditions and equipment limitations. Images are also disturbed by external noise during compression and transmission, which adversely affects consequent processing, like image segmentation, target recognition, and text detection. A two-dimensional amplitude [...] Read more.

During image collection, images are often polluted by noise because of imaging conditions and equipment limitations. Images are also disturbed by external noise during compression and transmission, which adversely affects consequent processing, like image segmentation, target recognition, and text detection. A two-dimensional amplitude image is one of the most common image categories, which is widely used in people’s daily life and work. Research on this kind of image-denoising algorithm is a hotspot in the field of image denoising. Conventional denoising methods mainly use the nonlocal self-similarity of images and sparser representatives in the converted domain for image denoising. In particular, the three-dimensional block matching filtering (BM3D) algorithm not only effectively removes the image noise but also better retains the detailed information in the image. As artificial intelligence develops, the deep learning-based image-denoising method has become an important research direction. This review provides a general overview and comparison of traditional image-denoising methods and deep neural network-based image-denoising methods. First, the essential framework of classic traditional denoising and deep neural network denoising approaches is presented, and the denoising approaches are classified and summarized. Then, existing denoising methods are compared with quantitative and qualitative analyses on a public denoising dataset. Finally, we point out some potential challenges and directions for future research in the field of image denoising. This review can help researchers clearly understand the differences between various image-denoising algorithms, which not only helps them to choose suitable algorithms or improve and innovate on this basis but also provides research ideas and directions for subsequent research in this field. Full article

(This article belongs to the Special Issue Digital Image Processing and Sensing Technologies—Second Edition)

► Show Figures

Figure 1

17 pages, 1619 KiB

Open AccessArticle

Malicious Traffic Detection Method for Power Monitoring Systems Based on Multi-Model Fusion Stacking Ensemble Learning

by Hao Zhang, Ye Liang, Yuanzhuo Li, Sihan Wang, Huimin Gong, Junkai Zhai and Hua Zhang

Sensors 2025, 25(8), 2614; https://doi.org/10.3390/s25082614 - 20 Apr 2025

Abstract

With the rapid development of the internet, the increasing amount of malicious traffic poses a significant challenge to the network security of critical infrastructures, including power monitoring systems. As the core part of the power grid operation, the network security of power monitoring [...] Read more.

With the rapid development of the internet, the increasing amount of malicious traffic poses a significant challenge to the network security of critical infrastructures, including power monitoring systems. As the core part of the power grid operation, the network security of power monitoring systems directly affects the stability of the power system and the safety of electricity supply. Nowadays, network attacks are complex and diverse, and traditional rule-based detection methods are no longer adequate. With the advancement of machine learning technologies, researchers have introduced them into the field of traffic detection to address this issue. Current malicious traffic detection methods mostly rely on single machine learning models, which face problems such as poor generalization, low detection accuracy, and instability. To solve these issues, this paper proposes a malicious traffic detection method based on multi-model fusion, using the stacking strategy to integrate models. Compared to single models, stacking enhances the model’s generalization and stability, improving detection accuracy. Experimental results show that the accuracy of the stacking model on the NSL-KDD test set is 96.5%, with an F1 score of 96.6% and a false-positive rate of 1.8%, demonstrating a significant improvement over single models and validating the advantages of multi-model fusion in malicious traffic detection. Full article

(This article belongs to the Special Issue IoT Network Security (Second Edition))

► Show Figures

Figure 1

14 pages, 1361 KiB

Open AccessArticle

Multiple Targets CFAR Detection Performance Based on an Intelligent Clustering Algorithm in K-Distribution Sea Clutter

by Mansoor M. Al-dabaa, Eugen Laslo, Ahmed A. Emran, Ahmed Yahya and Ashraf Aboshosha

Sensors 2025, 25(8), 2613; https://doi.org/10.3390/s25082613 - 20 Apr 2025

Abstract

Maintaining a Constant False Alarm Rate (CFAR) in the presence of K-distributed sea clutter is vital due to the dynamic and unpredictable nature of maritime environments. However, conventional CFAR detectors suffer significant performance degradation in multi-target scenarios, primarily due to the masking effect [...] Read more.

