Next Issue
Volume 26, September-1
Previous Issue
Volume 26, August-1
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
9.4
4.0
Journals
Sensors
Volume 26
Issue 16
sensors-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Sensors, Volume 26, Issue 16 (August-2 2026) – 310 articles

Cover Story (view full-size image): Carbon nanotube (CNT) network density plays a critical role in determining the electrical and sensing performance of electrolyte-gated carbon nanotubes field-effect transistors (CNT-FETs). Here we investigate how CNT density influences charge transport, device variability, electrostatic gating, and pH sensing. Sparse CNT networks provide enhanced pH sensitivity but exhibit greater device-to-device variability. Increasing CNT density improves electrical conductivity, device yield, and reproducibility, while excessive density reduces pH sensing sensitivity due to electrostatic screening. These results reveal a fundamental trade-off between sensitivity and device robustness, and provide a practical framework for optimizing CNT-FET architectures for scalable chemical and biosensing applications. View this paper
  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
31 pages, 824 KB  
Article
Recovery-Limiting Signal-Conditioned Action Decoding for Emergency Wireless Access–Backhaul Capacity Allocation
by Jingxiang Ma, Minglei Han, Hongbin Ma, Ying Jin and Youzhi Zhang
Sensors 2026, 26(16), 5317; https://doi.org/10.3390/s26165317 - 21 Aug 2026
Viewed by 491
Abstract
Post-disaster emergency wireless access–backhaul recovery requires limited capacity to be allocated across affected regions under nonnegativity and fixed-total-capacity constraints. We propose Recovery-Limiting Signal-Conditioned Action Decoding (RLS-CAD). Critical-region access outage, backhaul shortfall, traffic backlog, and end-to-end recovery deficit form regional score bases, while a [...] Read more.
Post-disaster emergency wireless access–backhaul recovery requires limited capacity to be allocated across affected regions under nonnegativity and fixed-total-capacity constraints. We propose Recovery-Limiting Signal-Conditioned Action Decoding (RLS-CAD). Critical-region access outage, backhaul shortfall, traffic backlog, and end-to-end recovery deficit form regional score bases, while a 15-dimensional policy controls signal fusion, score correction, and finite-support simplex projection. Experiments use five training seeds and 100 shared test scenarios and retain both training and scenario variability. RLS-CAD outperforms Graph-PPO and both information-matched controls on FinalCap, CTI, and TaskUtility. The evaluation concerns normalized planning-level simulation; physical-layer scheduling is outside its scope. Full article
(This article belongs to the Section Intelligent Sensors)
►▼ Show Figures

Figure 1

57 pages, 1940 KB  
Review
From Modality Performance to Graceful Degradation: A PRISMA 2020 Systematic Review of Sensor Architectures for Autonomous Vehicles
by Patrik Viktor
Sensors 2026, 26(16), 5316; https://doi.org/10.3390/s26165316 - 21 Aug 2026
Viewed by 536
Abstract
Autonomous vehicles depend on heterogeneous sensing systems whose performance varies with range, illumination, weather, object material, traffic geometry, contamination, calibration quality, and cyber-physical interference. This PRISMA 2020 and PRISMA-S systematic review synthesized 65 peer-reviewed primary studies selected from 2143 records identified through four [...] Read more.
Autonomous vehicles depend on heterogeneous sensing systems whose performance varies with range, illumination, weather, object material, traffic geometry, contamination, calibration quality, and cyber-physical interference. This PRISMA 2020 and PRISMA-S systematic review synthesized 65 peer-reviewed primary studies selected from 2143 records identified through four databases. After removal of 793 records before screening, 1350 titles and abstracts were screened; 273 full texts were assessed and 208 were excluded with documented reasons. The final evidence base covers cameras, LiDAR, radar, thermal and event cameras, GNSS/IMU localization, calibration, synchronization, multimodal fusion, adverse-weather perception, sensor-health monitoring, and fault-tolerant perception. No modality was universally superior: comparative performance depended on hardware generation, dataset, environmental severity, range, and metric. Direct evidence was strongest for component-level perception and controlled degradation, whereas health-conditioned fusion, ODD restriction, and minimum-risk behavior were supported mainly by partial experimental evidence and safety-oriented synthesis. The review therefore proposes, rather than claims to validate, a reliability-aware architecture that separates sensor health from task confidence, preserves uncertainty and provenance, adapts fusion, and constrains operation when residual evidence is insufficient. The review was retrospectively registered in PROSPERO on 30 July 2026 (CRD420261465869). Full article
(This article belongs to the Special Issue Artificial Intelligence and Sensor-Based Learning for Intelligent Driving and Smart Mobility)
►▼ Show Figures

Figure 1

28 pages, 2126 KB  
Article
Design and Evaluation of an Edge AI-Enabled Low-Power Magnetic Sensor for Real-Time Road Traffic Monitoring
by Michal Hodoň, Peter Šarafín, Lukáš Formanek and Andrea Kociánová
Sensors 2026, 26(16), 5315; https://doi.org/10.3390/s26165315 - 21 Aug 2026
Viewed by 360
Abstract
Road traffic surveys require sensing systems that can be deployed rapidly without modifying the road surface or requiring a permanent power connection. This paper presents the design, embedded implementation, and evaluation of a low-power roadside magnetic sensor that performs vehicle-event detection and classification [...] Read more.
Road traffic surveys require sensing systems that can be deployed rapidly without modifying the road surface or requiring a permanent power connection. This paper presents the design, embedded implementation, and evaluation of a low-power roadside magnetic sensor that performs vehicle-event detection and classification directly at the edge. The sensing node integrates two RM3100 three-axis magnetometers (PNI Sensor, Santa Rosa, CA, USA) with an NXP MK22FN512VLH12 microcontroller (NXP Semiconductors N.V., Eindhoven, The Netherlands) based on a 120 MHz Arm Cortex-M4F core with 512 kB Flash and 128 kB SRAM. Magnetic-field data are acquired at 250 Hz and processed locally using baseline removal, low-pass filtering, signal-energy calculation, and peak-based event detection. Detected magnetic signatures are classified using an integer-quantised one-dimensional convolutional neural network implemented directly on the microcontroller. The model processes four synchronised 512-sample channels representing the three magnetic-field axes and their combined signal energy. Model development was supported by approximately 50,000 annotated events obtained from 36 h of real-world traffic measurements at eight locations. The selected model achieved an overall classification accuracy of 91.1% for the considered operational categories. The implemented network requires 288,128 multiply–accumulate operations per inference, while its quantised weights and biases occupy approximately 23 kB of Flash memory. Complete three-axis event signatures are stored locally for subsequent verification, whereas only the timestamp and predicted vehicle category are transmitted through the wireless interface. Based on the capacity of the applied LiFePO4 battery and the estimated consumption of the implemented hardware, the expected autonomous operating period is approximately 41 days. The results demonstrate the feasibility of integrating magnetic sensing, embedded signal processing, and Edge AI on a conventional resource-constrained Cortex-M4 platform for non-invasive road traffic monitoring. Full article
(This article belongs to the Special Issue Recent Trends and Advances in Magnetic Sensors)
►▼ Show Figures