Maintaining a Constant False Alarm Rate (CFAR) in the presence of K-distributed sea clutter is vital due to the dynamic and unpredictable nature of maritime environments. However, conventional CFAR detectors suffer significant performance degradation in multi-target scenarios, primarily due to the masking effect caused by interfering targets. To address this challenge, this paper introduces an advanced detection scheme that integrates Linear Density-Based Spatial Clustering for Applications with Noise (Lin-DBSCAN) with CFAR processing. Lin-DBSCAN is specifically tailored to efficiently identify and isolate interfering targets and sea spikes, which typically manifest as outliers in the symmetric reference windows surrounding the Cell Under Test (CUT). By leveraging Lin-DBSCAN, the proposed Lin-DBSCAN-CFAR method effectively filters out anomalous signals from the background clutter, resulting in enhanced detection accuracy and robustness, especially under complex sea clutter conditions. Extensive simulations under varying conditions, including multiple target environments, varying false alarm rates, and different clutter shape parameters, demonstrate that Lin-DBSCAN-CFAR significantly outperforms conventional CFAR approaches. It is noteworthy that the proposed method achieves detection performance comparable to the more computationally intensive DBSCAN-CFAR while significantly reducing computational complexity. Simulation results reveal that Lin-DBSCAN-CFAR requires a 1 to 2 dB lower SNR to reach a detection probability of 0.8 compared with the nearest traditional CFAR techniques, confirming its superiority in both accuracy and efficiency. Full article

(This article belongs to the Topic Innovations in AI and Signal Processing for Advanced Sensing, Radar, RFID, and Communication Systems)

► Show Figures

Figure 1

17 pages, 10288 KiB

Open AccessArticle

An Improved Fast Prediction Method for Full-Space Bistatic Acoustic Scattering of Underwater Vehicles

by Ruichong Gu, Zilong Peng, Yaqiang Xue, Cong Xu and Changxiong Chen

Sensors 2025, 25(8), 2612; https://doi.org/10.3390/s25082612 - 20 Apr 2025

Abstract

This paper presents an improved rapid prediction method for solving the full-space bistatic scattering sound field of underwater vehicles. The scattering sound field is represented as the product of the acoustic scattering transfer function and the sound source density function. By utilizing target [...] Read more.

This paper presents an improved rapid prediction method for solving the full-space bistatic scattering sound field of underwater vehicles. The scattering sound field is represented as the product of the acoustic scattering transfer function and the sound source density function. By utilizing target surface mesh information and partial scattered sound pressure data as known inputs, the method predicts other bistatic scattering sound fields through numerical integration, matrix theory, and the least squares method. To reduce the data input required for predicting the scattering field, the monostatic to bistatic equivalence theorem is incorporated into the algorithm. A comparison with simulation results demonstrates that the proposed approach achieves favorable computational efficiency and reliability. Experimental tests on a double-layered ribbed cylindrical shell further validate the method’s performance. Full article

(This article belongs to the Section Physical Sensors)

► Show Figures

Figure 1

36 pages, 1802 KiB

Open AccessReview

A Review of Vision-Based Multi-Task Perception Research Methods for Autonomous Vehicles

by Hai Wang, Jiayi Li and Haoran Dong

Sensors 2025, 25(8), 2611; https://doi.org/10.3390/s25082611 - 20 Apr 2025

Abstract

Multi-task perception technology for autonomous driving significantly improves the ability of autonomous vehicles to understand complex traffic environments by integrating multiple perception tasks, such as traffic object detection, drivable area segmentation, and lane detection. The collaborative processing of these tasks not only improves [...] Read more.

Multi-task perception technology for autonomous driving significantly improves the ability of autonomous vehicles to understand complex traffic environments by integrating multiple perception tasks, such as traffic object detection, drivable area segmentation, and lane detection. The collaborative processing of these tasks not only improves the overall performance of the perception system but also enhances the robustness and real-time performance of the system. In this paper, we review the research progress in the field of vision-based multi-task perception for autonomous driving and introduce the methods of traffic object detection, drivable area segmentation, and lane detection in detail. Moreover, we discuss the definition, role, and classification of multi-task learning. In addition, we analyze the design of classical network architectures and loss functions for multi-task perception, introduce commonly used datasets and evaluation metrics, and discuss the current challenges and development prospects of multi-task perception. By analyzing these contents, this paper aims to provide a comprehensive reference framework for researchers in the field of autonomous driving and encourage more research work on multi-task perception for autonomous driving. Full article

(This article belongs to the Special Issue Advanced Environment Perception, Decision Planning, and Intelligent Control of Transport Equipment)