Figure 1

22 pages, 8970 KB  
Article
Lower Limb Motion Classification of Actions in Confined Environments Based on Multi-Source Signal Fusion and Muscle Symmetry Features
by Dingzhe Li, Xiaorong Guan, Zheng Wang, Changlong Jiang, Long He, Xiwang Mao and Qiang Zhou
Sensors 2026, 26(16), 5314; https://doi.org/10.3390/s26165314 - 21 Aug 2026
Viewed by 369
Abstract
In recent years, advances in exoskeleton technology have increased the demand for human motion classification in terms of both movement diversity and recognition accuracy, making the effective classification of more complex asymmetric movements increasingly important. This study integrated surface electromyography (sEMG) and inertial [...] Read more.
In recent years, advances in exoskeleton technology have increased the demand for human motion classification in terms of both movement diversity and recognition accuracy, making the effective classification of more complex asymmetric movements increasingly important. This study integrated surface electromyography (sEMG) and inertial measurement unit (IMU) signals and employed mutual information (MI) and muscle symmetry features (MSF) to analyze the characteristic differences between symmetrical muscles in both legs during asymmetric movements, with the aim of improving the accuracy of asymmetric motion classification. sEMG and IMU signals were collected from six movements, including three asymmetric postures: asymmetrical stance, single-knee kneeled position, and crouching advance. The acquired signals were processed through energy envelope analysis, active segment extraction, empirical mode decomposition (EMD), feature extraction, MI extraction, and MSF extraction. The relevance based on weight feature selection (RWFS) combined with conditional mutual information (CMI) method was then applied to reduce feature dimensionality, prioritize features with significant fluctuations, and preserve key characteristics. Finally, the CNN-LSTM-Attention algorithm was used for classification. Experimental results showed that fusing sEMG and IMU signals achieved 96.30% accuracy in lower limb motion recognition. The proposed method improves asymmetric movement classification and may provide a potential basis for exoskeleton motion classification in special environments. Full article
(This article belongs to the Special Issue Challenges and Future Trends in Biomedical Signal Processing)
►▼ Show Figures

Figure 1

17 pages, 37285 KB  
Article
Improvement of Initial Azimuth Estimation Time for a North-Finding System Using Low-Cost MEMS Sensors and a Compact 3-Axis Turntable in Challenging Environments
by Taisei Hayashi and Daisuke Terada
Sensors 2026, 26(16), 5313; https://doi.org/10.3390/s26165313 - 21 Aug 2026
Viewed by 293
Abstract
This paper proposes a method for reducing the initial azimuth estimation time of a north-finding system employing low-cost sensors and a compact 3-axis turntable. The system is capable of operating in non-horizontal environments, magnetically disturbed environments, and environments where Global Navigation Satellite System [...] Read more.
This paper proposes a method for reducing the initial azimuth estimation time of a north-finding system employing low-cost sensors and a compact 3-axis turntable. The system is capable of operating in non-horizontal environments, magnetically disturbed environments, and environments where Global Navigation Satellite System (GNSS) signals are unavailable. Detection of due north without prior azimuth information was evaluated through indoor experiments under the aforementioned conditions. During each rotation, the compact 3-axis turntable was kept horizontal and the acceleration and angular velocity were measured in 16 directions at 22.5° intervals. By including the final position coinciding with the initial one, a total of 17 measurement points were obtained per lap. This process was repeated for 77 laps. For statistical evaluation, 5000 bootstrap replications were generated. Detection of due north was then performed using these datasets and the relationship between the number of laps and the estimation error was statistically analyzed. Consequently, it was confirmed that the root mean square (RMS) error becomes less than 1° after ten laps, corresponding to a data acquisition time of approximately 1.3 h. Compared to our previous study, the required estimation time is reduced by approximately 2 h. Full article
(This article belongs to the Special Issue Multi-Sensor Technology for Tracking, Positioning and Navigation)
►▼ Show Figures

Graphical abstract

26 pages, 1733 KB  
Review
Microwave Sensors for Dielectric Characterization: Planar Architectures, Extraction Methods, and Emerging Applications
by Feifei Tan and Changjun Liu
Sensors 2026, 26(16), 5312; https://doi.org/10.3390/s26165312 - 21 Aug 2026
Viewed by 324
Abstract
Microwave dielectric characterization is essential for material evaluation, process monitoring, biomedical sensing, and nondestructive testing. This review critically evaluates planar microwave sensors for dielectric characterization, with the core scope restricted to printed microstrip and coplanar-waveguide structures, SRR/CSRR and DGS configurations, substrate-integrated waveguides, interferometric [...] Read more.
Microwave dielectric characterization is essential for material evaluation, process monitoring, biomedical sensing, and nondestructive testing. This review critically evaluates planar microwave sensors for dielectric characterization, with the core scope restricted to printed microstrip and coplanar-waveguide structures, SRR/CSRR and DGS configurations, substrate-integrated waveguides, interferometric sensors, and microfluidic platforms. Adjacent non-planar or system-level techniques are included only when they provide transferable lessons in calibration, inversion, or deployment. Unlike earlier surveys that primarily catalog devices or extraction methods, the literature is organized here through a design-decision hierarchy linking architecture, operating principle, readout mechanism, sample interface, and application. Representative approaches are compared not only by frequency, sensitivity, Q-factor, and sample volume, but also by calibration burden, fabrication tolerance, environmental robustness, cost, and scalability. Particular attention is given to uncertainty sources in practical measurement chains, FR-4 and PCB manufacturing variability, long-term drift and sensor aging, and the application-specific limitations of machine-learning-assisted inversion. The resulting synthesis provides design-oriented guidance for selecting and translating planar microwave sensors into reliable industrial, biomedical, and microwave-processing measurement systems. Full article
(This article belongs to the Special Issue Advances in Microwave and Millimeter-Wave Sensing)
►▼ Show Figures

Figure 1

20 pages, 4904 KB  
Article
A Neuro-Inspired Rate-Encoded Descriptor for High-Speed Asynchronous Robotic Vision
by Shane Harrigan, Sonya Coleman, Dermot Kerr, Pratheepan Yogarajah, Chengdong Wu and Zheng Fang
Sensors 2026, 26(16), 5311; https://doi.org/10.3390/s26165311 - 21 Aug 2026
Viewed by 283
Abstract
This paper presents the Post-Stimulus Time-Dependent Event Descriptor (P-TED), a novel “pure event” feature descriptor designed for neuromorphic vision data. Unlike conventional frame-based approaches or hybrid methods that transform event data into intermediate representations, P-TED operates directly on asynchronous event streams, thereby preserving [...] Read more.
This paper presents the Post-Stimulus Time-Dependent Event Descriptor (P-TED), a novel “pure event” feature descriptor designed for neuromorphic vision data. Unlike conventional frame-based approaches or hybrid methods that transform event data into intermediate representations, P-TED operates directly on asynchronous event streams, thereby preserving the intrinsic low-latency and high-temporal-resolution advantages of event-based sensors. The descriptor integrates two complementary feature sets: a motion feature vector, which aggregates spatial relationships within a Moore neighbourhood to quantify stimulus direction, and a pattern feature vector, which employs rate encoding to capture temporal excitation signatures. The efficacy of the P-TED framework is validated through three distinct experiments: object and character recognition (MNIST-DVS and CIFAR10-DVS), mobile robot movement analysis, and complex non-rigid robotic hand gesture recognition (RoShamBo). Experimental results demonstrate that the P-TED achieves a significant reduction in classification latency, requiring only 2.7 ms compared to the 10.3 ms recorded by the state-of-the-art Distribution-Aware Retinal Transform (DART) framework. Additionally, P-TED exhibits superior robustness in disambiguating symmetric and mirrored motions, as well as in maintaining stability under non-linear fluctuations in event density caused by changing scale. This work establishes P-TED as a high-speed, computationally efficient, and explainable solution for real-time neuromorphic robotic vision systems. Full article
(This article belongs to the Special Issue Event-Based Vision and Multimodal Sensor Fusion)
►▼ Show Figures