► Show Figures

Figure 1

11 pages, 1486 KiB

Open AccessArticle

High Concordance of E-Nose-Derived Breathprints in a Healthy Population: A Cross-Sectional Observational Study

by Silvano Dragonieri, Vitaliano Nicola Quaranta, Andrea Portacci, Teresa Ranieri and Giovanna Elisiana Carpagnano

Sensors 2025, 25(8), 2610; https://doi.org/10.3390/s25082610 - 20 Apr 2025

Abstract

Exhaled breath analysis using electronic noses (e-noses) is a promising non-invasive diagnostic tool. However, a lack of standardized protocols limits clinical implementation. This study evaluates the consistency of breathprints in healthy subjects using the Cyranose 320 e-nose to support standardization efforts. Breath samples [...] Read more.

Exhaled breath analysis using electronic noses (e-noses) is a promising non-invasive diagnostic tool. However, a lack of standardized protocols limits clinical implementation. This study evaluates the consistency of breathprints in healthy subjects using the Cyranose 320 e-nose to support standardization efforts. Breath samples from 139 healthy non-smoking subjects (age range 18–65 years) were collected using a standardized protocol. Participants exhaled into a Tedlar bag for immediate analysis with the Cyranose 320. Principal Component Analysis (PCA) was used to reduce data dimensionality, and K-means clustering grouped subjects based on breathprints. PCA identified four principal components explaining 97.15% of variance. K-means clustering revealed two clusters: 1 outlier and 138 subjects with highly similar breathprints. The median distance from the cluster center was 0.21 (IQR: 0.18–0.24), indicating low variability. Box plots confirmed breathprint consistency across subjects. The high consistency of breathprints in healthy subjects supports the feasibility of standardizing e-nose protocols. These findings highlight the potential of e-noses for clinical diagnostics, warranting further research in diverse populations and disease cohorts. Full article

(This article belongs to the Special Issue Gas Recognition in E-Nose System)

► Show Figures

Figure 1

15 pages, 1928 KiB

Open AccessArticle

High-Precision 2D-DOA Estimation Method for Millimeter-Wave Radar Based on Double-Parallel Linear Array and Joint IAA-RIT

by Danyang Yu, Lei Du, Jie Bai and Yulin Chen

Sensors 2025, 25(8), 2609; https://doi.org/10.3390/s25082609 - 20 Apr 2025

Abstract

High-precision two-dimensional direction of arrival (2D-DOA) estimation is an important mean of millimeter-wave radar for accurate target location. Aiming at the problems such as limited antenna aperture, signal coherence, and a few snapshots in millimeter-wave radar target detection, this paper proposes a 2D-DOA [...] Read more.

High-precision two-dimensional direction of arrival (2D-DOA) estimation is an important mean of millimeter-wave radar for accurate target location. Aiming at the problems such as limited antenna aperture, signal coherence, and a few snapshots in millimeter-wave radar target detection, this paper proposes a 2D-DOA estimation method by using a joint iterative adaptive approach and rotational invariance technique (IAA-RIT) based on the double-parallel linear array. This method first constructs an iterative adaptive approach spectrum based on subarray 1 in the double-parallel linear array and then calculates the coupling angle estimate with the azimuth and the elevation. Secondly, based on the rotational invariance relationship between the two subarrays, the extended covariance matrices are respectively constructed, and the spatial smoothing technique is employed to decorrelate the signals. Then, the signal direction matrix is reconstructed based on the coupling angle estimate, and the rotational invariance relationship between the two subarrays is calculated to obtain another set of coupling angle estimates. Finally, the azimuth and the elevation are decoupled based on two sets of estimated coupling angles and the spatial geometry relation. Our experimental results show that IAA-RIT can estimate the coherent signal with high-precision 2D-DOA with a few snapshots and no additional angle matching. Full article

(This article belongs to the Special Issue Innovative Applications of mmWave Sensors)

► Show Figures

Figure 1

22 pages, 9294 KiB

Open AccessArticle

Deep Layered Network Based on Rotation Operation and Residual Transform for Building Segmentation from Remote Sensing Images

by Shuzhe Zhang, Taoyi Chen, Fei Su, Hao Xu, Yan Li and Yaohui Liu

Sensors 2025, 25(8), 2608; https://doi.org/10.3390/s25082608 - 20 Apr 2025

Abstract

Deep learning has been widely applied in building segmentation from high-resolution remote sensing (HRS) images. However, HRS images suffer from insufficient complementary representation of target points in terms of capturing details and global information. To this end, we propose a novel building segmentation [...] Read more.

Deep learning has been widely applied in building segmentation from high-resolution remote sensing (HRS) images. However, HRS images suffer from insufficient complementary representation of target points in terms of capturing details and global information. To this end, we propose a novel building segmentation model for HRS images, termed C_ASegformer. Specifically, we design a Deep Layered Enhanced Fusion (DLEF) module to integrate hierarchical information from different receptive fields, thereby enhancing the feature representation capability of HRS information from global to detailed levels. Additionally, we introduce a Triplet Attention (TA) Module, which establishes dependency relationships between buildings and the environment through multi-directional rotation operations and residual transformations. Furthermore, we propose a Multi-Level Dilated Connection (MDC) Module to efficiently capture contextual relationships across different scales at a low computational cost. We conduct comparative experiments with several state-of-the-art models on three datasets, including the Massachusetts dataset, the INRIA dataset, and the WHU dataset. On the Massachusetts dataset, C_ASegformer achieves 95.42%, 85.69%, and 75.46% for OA, F1score, and mIoU, respectively. C_ASegformer shows more accurate performance, demonstrating the validity and sophistication of the model. Full article

(This article belongs to the Special Issue Computer Vision and Pattern Recognition Based on Remote Sensing)

► Show Figures

Figure 1

19 pages, 6097 KiB

Open AccessArticle

Forced Oscillation Detection via a Hybrid Network of a Spiking Recurrent Neural Network and LSTM

by Xiaomei Yang, Jinfei Wang, Xingrui Huang, Yang Wang and Xianyong Xiao

Sensors 2025, 25(8), 2607; https://doi.org/10.3390/s25082607 - 20 Apr 2025

Abstract

The detection of forced oscillations, especially distinguishing them from natural oscillations, has emerged as a major concern in power system stability monitoring. Deep learning (DL) holds significant potential for detecting forced oscillations correctly. However, existing artificial neural networks (ANNs) face challenges when employed [...] Read more.

The detection of forced oscillations, especially distinguishing them from natural oscillations, has emerged as a major concern in power system stability monitoring. Deep learning (DL) holds significant potential for detecting forced oscillations correctly. However, existing artificial neural networks (ANNs) face challenges when employed in edge devices for timely detection due to their inherent complex computations and high power consumption. This paper proposes a novel hybrid network that integrates a spiking recurrent neural network (SRNN) with long short-term memory (LSTM). The SRNN achieves computational and energy efficiency, while the integration with LSTM is conducive to effectively capturing temporal dependencies in time-series oscillation data. The proposed hybrid network is trained using the backpropagation-through-time (BPTT) optimization algorithm, with adjustments made to address the discontinuous gradient in the SRNN. We evaluate our proposed model on both simulated and real-world oscillation datasets. Overall, the experimental results demonstrate that the proposed model can achieve higher accuracy and superior performance in distinguishing forced oscillations from natural oscillations, even in the presence of strong noise, compared to pure LSTM and other SRNN-related models. Full article

(This article belongs to the Special Issue Diagnosis and Risk Analysis of Electrical Systems)

► Show Figures

Figure 1

18 pages, 4983 KiB

Open AccessArticle

Small Defects Detection of Galvanized Strip Steel via Schatten-p Norm-Based Low-Rank Tensor Decomposition

by Shiyang Zhou, Xuguo Yan, Huaiguang Liu and Caiyun Gong

Sensors 2025, 25(8), 2606; https://doi.org/10.3390/s25082606 - 20 Apr 2025

Abstract

Accurate and efficient white-spot defects detection for the surface of galvanized strip steel is one of the most important guarantees for the quality of steel production. It is a fundamental but “hard” small target detection problem due to its small pixel occupation in [...] Read more.