Figure 1

24 pages, 4913 KB  
Article
Privacy-Preserving Head Pose Estimation System for Measuring Cervical Range of Motion
by Zhuofu Liu, Lichao Zhang, Gaohan Li and Peter W. McCarthy
Sensors 2026, 26(16), 5310; https://doi.org/10.3390/s26165310 - 21 Aug 2026
Viewed by 416
Abstract
Cervical range of motion (CROM) has been used in research and clinically for assessing cervical health. Gold-standard goniometers tend to be cumbersome. However, Inertial Measurement Units (IMUs) or vision-based alternatives demand frequent calibration and/or costly hardware; moreover, the subject is aware of being [...] Read more.
Cervical range of motion (CROM) has been used in research and clinically for assessing cervical health. Gold-standard goniometers tend to be cumbersome. However, Inertial Measurement Units (IMUs) or vision-based alternatives demand frequent calibration and/or costly hardware; moreover, the subject is aware of being measured and there is a risk of breaching privacy. In response, we have developed a non-contact HPNet system for head pose estimation (HPE) that can use a rear-facing camera to quantify CROM accurately. A Re-parameterized Visual Geometry Group (RepVGG)-D2se model is employed as the backbone of the network, and a Spatial Feature Enhancement (SCFE) module is incorporated to improve feature extraction. HPNet was evaluated on the large-scale Carnegie Mellon University (CMU) Panoptic dataset, achieving a mean absolute error (MAE) of 3.48°, 3.22°and 3.34° for yaw, pitch and roll respectively. Inter-instrument reliability was excellent for all six cervical movements when compared with the research/clinical-grade CROM device, with intraclass correlation coefficients (ICCs) averaging 0.939. Bland–Altman plots confirmed close agreement between the two methods. Cervical movement trajectory curves further confirmed the concordance between the clinical device and our method. The system is fully automatic, requires only a rear-facing camera, effectively preserves patient privacy, and provides accurate cervical posture estimation. This technology may provide a basis for future applications in neck-disorder screening, remote health monitoring, and personalized musculoskeletal wellness management, although further task-specific clinical validation will be required. To date, HPNet has been validated primarily on a computer-based platform and has not yet been deployed on smartphones. Future work will focus on model lightweighting, mobile deployment, and cross-device adaptation to facilitate its practical implementation on mobile devices. Full article
(This article belongs to the Special Issue Intelligent Sensing Technologies to Facilitate Clinical/Medical Decision Making)
►▼ Show Figures

Figure 1

26 pages, 9400 KB  
Article
Around the Clock: Wearing a Flex-Printed trEEGrid Electrode Patch and Miniaturized Amplifier for Nearly 24 Hours Across Daytime and Overnight EEG Sessions
by Joanna E. M. Scanlon, Axel H. Winneke, Wiebke Pätzold and Karen Insa Wolf
Sensors 2026, 26(16), 5309; https://doi.org/10.3390/s26165309 - 21 Aug 2026
Viewed by 483
Abstract
Long-term EEG measurements (over eight hours) can allow a deeper understanding of everyday life brain functioning. However, most EEG systems can only be used for short periods in controlled settings, due to limitations in comfort, signal quality of EEG electrode sensors and amplifier [...] Read more.
Long-term EEG measurements (over eight hours) can allow a deeper understanding of everyday life brain functioning. However, most EEG systems can only be used for short periods in controlled settings, due to limitations in comfort, signal quality of EEG electrode sensors and amplifier size. In this study, we measured brain activity during day and night on ten lab member participants using our new flex-printed trEEGrid patch electrodes and a miniaturized and modularized amplifier prototype. The patch was left on participants up to 24 h while they went about their day. EEG was recorded during several exploratory tasks and overnight. Daytime scenarios included auditory and visual oddball tasks, as well as a resting-state measurement and sudoku cognitive load task. Impedances were recorded throughout all tasks. Ag/AgCl ‘ring’ electrodes were used as comparison. Daytime tasks were performed twice (i.e., once in the afternoon and again the next morning), to assess signal quality changes over time. P3 oddball, resting state and workload-related spectral effects were observed on both days. Auditory N1 amplitudes were similar between the patch and ring electrodes. The system shows promising data quality over the nearly 24 h wearing time and allows new possibilities for daytime and sleep EEG studies. Full article
(This article belongs to the Special Issue Biomedical Electronics and Wearable Systems—2nd Edition)
►▼ Show Figures

Figure 1

16 pages, 16824 KB  
Article
An Ultrasensitive Electrochemical Biosensor for Nucleic Acid Detection Based on Silver Nanoflower-Stem-Loop Probes
by Yingying Yuan, Xiaoyu Lei, Fengyu Li, Yuchen Su, Bo Liu, Lei Luo and Hangyu Zhang
Sensors 2026, 26(16), 5308; https://doi.org/10.3390/s26165308 - 21 Aug 2026
Viewed by 292
Abstract
Nucleic acids are critical biomarkers that provide essential information throughout disease progression, making their detection critical to early diagnosis of both infectious and non-infectious diseases. However, existing detection methods, including classical analytical techniques and even most reported biosensors, are often constrained by complex [...] Read more.
Nucleic acids are critical biomarkers that provide essential information throughout disease progression, making their detection critical to early diagnosis of both infectious and non-infectious diseases. However, existing detection methods, including classical analytical techniques and even most reported biosensors, are often constrained by complex procedures, high costs, and limited sensitivity, with the majority operating at the femtomolar level and failing to achieve single-molecule detection needed for early-stage diagnosis. Here, we report an electrochemical biosensor based on silver nanoflowers (AgNFs) integrated with stem-loop probes (SPs) for universal nucleic acid detection, using Norovirus RNA as a model target to validate the platform. The SPs serve as critical elements in a signal amplification system, converting target binding into a biotin–streptavidin recognition event, which leads to the accumulation of AgNFs-SP complexes on laser-induced graphene (LIG) electrodes and generates a strong electrochemical signal. Under optimized conditions with a 50 min hybridization incubation (total assay time ~60 min), the sensor exhibits a linear response to Norovirus RNA concentrations from 1 aM to 10 fM, with a measured detection limit of 1 aM, achieving single-molecule-level detection capability. For applications requiring faster turnaround, a 20 min hybridization incubation (~30 min total assay time) shifts the linear range to 0.1 fM–1 pM with a measured detection limit of 0.1 fM, offering more rapid quantification when the maximum sensitivity is not required. The proposed biosensor is cost-effective, amenable to miniaturization, and designed as a versatile platform adaptable to other nucleic acid targets by simply modifying the probe sequence, showing broad potential for early diagnosis of various diseases. Full article
(This article belongs to the Section Biosensors)
►▼ Show Figures

Figure 1

27 pages, 8740 KB  
Article
Research on the High-Precision Position Datum Long Baseline Coordinate Transfer Strategy Using BDS-3 High- and Low-Frequency Signals
by Mingduan Zhou, Wenxuan Zhang, Haodong Cai, Zichun Wang, Shuzhan Xia, Qiao Song, Shiqi Lin and Lu Qin
Sensors 2026, 26(16), 5307; https://doi.org/10.3390/s26165307 - 21 Aug 2026
Viewed by 376
Abstract
High-precision long-baseline coordinate transfer is essential for maintaining spatial reference frames, and the modernized multi-frequency signals of BDS-3 provide new opportunities for this task. However, existing long-baseline network solutions still rely mainly on legacy frequency combinations, and quantitative evidence for pure new-frequency BDS-3 [...] Read more.
High-precision long-baseline coordinate transfer is essential for maintaining spatial reference frames, and the modernized multi-frequency signals of BDS-3 provide new opportunities for this task. However, existing long-baseline network solutions still rely mainly on legacy frequency combinations, and quantitative evidence for pure new-frequency BDS-3 combinations in large-scale coordinate transfer remains limited. This study evaluates the applicability of BDS-3 high- and low-frequency signal combinations for long-baseline position datum transfer and investigates frequency-combination selection. Seven continuous stations were used to form 21 long baselines. Five dual-frequency schemes were tested, including four BDS-3 combinations, namely B1I/B3I, B1I/B2a, B1C/B2a, and B1C/B3I, and one GPS reference combination, L1/L5. Double-differenced ionosphere-free baseline processing and three-dimensional constrained network adjustment were applied. Performance was assessed using carrier-phase precision, normalized root mean square (NRMS), baseline vector quality, and point-transfer differences. The results show that the BDS-3 B1C/B2a new-frequency combination achieved the best overall consistency among the BDS-3 schemes, with an average high-frequency carrier-phase precision of 6.3 mm, a mean NRMS of 0.23, millimeter-level baseline vector formal-error RMS, and a 19.7 mm point difference at the unknown station DCMS. Given that the evaluation is based on seven consecutive days of observations, the long-term applicability of the proposed strategy requires further validation. Full article
(This article belongs to the Special Issue Advances in GNSS Signal Processing and Navigation—Third Edition)
►▼ Show Figures

Figure 1

18 pages, 3935 KB  
Article
Lightweight Monocular Depth Estimation with Local Feature Enhancement Modules and Guided Data Augmentation
by Jae-young Lee and Soon-kak Kwon
Sensors 2026, 26(16), 5306; https://doi.org/10.3390/s26165306 - 21 Aug 2026
Viewed by 291
Abstract
Although research on lightweight monocular depth estimation models has been actively conducted, achieving real-time deployment on edge devices remains challenging due to limited computational resources and memory capacity. To address these limitations, we propose a lightweight model for self-supervised monocular depth estimation. We [...] Read more.
Although research on lightweight monocular depth estimation models has been actively conducted, achieving real-time deployment on edge devices remains challenging due to limited computational resources and memory capacity. To address these limitations, we propose a lightweight model for self-supervised monocular depth estimation. We cut the iterations of each feature-extracting block nearly in half, while our proposed Asymmetric Dilated Convolution module and a StarNext module compensate for the reduced model capacity. Specifically, the Asymmetric Dilated Convolution module captures horizontal and vertical structural features through asymmetric kernels, and the StarNext module fuses multi-scale features via element-wise multiplication. In model training, random cropping and scaling are applied for inducing the model to focus on localized object features. Additionally, we introduce a Disparity-guided Cutout technique based on the pre-inferred disparity map to randomly mask adjacent pixels. Simulation results on the KITTI dataset demonstrate that the proposed model reduces the number of parameters and GFLOPs by approximately 50% and 58%, respectively, without significant degradation in depth estimation accuracy compared to the baseline Lite-Mono. Furthermore, inference benchmarks on the Jetson Orin Nano platform demonstrate speedups of approximately 46.0%, 46.1%, 46.3%, and 52.0% across the MaxN, 25 W, 15 W, and 7 W power modes, respectively. Full article
(This article belongs to the Special Issue Sensor-Based Image Processing and Sensing Techniques for Enhanced Object Detection)
►▼ Show Figures

Figure 1

36 pages, 11654 KB  
Article
Interpretable Graph–Temporal–Spectral Fusion for Precursor-Related Anomaly Detection in Underground Mine Sensor Networks
by Shuren Mao, Yunpei Liang and Quangui Li
Sensors 2026, 26(16), 5305; https://doi.org/10.3390/s26165305 - 21 Aug 2026
Viewed by 334
Abstract
Underground coal mine safety monitoring relies on multi-source sensor networks, but abnormal-state detection remains challenging because methane, airflow, dust, and equipment-operation signals are non-stationary, heterogeneous, and constrained by ventilation and mining disturbances. This study proposes GasNet, an interpretable engineering-prior graph–temporal–spectral framework for precursor-related [...] Read more.
Underground coal mine safety monitoring relies on multi-source sensor networks, but abnormal-state detection remains challenging because methane, airflow, dust, and equipment-operation signals are non-stationary, heterogeneous, and constrained by ventilation and mining disturbances. This study proposes GasNet, an interpretable engineering-prior graph–temporal–spectral framework for precursor-related anomaly detection. Variables are organized into methane-related core sensors and environmental–operational modulation sensors. A directed sensor graph is constructed using ventilation causality, sensor deployment, and shearer-coupling relationships. GasNet then integrates a graph convolutional network for spatial–topological modeling, TimesNet for temporal–spectral pattern extraction, and cross-attention for adaptive feature fusion. An unsupervised reconstruction strategy identifies intervals deviating from learned normal production patterns. Field validation was conducted on the 31002 fully mechanized working face of the Xinyuan Coal Mine, where eight precursor-related abnormal intervals were annotated from monitoring data and field records. GasNet achieved a Precision of 0.881, a Recall of 1.000, an F1-score of 0.937, a false-alarm rate of 0.0017, and zero missed detections, with the highest F1-score among seven time-series baselines. Interpretability analysis further provided feature-fusion and sensor-time evidence for warning review. These results support the feasibility of GasNet for interpretable anomaly detection in the investigated working face. Full article
(This article belongs to the Section Sensor Networks)
►▼ Show Figures

Figure 1

14 pages, 1233 KB  
Article
Structure-Aware Noise Scheduling for Fixed-View Visual Sensor Reconstruction
by Xingyu Lu and Zengshan Yao
Sensors 2026, 26(16), 5304; https://doi.org/10.3390/s26165304 - 21 Aug 2026
Viewed by 295
Abstract
Fixed-view visual sensors require normal-image reconstruction that preserves structural detail while exposing a local deviation in a residual map. Standard denoising diffusion probabilistic models (DDPMs) use spatially uniform corruption. We study an image-derived, structure-aware noise schedule for reference-image reconstruction, where the input image [...] Read more.
Fixed-view visual sensors require normal-image reconstruction that preserves structural detail while exposing a local deviation in a residual map. Standard denoising diffusion probabilistic models (DDPMs) use spatially uniform corruption. We study an image-derived, structure-aware noise schedule for reference-image reconstruction, where the input image is available and the same importance map can be fixed in the forward and reverse processes. A matched three-seed diagnostic further isolates the effect of the semantic-training/gradient-reverse map substitution and of a lower bound on local noise. Map consistency recovers much of the unconditional FID loss, but neither it nor the attenuation floor yields a universal advantage over DDPM. The six-category MVTec AD residual study is likewise category dependent: the gradient/edge special case improves selected texture-localization outcomes but degrades several object-level outcomes. We therefore present the method as a reconstruction diagnostic, not as a competitive industrial anomaly detector or a universally superior generator. Full article
(This article belongs to the Section Intelligent Sensors)
►▼ Show Figures

Figure 1

29 pages, 11764 KB  
Article
Optimization Scheme for Hybrid RIS-Assisted ISAC System for Controllable Communication and Sensing
by Zhishuo Deng, Bo Li and Hehang Wang
Sensors 2026, 26(16), 5303; https://doi.org/10.3390/s26165303 - 21 Aug 2026
Viewed by 277
Abstract
Integrated sensing and communication (ISAC) systems are envisioned as a key enabler for next-generation wireless networks. To simultaneously achieve multi-function of communication and sensing for this system, a hybrid reconfigurable intelligent surface (RIS) comprising both active and passive reflecting elements is proposed. An [...] Read more.
Integrated sensing and communication (ISAC) systems are envisioned as a key enabler for next-generation wireless networks. To simultaneously achieve multi-function of communication and sensing for this system, a hybrid reconfigurable intelligent surface (RIS) comprising both active and passive reflecting elements is proposed. An index-wise importance score matrix and a factorized representation of complex reflection coefficients are optimized by introducing a communication-sensing controllable coefficient. A novel ISAC system is investigated, which exploits the low-power advantage of passive reflecting elements while retaining the signal amplification capability of active reflecting elements. In the multiple-input multiple-output (MIMO) communication networks, the RIS reflection coefficient matrix is adaptively optimized via a Riemannian Hessian-based method. At the same time, the transmit precoding matrix is obtained using an extended weighted minimum mean square error (WMMSE) and Lagrange multiplier methods. Compared with baseline schemes including active-only RIS, passive-only RIS, random-phase RIS, and non-RIS, the proposed scheme enables multi-mode adjustability of communication and sensing which can be easily transplanted in current ISAC system. Under the constraint of transmit power, it exhibits more robust performance on communication-sensing with varying number of RIS elements and different signal-to-noise ratios (SNRs). Full article
(This article belongs to the Section Communications)
►▼ Show Figures

Figure 1

37 pages, 67321 KB  
Article
Improving Autism Diagnosis Across Ages Using Eye-Tracking and Temporal Transformer Models
by Mohammed A. AlZain, Mahmoud Rokaya, Dalia I. Hemdan, Ibrahim Gad, Malik Almaliki and Elsayed Atlam
Sensors 2026, 26(16), 5302; https://doi.org/10.3390/s26165302 - 21 Aug 2026
Viewed by 347
Abstract
Variation in gaze behavior due to age is currently a considerable challenge in building reliable eye-tracking systems for Autism Spectrum Disorder (ASD) diagnosis. However, existing strategies often focus on static gaze representation or dataset-based information, which can lead to limited generalization of findings [...] Read more.
Variation in gaze behavior due to age is currently a considerable challenge in building reliable eye-tracking systems for Autism Spectrum Disorder (ASD) diagnosis. However, existing strategies often focus on static gaze representation or dataset-based information, which can lead to limited generalization of findings depending on developmental groups and heterogeneous recording conditions. In this paper, we present a temporal transformer-based system for ASD classification using eye-tracking sequences. This allows you to model gaze behavior as a structured temporal process in the context of contextual attention, as well as employing entropy-based modeling for various distributions of variability over time and temporal consistency constraints to capture sequential gaze dynamics related to ASD behavioral patterns. The framework was evaluated using public eye-tracking corpus containing temporally ordered gaze recordings from ASD and TD participants across age groups. Five sequential experiments on baseline classification, class-balancing analysis, cross-age evaluation, ablation analysis, and cross-dataset transfer learning were performed to conduct experiment-based evaluations. Model performed 0.91 in in-domain Area Under the Receiver Operating Characteristic Curve (AUC) and 0.81 in F1-score on the primary eye-tracking dataset. In the cross-dataset assessment stage, the framework presented a relatively stable performance, with an AUC of 0.85 and an average F1-score of 0.74, irrespective of differences in participant distributions and recording conditions. Ablation analysis also revealed that entropy regularization and temporal consistency mechanisms played a significant role in model stability and classification performance. The ablation analysis provides additional insight into the contribution of the proposed framework components beyond the overall classification performance. Removing the entropy-based regularization reduced the model’s ability to represent variability in gaze allocation, whereas removing the temporal-consistency regularization resulted in less stable sequence representations during learning. These observations indicate that the proposed components complement the transformer-based sequence encoder by improving representation stability and preserving diagnostically relevant temporal information. Rather than acting as independent classifiers, the regularization mechanisms serve as supporting constraints that enhance the quality and robustness of the learned temporal representations. The results indicate that temporally structured gaze modeling is more robust, interpretable, and general in comparison to static gaze representations. In summary, the presented framework can represent a scalable and developmentally appropriate approach to gaze-based ASD classification and support the implementation of trusted neurodevelopmental screening systems. Full article
(This article belongs to the Special Issue Integrated IoT and Sensing in Healthcare System)
►▼ Show Figures

Figure 1

29 pages, 3497 KB  
Article
BDC-YOLO: A Novel Architecture Coupling Dynamic Serpentine Convolutions with Bi-Level Routing Attention for Road Defect Detection
by Bo Yang, Hongli Sheng, Chen Geng, Chen Chen and Huiqing Lian
Sensors 2026, 26(16), 5301; https://doi.org/10.3390/s26165301 - 21 Aug 2026
Viewed by 349
Abstract
Accurate pavement distress identification is essential for infrastructure maintenance. However, prevailing models frequently underperform in complicated environments due to extreme scale variations, atypical defect geometries, and severe background noise. To mitigate these limitations, this study presents BDC-YOLO, an upgraded detection network built upon [...] Read more.
Accurate pavement distress identification is essential for infrastructure maintenance. However, prevailing models frequently underperform in complicated environments due to extreme scale variations, atypical defect geometries, and severe background noise. To mitigate these limitations, this study presents BDC-YOLO, an upgraded detection network built upon the YOLOv8 baseline. The proposed architecture structurally incorporates three specialized mechanisms: Bi-level Routing Attention (BRA) to isolate relevant target features from background artifacts; Dynamic Snake Convolution (DySnakeConv) to capture the topological characteristics of elongated and irregularly shaped cracks; and Content-Aware ReAssembly of FEatures (CARAFE) to minimize information degradation during upsampling and refine multi-scale feature fusion. Evaluated on the RDDChina dataset, BDC-YOLO demonstrates superior accuracy over the baseline and comparative state-of-the-art methods. Specifically, the framework yields a mAP0.5 of 88.9%, representing an absolute gain of 4.7% against the standard YOLOv8 model while achieving an inference speed of 175.4 FPS. Full article
(This article belongs to the Special Issue Artificial Intelligence-Based Target Recognition and Remote Sensing Data Processing)
►▼ Show Figures

Figure 1

18 pages, 7772 KB  
Article
Hierarchically Structured V2O5/PANI Heterostructures for Room-Temperature Ammonia Sensing
by Chunmei Shangguan, Anan Xu, Fang Wang, Ying Li, Jiao Jia and Zhenchen Liu
Sensors 2026, 26(16), 5300; https://doi.org/10.3390/s26165300 - 21 Aug 2026
Viewed by 271
Abstract
Ammonia, a toxic and volatile pollutant commonly found in chemical industrial environments, requires reliable real-time detection to ensure industrial safety and effective environmental monitoring. Conventional gas sensors typically operate at elevated temperatures, resulting in high power consumption. Moreover, pure metal oxides and conductive [...] Read more.
Ammonia, a toxic and volatile pollutant commonly found in chemical industrial environments, requires reliable real-time detection to ensure industrial safety and effective environmental monitoring. Conventional gas sensors typically operate at elevated temperatures, resulting in high power consumption. Moreover, pure metal oxides and conductive polymers often suffer from significant aggregation and exhibit suboptimal sensing performance under ambient conditions, limiting their practical applications. In this study, hierarchical porous V2O5/PANI composites were synthesized via a straightforward one-step coprecipitation method combined with in situ polymerization. The interlaced architecture of polyaniline (PANI) and vanadium pentoxide (V2O5) effectively reduces structural aggregation and increases the availability of surface active sites. Furthermore, the synergistic interaction at the bi-phase interface significantly enhances charge carrier transport, leading to improved ammonia-sensing capabilities at room temperature. Notably, the composite containing 20% V2O5 demonstrated superior response, selectivity, and reproducibility toward 10 ppm NH3. Due to its simple fabrication process and room-temperature operation without external heating, the developed V2O5/PANI composite sensor holds significant potential for practical applications in low-concentration ammonia detection under ambient conditions. Full article
(This article belongs to the Special Issue Smart Gas Sensor Applications in Environmental Change Monitoring)
►▼ Show Figures

Graphical abstract

23 pages, 2766 KB  
Article
Cloud–Edge Collaborative Personalized Deployment of Knowledge Bases in Semantic Communications
by Kaixiang Yang, Yushen Han, Yikai Xu and Mingkai Chen
Sensors 2026, 26(16), 5299; https://doi.org/10.3390/s26165299 - 21 Aug 2026
Viewed by 360
Abstract
With the rapid evolution of next-generation mobile communications, semantic communication has emerged as an intelligent communication paradigm capable of surpassing the Shannon limit. A fundamental prerequisite for this paradigm is the synchronization of background knowledge between the transmitter and receiver, making the semantic [...] Read more.
With the rapid evolution of next-generation mobile communications, semantic communication has emerged as an intelligent communication paradigm capable of surpassing the Shannon limit. A fundamental prerequisite for this paradigm is the synchronization of background knowledge between the transmitter and receiver, making the semantic knowledge base (SKB) a critical cornerstone. However, effectively selecting appropriate content from massive cloud-based knowledge repositories for edge deployment remains a significant challenge. This paper conducts systematic research to address the key issues in the flow deployment of SKBs at the edge, including insufficient adaptation to personalized preferences, inadequate timeliness management, and the complexity of multi-objective optimization. First, a comprehensive system model is constructed, integrating user preferences, knowledge relevance, transceiver matching degree, and the Age of Information (AOI). Second, the Generative Adversarial Network (GAN)-assisted Preference-based Reinforcement Learning (GaPbRL) algorithm is proposed. The experimental results demonstrate that this method outperforms traditional schemes in terms of knowledge-base hit rate, transceiver matching degree, and algorithm convergence speed, while significantly reducing the overhead of manual fine-tuning. This study provides a robust framework for the personalized and efficient cloud–edge collaborative deployment of SKBs. Full article
(This article belongs to the Special Issue The Key Technologies for Wireless Communication, Computing and Sensing)
►▼ Show Figures

Figure 1

22 pages, 5725 KB  
Article
A Priority-Aware Multi-Agent Reinforcement Learning Framework for Collaborative Intelligent Sensing in Social IoT
by Jing Zhu
Sensors 2026, 26(16), 5298; https://doi.org/10.3390/s26165298 - 21 Aug 2026
Viewed by 263
Abstract
Collaborative intelligent sensing in the Social Internet of Things (Social IoT) relies on distributed AI-enabled sensors to support complementary information sharing, multimodal perception, and real-time autonomous decision-making. Under high-load conditions, mismatches between resource provisioning and sensing quality of experience (QoE) can significantly degrade [...] Read more.
Collaborative intelligent sensing in the Social Internet of Things (Social IoT) relies on distributed AI-enabled sensors to support complementary information sharing, multimodal perception, and real-time autonomous decision-making. Under high-load conditions, mismatches between resource provisioning and sensing quality of experience (QoE) can significantly degrade system performance in applications such as smart cities. To address this issue, this paper proposes a service priority-aware collaborative sensing support framework based on a joint next-generation passive optical network (NG-PON) and cooperative intelligent service-based radio access network (CIS-RAN) architecture. The framework enables edge AI-driven inference and distributed sensor collaboration in heterogeneous Social IoT environments. Service-slice-specific priority weights are assigned to optical network units (ONUs) and wavelengths according to the QoE requirements and latency sensitivity of sensing tasks, allowing dynamic wavelength tuning that prioritizes high-impact collaborative services. The utility of a centralized intelligent processing pool is formulated to achieve priority-consistent and efficient resource coordination under collaborative constraints. In addition, a multi-agent AI-driven optimization framework is employed to derive adaptive resource allocation strategies that incorporate service priorities while satisfying stringent service-level agreements (SLAs). Simulation results show that the proposed framework improves system-level proxy metrics, including total utility, wavelength satisfaction, and resource utilization, compared with representative baseline schemes. Full article
(This article belongs to the Special Issue Collaborative Intelligent Sensing for Social IoT)
►▼ Show Figures

Figure 1

25 pages, 15896 KB  
Article
Privacy-Preserving and Poisoning-Robust Federated Learning for Industrial IoT
by Huan Yin, Congwen Chen, Jingyi Zhang, Dian Yu and Shuanggen Liu
Sensors 2026, 26(16), 5297; https://doi.org/10.3390/s26165297 - 21 Aug 2026
Viewed by 288
Abstract
With the rapid development of Industrial Internet of Things (IIoT), large amounts of sensor data generated by industrial devices and edge nodes have become the basis of intelligent manufacturing applications. Collaborative modeling on these distributed data is important for tasks such as anomaly [...] Read more.
With the rapid development of Industrial Internet of Things (IIoT), large amounts of sensor data generated by industrial devices and edge nodes have become the basis of intelligent manufacturing applications. Collaborative modeling on these distributed data is important for tasks such as anomaly detection, equipment monitoring, and predictive maintenance. Federated learning offers a practical way to train models without exposing raw sensor data, but it still faces privacy leakage and malicious poisoning attacks. To address these issues, this paper proposes a hierarchical privacy protection and poisoning-robust defense framework for industrial federated learning. Starting from the sensitivity differences among parameters at different model layers, the proposed method designs a hierarchical privacy-budget allocation strategy that enhances protection for sensitive information while minimizing the performance impact of perturbation. Meanwhile, a multi-layer, multi-feature anomaly-detection mechanism is adopted to identify malicious updates by jointly exploiting directional consistency, scale stability, and inter-layer similarity, and majority voting together with update clipping is used to further improve system robustness. Experiments on Fashion-MNIST, MVTec AD, and C-MAPSS demonstrate that the proposed method can effectively suppress global-model degradation under multiple poisoning attacks and achieves a favorable balance among privacy protection strength, robustness, and training efficiency. Full article
(This article belongs to the Special Issue Cyber Security and Privacy in Internet of Things (IoT))
►▼ Show Figures

Figure 1

16 pages, 3980 KB  
Article
A Gear-Driven Plantar Energy Harvester with Integrated Self-Sensing for Human Locomotion Recognition
by Xinrui Wang, Weiqi Lin, Wenda Wang, Yang Yu, Moyue Cong, Yongzhuo Gao and Wei Dong
Sensors 2026, 26(16), 5296; https://doi.org/10.3390/s26165296 - 21 Aug 2026
Viewed by 301
Abstract
Wearable electronic systems require compact and sustainable power sources together with reliable motion-sensing functions. This study presents a gear-driven plantar energy harvester that integrates biomechanical energy conversion with self-sensing locomotion recognition. The device converts low-frequency vertical foot loading into rotary motion through a [...] Read more.
Wearable electronic systems require compact and sustainable power sources together with reliable motion-sensing functions. This study presents a gear-driven plantar energy harvester that integrates biomechanical energy conversion with self-sensing locomotion recognition. The device converts low-frequency vertical foot loading into rotary motion through a wedge–lever transmission and amplifies the rotational speed using a multistage gear train with a total transmission ratio of 12. A one-way bearing enables directional power transmission during loading and prevents reverse rotation during recovery. The generated voltage serves both as the electrical output and as the sensing signal for locomotion recognition. Human-subject experiments were conducted under six locomotion modes: walking at 1, 2 and 3 m/s; running; ascending; and descending. Voltage signals were sampled at 2000 Hz and segmented into overlapping sequences. A CNN–LSTM model was used to extract local waveform features and temporal dependencies from the nonstationary signals. The model achieved an overall recognition accuracy of 98.8%, with most errors occurring between ascending and descending. The results demonstrate that a single plantar device can simultaneously harvest biomechanical energy and provide motion-related information, offering a compact solution for integrated energy harvesting and self-sensing in wearable systems. Full article
(This article belongs to the Special Issue Self-Powered Wearable Sensors for Mobile Autonomous Systems and Human–Robot Interaction)
►▼ Show Figures

Figure 1

24 pages, 5349 KB  
Article
MarShip-DET: A Frequency-Aware Multi-Scale Fusion Algorithm for Ship Detection in Maritime Remote Sensing Imagery
by Keren Chen, Xufang Zhu, Zhikun Liu, Fuyan Zhao and Kang Wang
Sensors 2026, 26(16), 5295; https://doi.org/10.3390/s26165295 - 21 Aug 2026
Viewed by 295
Abstract
To address the challenges of multi-scale target variation, complex background interference, and insufficient feature fusion quality in maritime remote sensing ship detection, this paper proposes MarShip-DET, a frequency-aware multi-scale fusion detection algorithm based on YOLO11n. Three core modules are introduced: Channel-Decoupled Progressive Feature [...] Read more.
To address the challenges of multi-scale target variation, complex background interference, and insufficient feature fusion quality in maritime remote sensing ship detection, this paper proposes MarShip-DET, a frequency-aware multi-scale fusion detection algorithm based on YOLO11n. Three core modules are introduced: Channel-Decoupled Progressive Feature Extraction Module (CDPFEM), which employs asymmetric channel decoupling with dual-statistic channel attention and image-relative-position-encoded multi-head self-attention to enhance discriminative feature extraction; Edge-Aware Region Context Fusion Module (EARCFusion), which integrates learnable Sobel edge sensing and cross-attention correction to achieve precise foreground refinement; and Wavelet-guided Prototype Attention Module (WavePAM), which combines Haar wavelet frequency decomposition with prototype-guided spatial compression attention to strengthen deep semantic representation. Experiments on HRSC2016 demonstrate that MarShip-DET achieves an mAP50 of 94.9% and an mAP50-95 of 84.4%, improving by 3.7% and 4.3% over the baseline, respectively. Zero-shot experiments on HRSID and SSDD, including comparisons with YOLO11n, D-FINE-N, and YOLOv13n, provide additional evidence of cross-domain transferability under the evaluated protocol. Full article
(This article belongs to the Section Remote Sensors)
►▼ Show Figures

Figure 1

35 pages, 6221 KB  
Article
A Dim Space Target Detection and Track Association Method for Dense Stellar Backgrounds
by Cheng Jiang, Zhixia Yang, Zhongqi Ma, Chiming Tong and Jinshen Wang
Sensors 2026, 26(16), 5294; https://doi.org/10.3390/s26165294 - 21 Aug 2026
Viewed by 317
Abstract
In space target surveillance missions, effective detection of dim space targets remains a challenge due to dense star interference and the random nature of target motion. To quickly and accurately extract dim space targets from complex star backgrounds, this paper proposes a dim [...] Read more.
In space target surveillance missions, effective detection of dim space targets remains a challenge due to dense star interference and the random nature of target motion. To quickly and accurately extract dim space targets from complex star backgrounds, this paper proposes a dim space target detection and track association method for dense star backgrounds. This paper analyzes various features of the target and background from a combined spatiotemporal perspective, including three main stages. First, inter-frame registration is used to filter out bright stars, followed by connected domain post-processing, which simplifies the star map background while enhancing the signal-to-noise ratio of dim targets. Secondly, an image difference fusion coarse processing module is proposed. The reconstructed multi-impulse function is derived from the spectral phase difference to estimate the displacement parameters of different components, after which the image difference fusion is designed to obtain candidate targets. Third, a directional track association algorithm is designed, with the candidate targets as the center and the motion parameters as thresholds, narrowing the association range to a fan-shaped region. This enables fast detection of target tracks while removing excess false alarms. The experimental results on four datasets demonstrate that this method outperforms traditional baseline methods in terms of target detection and localization accuracy. Full article
(This article belongs to the Section Navigation and Positioning)
►▼ Show Figures

Figure 1

14 pages, 2996 KB  
Article
A Static and Dynamic Combined Center of Mass Measurement Method Based on Multi-View Vision
by Daojing Qu, Xuhao Zhang, Genyou Wei, Meibao Wang and Zhiyao Xiang
Sensors 2026, 26(16), 5293; https://doi.org/10.3390/s26165293 - 21 Aug 2026
Viewed by 252
Abstract
The position of the center of mass directly affects the attitude control and flight safety of moving bodies such as unmanned aerial vehicles (UAVs). Therefore, high-precision measurement of the center of mass is required. Existing methods require changing the posture of the measured [...] Read more.
The position of the center of mass directly affects the attitude control and flight safety of moving bodies such as unmanned aerial vehicles (UAVs). Therefore, high-precision measurement of the center of mass is required. Existing methods require changing the posture of the measured object multiple times. This introduces repeated positioning errors and suffers from poor equipment versatility. To address these issues, this paper proposes a static and dynamic combined measurement method for the center of mass based on multi-view vision. First, the relationship between the swing period and the pendulum length under the simple pendulum principle is analyzed. The basic principle of determining the direction of the center of mass using the line of gravity is also examined. Second, an under-constrained compound pendulum fixture is designed. A binocular vision system is used to track circular markers, perform FFT-based period verification, and fit the gravity line using singular value decomposition (SVD). Third, using a standard cubic iron block as the test object, the influence of pendulum length and swing angle on measurement accuracy is studied. Finally, experiments verify that the proposed method can obtain three-dimensional coordinates of the center of mass under a single suspension condition. The results show that with a pendulum length of 330 mm and an initial swing angle of 4°, the root mean square error of the center of mass measurement is 0.70 mm, and the maximum deviation over five repeated measurements is 1.45 mm. This method does not require repeated lifting or changes in posture. It can meet the need for in-situ, high-precision center of mass measurement of UAVs and other aircraft. Full article
(This article belongs to the Section Sensing and Imaging)
►▼ Show Figures

Figure 1

18 pages, 3993 KB  
Article
Rail Light-Strip Abnormality Analysis from Color Inspection Images Using an Improved SegFormer and Geometric Rules
by Haoran Song, Yuntao Gou, Ning Wang, Le Wang, Junbo Liu, Shengchun Wang, Chengliang Xia, Qiang Han and Zichen Gu
Sensors 2026, 26(16), 5292; https://doi.org/10.3390/s26165292 - 21 Aug 2026
Viewed by 238
Abstract
Rail light-strip morphology reflects the wheel-rail contact condition. Reliable automatic analysis remains difficult. The strip is narrow and has weak boundaries, while specular reflection, rail-head texture and trackside background interfere with color inspection images. This study proposes a segmentation-guided geometric method for rail [...] Read more.
Rail light-strip morphology reflects the wheel-rail contact condition. Reliable automatic analysis remains difficult. The strip is narrow and has weak boundaries, while specular reflection, rail-head texture and trackside background interfere with color inspection images. This study proposes a segmentation-guided geometric method for rail light-strip abnormality analysis. An improved SegFormer jointly segments the background, rail-head and light-strip regions. A boundary detail enhancement module refines weak rail-head and light-strip contours. Focal Loss emphasizes minority and hard boundary pixels. The rail-head mask provides the geometric reference for extracting the light-strip centerline, eccentricity, width sequence and connected-component morphology. The predicted masks are ordered using the corrected mileage record. Every 1000 original-resolution rows then form a consecutive 1 m detection unit. When a geometric rule is triggered, the method reports that unit’s 1 m mileage interval together with its eccentricity, width-change or local-integrity measurement. The model achieves 95.67% mean Intersection over Union (mIoU) on 3520 annotated images. It detects 845 of 876 positive units, with 96.46% recall, 89.23% precision and 92.70% F1-score. The resulting records identify abnormal 1 m mileage intervals and report the corresponding eccentricity, width-change, or local-integrity measurements for targeted manual review. Full article
(This article belongs to the Special Issue Sensor-Based Image Processing and Sensing Techniques for Enhanced Object Detection)
►▼ Show Figures

Figure 1

14 pages, 1483 KB  
Article
Plasmonic Field-Enhanced Raman Sensing Enables Rapid Trace Methanol Detection in Transformer Oil
by Xiaoqin Zhang, Hongbin Zhu, Hao Liu, Jin Cao, Han Shi and Shanyuan Niu
Sensors 2026, 26(16), 5291; https://doi.org/10.3390/s26165291 - 21 Aug 2026
Viewed by 286
Abstract
Methanol is a critical molecular marker for the early aging of oil-paper insulation, and its rapid detection is highly valuable for insulation condition assessment and the fault warning of power transformers. Widely used chromatographic methods require sophisticated pretreatment workflow and are not suitable [...] Read more.
Methanol is a critical molecular marker for the early aging of oil-paper insulation, and its rapid detection is highly valuable for insulation condition assessment and the fault warning of power transformers. Widely used chromatographic methods require sophisticated pretreatment workflow and are not suitable for in situ monitoring. Non-destructive spectroscopic methods remain challenging due to the intrinsically small cross section of trace molecules in complex liquid environments. The rapid, direct detection of trace methanol in an oil mixture has yet to be demonstrated. In this study, a high-performance Raman-enhancing substrate was developed through hierarchical microstructure regulation, combining microscale light-trapping structures and nanoscale field-confinement sites to sense the weak Raman response of methanol in transformer oil. Direct detection of ppm-level methanol in the oil matrix was achieved, without additional adsorption enrichment or other complicated pretreatment procedures. The characteristic Raman band of methanol in transformer oil was identified, and a quantitative sensing method was established. Furthermore, the intrinsic temperature-dependent Raman response of methanol was investigated to evaluate the stability of its characteristic fingerprint bands over a broad temperature range. This work demonstrates a rapid, sensitive, and pretreatment-free spectroscopic strategy for trace methanol detection in complex oil matrices, and also sheds light on the high-sensitivity detection of small molecular markers in complex liquid environments. Full article
(This article belongs to the Section Electronic Sensors)
►▼ Show Figures

Figure 1

27 pages, 4720 KB  
Article
SEMU-Net: A Structure-Enhanced Multi-Branch U-Shaped Network for High-Resolution Remote Sensing Land-Cover Segmentation
by Bingyan Lu, Mei Li, Xiaorong Xue, Wen Zhang, Xin Zhao, Jingtong Yang, Yishuo Tian and Wancheng Wang
Sensors 2026, 26(16), 5290; https://doi.org/10.3390/s26165290 - 20 Aug 2026
Viewed by 496
Abstract
High-resolution remote sensing semantic segmentation remains challenging because repeated downsampling progressively weakens the fine-grained spatial information of small objects, while direct fusion of heterogeneous multi-scale features may introduce semantic discrepancies and redundant background responses. To address these issues, this study proposes SEMU-Net, a [...] Read more.
High-resolution remote sensing semantic segmentation remains challenging because repeated downsampling progressively weakens the fine-grained spatial information of small objects, while direct fusion of heterogeneous multi-scale features may introduce semantic discrepancies and redundant background responses. To address these issues, this study proposes SEMU-Net, a structure-enhanced multi-branch U-shaped network. First, an independent multi-scale complementary branch is constructed outside the main encoder pathway to provide auxiliary hierarchical representations and compensate for information degradation during progressive semantic abstraction. Second, a scale-consistent feature embedding module is introduced to project and normalize side-branch features before residual injection, thereby improving the compatibility of cross-path feature representations. Third, a discriminative channel modulation module is incorporated into the decoder to adaptively strengthen task-relevant channel responses and suppress redundant background activations. Experiments were conducted on the ISPRS Vaihingen dataset and a self-annotated high-resolution remote sensing dataset. On the Vaihingen dataset, SEMU-Net achieved a mIoU of 72.21% and an Average F1 score of 83.64%, outperforming the strongest competing method by 0.59 and 0.44 percentage points, respectively. The IoU of the Car class increased by 3.50 percentage points. On the self-annotated dataset, the IoU of the narrow Road class improved by 4.12 percentage points. These results demonstrate that SEMU-Net improves overall segmentation accuracy and enhances the recognition of small objects, with the observed improvements being consistent with the design objectives of multi-scale information compensation, cross-path feature adaptation, and channel recalibration. Full article
(This article belongs to the Section Remote Sensors)
►▼ Show Figures

Figure 1

16 pages, 3582 KB  
Article
Accurate Classification of Minor Leaks in Primary Loop Systems Using a Multi-Scale Attention Temporal Convolutional Autoencoder
by Jinghua Yang, Xiaohua Yang, Jie Liu, Zhuoran Xu and Guorui Huang
Sensors 2026, 26(16), 5289; https://doi.org/10.3390/s26165289 - 20 Aug 2026
Viewed by 300
Abstract
Minor leaks in the reactor coolant system (RCS) can provide early warning of degradation that may progress towards safety-relevant failures in nuclear power plants. Their timely localization remains difficult because they generate weak thermal-hydraulic perturbations and labeled leak data are rarely available. Here, [...] Read more.
Minor leaks in the reactor coolant system (RCS) can provide early warning of degradation that may progress towards safety-relevant failures in nuclear power plants. Their timely localization remains difficult because they generate weak thermal-hydraulic perturbations and labeled leak data are rarely available. Here, we present MATCA-TM, a physics-informed zero-shot framework that combines a Multi-scale Attention Temporal Convolutional Autoencoder (MATCA) with theoretical response-template matching for early leak localization. The framework learns multivariate temporal dependencies solely from normal operating data, enabling reconstruction residuals to highlight subtle anomalies caused by leaks. Physics-informed templates constructed from thermal-hydraulic response characteristics then provide a physically motivated basis for localizing leak sources. On the II + CPR1000 simulation platform, which represents a CPR1000 pressurized water reactor model, MATCA-TM achieved 65.3% localization accuracy for an equivalent leakage diameter of 6.7 × 10−3 cm and outperformed the evaluated machine learning and deep learning baselines under the same protocol. By combining data-driven representation learning with domain physics, MATCA-TM provides a physics-informed, simulation-validated approach to diagnosing incipient RCS leaks and a basis for further validation under plant-relevant operating conditions. Full article
(This article belongs to the Section Industrial Sensors)
►▼ Show Figures

Figure 1

17 pages, 5850 KB  
Article
A Federated Machine Learning Approach for the Detection and Visualisation of Eye Diseases Using Activation Maps
by Filomena Niro, Miriam Di Renzo, Patrizia Agnello, Marta Petyx, Fabio Martinelli, Maurizio Maddalena, Mario Cesarelli, Antonella Santone and Francesco Mercaldo
Sensors 2026, 26(16), 5288; https://doi.org/10.3390/s26165288 - 20 Aug 2026
Viewed by 312
Abstract
Eye diseases, particularly glaucoma and cataracts, are the leading causes of visual impairment, compromising quality of life. Automatic classification of these diseases can lead to more accurate and timely diagnosis, thereby limiting their progression and complications. In recent years, advances in Deep Learning [...] Read more.
Eye diseases, particularly glaucoma and cataracts, are the leading causes of visual impairment, compromising quality of life. Automatic classification of these diseases can lead to more accurate and timely diagnosis, thereby limiting their progression and complications. In recent years, advances in Deep Learning (DL) have shown promising results in the study of images in ophthalmology. However, traditional DL models are based on a centralised approach to data, sharing sensitive patient information and compromising privacy; moreover, the models are often difficult to interpret. In this paper we propose a method aimed to solve the issues of privacy and transparency in decision-making related to eye diseases detection and localisation. As a matter of fact, we consider Federated Learning (FL), an approach based on data decentralisation that enables collaborative learning between different clients and sends only the model weights to the central server. In this way, sensitive patient data are not shared, ensuring security and privacy. With regard to eye disease classification we exploit a Vision Transformer, which allows global relationships within retinal images to be highlighted, improving representation capabilities compared to traditional convolutional architectures. Furthermore, the proposed method also aims to make the model explainable using explainability techniques, in this way we make diagnostic decisions transparent. The experimental analysis shows an accuracy of 0.8480, a precision of 0.8645, a recall of 0.8477, showing the effectiveness of the proposed method on eye disease detection. Full article
(This article belongs to the Special Issue Emerging Trends in Artificial Intelligence for Biomedical Image Analysis: 2nd Edition)
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-310
Previous Issue
Next Issue
Back to TopTop