Accurate and efficient white-spot defects detection for the surface of galvanized strip steel is one of the most important guarantees for the quality of steel production. It is a fundamental but “hard” small target detection problem due to its small pixel occupation in low-contrast images. By fully exploiting the low-rank and sparse prior information of a surface defect image, a Schatten-p norm-based low-rank tensor decomposition (SLRTD) method is proposed to decompose the defect image into low-rank background, sparse defect, and random noise. Firstly, the original defect images are transformed into a new patch-based tensor mode through data reconstruction for mining valuable information of the defect image. Then, considering the over-shrinkage problem in the low-rank component estimation caused by a vanilla nuclear norm and a weighted nuclear norm, a nonlinear reweighting strategy based on a Schatten

p

-norm is incorporated to improve the decomposition performance. Finally, a solution framework is proposed via a well-designed alternating direction method of multipliers to obtain the white-spot defect target image by a simple segmenting algorithm. The white-spot defect dataset from a real-world galvanized strip steel production line is constructed, and the experimental results demonstrate that the proposed SLRTD method outperforms existing state-of-the-art methods qualitatively and quantitatively. Full article

(This article belongs to the Special Issue Sensing and Imaging for Defect Detection: 2nd Edition)

► Show Figures

Figure 1

15 pages, 608 KiB

Open AccessReview

Revisiting Wireless Cyberattacks on Vehicles

by Roberto Gesteira-Miñarro, Gregorio López and Rafael Palacios

Sensors 2025, 25(8), 2605; https://doi.org/10.3390/s25082605 - 20 Apr 2025

Abstract

The automotive industry has been a prime target for cybercriminals for decades, with attacks becoming more sophisticated as vehicles integrate advanced digital technologies. In response, new standards and regulations have been introduced, requiring manufacturers to implement robust cybersecurity measures to obtain necessary certifications. [...] Read more.

The automotive industry has been a prime target for cybercriminals for decades, with attacks becoming more sophisticated as vehicles integrate advanced digital technologies. In response, new standards and regulations have been introduced, requiring manufacturers to implement robust cybersecurity measures to obtain necessary certifications. Modern vehicles have an extensive attack surface due to the increasing number of interconnected electronic components and wireless communication features. While new technologies improve connectivity, automation, and comfort, they also introduce new vulnerabilities that can be exploited by attackers. This paper presents a comprehensive analysis of the attack surface of modern vehicles, focusing on the security risks associated with wireless communication technologies. Each technology is examined in detail, highlighting existing research, known vulnerabilities, and potential countermeasures. Furthermore, this study identifies key research gaps in the field, providing insights into critical areas that require further investigation. This work aims to guide future research efforts in order to enhance vehicle cybersecurity in the evolving landscape of smart, autonomous, and connected vehicles. Full article

(This article belongs to the Section Vehicular Sensing)

► Show Figures

Figure 1

18 pages, 10309 KiB

Open AccessArticle

Accurate Conversion of Land Surface Reflectance for Drone-Based Multispectral Remote Sensing Images Using a Solar Radiation Component Separation Approach

by Huasheng Sun, Lei Guo and Yuan Zhang

Sensors 2025, 25(8), 2604; https://doi.org/10.3390/s25082604 - 20 Apr 2025

Abstract

Land surface reflectance is a basic physical parameter in many quantitative remote sensing models. However, the existing reflectance conversion techniques for drone-based (or UAV-based) remote sensing need further improvement and optimization due to either cumbersome operational procedures or inaccurate results. To tackle this [...] Read more.

Land surface reflectance is a basic physical parameter in many quantitative remote sensing models. However, the existing reflectance conversion techniques for drone-based (or UAV-based) remote sensing need further improvement and optimization due to either cumbersome operational procedures or inaccurate results. To tackle this problem, this study proposes a novel method to mathematically implement the separation of direct and scattering radiation using a self-developed multi-angle light intensity device. The verification results from practical experiments demonstrate that the proposed method has strong adaptability, as it can obtain accurate surface reflectance even under complicated conditions where both illumination intensity and component change simultaneously. Among the six selected typical land cover types (i.e., lake water, slab stone, shrub, green grass, red grass, and dry grass), green grass has the highest error among the five multispectral bands with a mean absolute error (MAE) of 1.59%. For all land cover types, the highest MAE of 1.01% is found in the red band. The above validation results indicate that the proposed land surface reflectance conversion method has considerably high accuracy. Therefore, the study results may provide valuable references for quantitative remote sensing applications of drone-based multispectral data, as well as the design of future multispectral drones. Full article

(This article belongs to the Special Issue Hyperspectral/Multispectral Sensing Technologies for Spectral Cameras and Image Sensors)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Sensors, Volume 25, Issue 8 (April-2 2025) – 292 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI