Sensors

17 pages, 14678 KB

Open AccessArticle

Preamble Injection-Based Jamming Method for UAV LoRa Communication Links

by Teng Wu, Runze Mao, Yan Du, Quan Zhu, Shengjun Wei and Changzhen Hu

Sensors 2026, 26(2), 614; https://doi.org/10.3390/s26020614 (registering DOI) - 16 Jan 2026

The widespread use of low-cost, highly maneuverable unmanned aerial vehicles (UAVs), such as racing drones, has raised numerous security concerns. These UAVs commonly employ LoRa (Long Range) communication protocols, which feature long-range transmission and strong anti-interference capabilities. However, traditional countermeasure techniques targeting LoRa-based [...] Read more.

The widespread use of low-cost, highly maneuverable unmanned aerial vehicles (UAVs), such as racing drones, has raised numerous security concerns. These UAVs commonly employ LoRa (Long Range) communication protocols, which feature long-range transmission and strong anti-interference capabilities. However, traditional countermeasure techniques targeting LoRa-based links often suffer from delayed response, poor adaptability, and high power consumption. To address these challenges, this study first leverages neural networks to achieve efficient detection and reverse extraction of key parameters from LoRa signals in complex electromagnetic environments. Subsequently, a continuous preamble injection jamming method is designed based on the extracted target signal parameters. By protocol-level injection, this method disrupts the synchronization and demodulation processes of UAV communication links, significantly enhancing jamming efficiency while reducing energy consumption. Experimental results demonstrate that, compared with conventional approaches, the proposed continuous preamble injection jamming method achieves improved signal detection accuracy, jamming energy efficiency, and effective range. To the best of our knowledge, this protocol-aware scheme, which integrates neural network-based signal perception and denoising, offers a promising and cost-effective technical pathway for UAV countermeasures. Full article

(This article belongs to the Special Issue LoRa Communication Technology for IoT Applications—2nd Edition)

61 pages, 7462 KB

Open AccessArticle

An Integrated Cyber-Physical Digital Twin Architecture with Quantitative Feedback Theory Robust Control for NIS2-Aligned Industrial Robotics

by Vesela Karlova-Sergieva, Boris Grasiani and Nina Nikolova

Sensors 2026, 26(2), 613; https://doi.org/10.3390/s26020613 (registering DOI) - 16 Jan 2026

Abstract

This article presents an integrated framework for robust control and cybersecurity of an industrial robot, combining Quantitative Feedback Theory (QFT), digital twin (DT) technology, and a programmable logic controller–based architecture aligned with the requirements of the NIS2 Directive. The study considers a five-axis [...] Read more.

This article presents an integrated framework for robust control and cybersecurity of an industrial robot, combining Quantitative Feedback Theory (QFT), digital twin (DT) technology, and a programmable logic controller–based architecture aligned with the requirements of the NIS2 Directive. The study considers a five-axis industrial manipulator modeled as a set of decoupled linear single-input single-output systems subject to parametric uncertainty and external disturbances. For position control of each axis, closed-loop robust systems with QFT-based controllers and prefilters are designed, and the dynamic behavior of the system is evaluated using predefined key performance indicators (KPIs), including tracking errors in joint space and tool space, maximum error, root-mean-square error, and three-dimensional positional deviation. The proposed architecture executes robust control algorithms in the MATLAB/Simulink environment, while a programmable logic controller provides deterministic communication, time synchronization, and secure data exchange. The synchronized digital twin, implemented in the FANUC ROBOGUIDE environment, reproduces the robot’s kinematics and dynamics in real time, enabling realistic hardware-in-the-loop validation with a real programmable logic controller. This work represents one of the first architectures that simultaneously integrates robust control, real programmable logic controller-based execution, a synchronized digital twin, and NIS2-oriented mechanisms for observability and traceability. The conducted simulation and digital twin-based experimental studies under nominal and worst-case dynamic models, as well as scenarios with externally applied single-axis disturbances, demonstrate that the system maintains robustness and tracking accuracy within the prescribed performance criteria. In addition, the study analyzes how the proposed architecture supports the implementation of key NIS2 principles, including command traceability, disturbance resilience, access control, and capabilities for incident analysis and event traceability in robotic manufacturing systems. Full article

(This article belongs to the Section Sensors and Robotics)

21 pages, 4103 KB

Open AccessArticle

Model-Centric or Data-Centric Approach? A Case Study on the Classification of Surface Defects in Steel Hot Rolling Using Convolutional Neural Networks

by Francisco López de la Rosa, José L. Gómez-Sirvent, Roberto Sánchez-Reolid, Rafael Morales and Antonio Fernández-Caballero

Sensors 2026, 26(2), 612; https://doi.org/10.3390/s26020612 (registering DOI) - 16 Jan 2026

Abstract

Any industrial application that uses convolutional neural networks (CNNs) requires initial data and resources in order to train the models. However, the selection of models must be appropriate to the quality and quantity of the available data and computational resources. This study analyses [...] Read more.

Any industrial application that uses convolutional neural networks (CNNs) requires initial data and resources in order to train the models. However, the selection of models must be appropriate to the quality and quantity of the available data and computational resources. This study analyses the influence of data quantity and quality on the performance of CNN models of different complexity. Image preprocessing and image transformation data augmentation techniques are applied to generate different amounts of synthetic data with which to train the aforementioned models, shedding light on the following question: does the quality and quantity of the data or the depth of the model have more influence? Different experiments are performed using the Northeastern University (NEU) Steel Surface Defects Database, which contains surface defects found in hot-rolled steel. After analyzing the results, the authors conclude that data quality and quantity have a much greater influence than model choice. As resources and time are often limited in industry and the ultimate goal is to maximize profit by increasing efficiency, the authors encourage researchers to carefully consider the industrial application at hand and analyze the available data and resources before selecting CNN models. Full article

(This article belongs to the Section Intelligent Sensors)

► Show Figures

Figure 1

27 pages, 1319 KB

Open AccessArticle

EMO-PEGASIS: A Dual-Phase Machine Learning Protocol for Energy Delay Optimisation in WSNs

by Abdulla Juwaied

Sensors 2026, 26(2), 611; https://doi.org/10.3390/s26020611 (registering DOI) - 16 Jan 2026

Abstract

Wireless sensor networks (WSNs) contend with the critical challenge of balancing energy conservation against data transmission delay, a trade-off that protocols such as PEGASIS—while being strong in energy efficiency—fail to manage optimally due to resulting high latency, unbalanced load distribution, and suboptimal cluster [...] Read more.

Wireless sensor networks (WSNs) contend with the critical challenge of balancing energy conservation against data transmission delay, a trade-off that protocols such as PEGASIS—while being strong in energy efficiency—fail to manage optimally due to resulting high latency, unbalanced load distribution, and suboptimal cluster formation. To address these limitations, this paper introduces the Enhanced Multi-Objective PEGASIS (EMO-PEGASIS) protocol, which is designed and implemented using a dual-phase machine learning strategy. This multi-objective approach works in two stages. First, it utilises K-means clustering to achieve robust spatial partitioning of the network. Second, it employs K-Nearest Neighbours (K-NN) classification to enable adaptive and intelligent routing. The simulation was performed using MATLAB R2025a, and the results show that EMO-PEGASIS addresses this multi-objective optimisation problem. The proposed EMO-PEGASIS protocol achieves a 45% reduction in average energy consumption, a 38% decrease in end-to-end delay, and a 67% increase in network lifetime compared to the original PEGASIS protocol. Additionally, EMO-PEGASIS demonstrates enhanced stability and effective load balancing under heterogeneous network configurations, while maintaining an excellent packet delivery ratio of 96.8%. These findings underscore the effectiveness of integrating machine learning techniques, which ultimately yield enhanced performance and enable reliable multi-objective optimisation within energy- and delay-constrained WSN environments. Full article

(This article belongs to the Special Issue Sensors Solutions for the Development of Climate Neutral and Smart Cities)

28 pages, 3390 KB

Open AccessArticle

SDC-YOLOv8: An Improved Algorithm for Road Defect Detection Through Attention-Enhanced Feature Learning and Adaptive Feature Reconstruction

by Hao Yang, Yulong Song, Yue Liang, Enhao Tang and Danyang Cao

Sensors 2026, 26(2), 609; https://doi.org/10.3390/s26020609 - 16 Jan 2026

Abstract

Road defect detection is essential for timely road damage repair and traffic safety assurance. However, existing object detection algorithms suffer from insufficient accuracy in detecting small road surface defects and are prone to missed detections and false alarms under complex lighting and background [...] Read more.

Road defect detection is essential for timely road damage repair and traffic safety assurance. However, existing object detection algorithms suffer from insufficient accuracy in detecting small road surface defects and are prone to missed detections and false alarms under complex lighting and background conditions. To address these challenges, this study proposes SDC-YOLOv8, an improved YOLOv8-based algorithm for road defect detection that employs attention-enhanced feature learning and adaptive feature reconstruction. The model incorporates three key innovations: (1) an SPPF-LSKA module that integrates Fast Spatial Pyramid Pooling with Large Separable Kernel Attention to enhance multi-scale feature representation and irregular defect modeling capabilities; (2) DySample dynamic upsampling that replaces conventional interpolation methods for adaptive feature reconstruction with reduced computational cost; and (3) a Coordinate Attention module strategically inserted to improve spatial localization accuracy under complex conditions. Comprehensive experiments on a public pothole dataset demonstrate that SDC-YOLOv8 achieves 78.0% mAP@0.5, 81.0% Precision, and 70.7% Recall while maintaining real-time performance at 85 FPS. Compared to the baseline YOLOv8n model, the proposed method improves mAP@0.5 by 2.0 percentage points, Precision by 3.3 percentage points, and Recall by 1.8 percentage points, yielding an F1 score of 75.5%. These results demonstrate that SDC-YOLOv8 effectively enhances small-target detection accuracy while preserving real-time processing capability, offering a practical and efficient solution for intelligent road defect detection applications. Full article

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

► Show Figures

Figure 1

20 pages, 7030 KB

Open AccessArticle

Latency-Aware Benchmarking of Large Language Models for Natural-Language Robot Navigation in ROS 2

by Murat Das, Zawar Hussain and Muhammad Nawaz

Sensors 2026, 26(2), 608; https://doi.org/10.3390/s26020608 - 16 Jan 2026

Abstract

A growing challenge in mobile robotics is the reliance on complex graphical interfaces and rigid control pipelines, which limit accessibility for non-expert users. This work introduces a latency-aware benchmarking framework that enables natural-language robot navigation by integrating multiple Large Language Models (LLMs) with [...] Read more.

A growing challenge in mobile robotics is the reliance on complex graphical interfaces and rigid control pipelines, which limit accessibility for non-expert users. This work introduces a latency-aware benchmarking framework that enables natural-language robot navigation by integrating multiple Large Language Models (LLMs) with the Robot Operating System 2 (ROS 2) Navigation 2 (Nav2) stack. The system allows robots to interpret and act upon free-form text instructions, replacing traditional Human–Machine Interfaces (HMIs) with conversational interaction. Using a simulated TurtleBot4 platform in Gazebo Fortress, we benchmarked a diverse set of contemporary LLMs, including GPT-3.5, GPT-4, GPT-5, Claude 3.7, Gemini 2.5, Mistral-7B Instruct, DeepSeek-R1, and LLaMA-3.3-70B, across three local planners, namely Dynamic Window Approach (DWB), Timed Elastic Band (TEB), and Regulated Pure Pursuit (RPP). The framework measures end-to-end response latency, instruction-parsing accuracy, path quality, and task success rate in standardised indoor scenarios. The results show that there are clear trade-offs between latency and accuracy, where smaller models respond quickly but have less spatial reasoning, while larger models have more consistent navigation intent but take longer to respond. The proposed framework is the first reproducible multi-LLM system with multi-planner evaluations within ROS 2, supporting the development of intuitive and latency-efficient natural-language interfaces for robot navigation. Full article

(This article belongs to the Special Issue Multimodal Sensing, Fusion, and VLMs for Scene Understanding and Robot Vision)

► Show Figures

Figure 1

18 pages, 2929 KB

Open AccessArticle

Vector Bending Sensor Based on Power-Monitored Tapered Few-Mode Multi-Core Fiber

by Qixuan Wu, Zhuyixiao Liu, Hao Wu and Ming Tang

Sensors 2026, 26(2), 607; https://doi.org/10.3390/s26020607 - 16 Jan 2026

Abstract

We propose a vector bending sensor based on a tapered few-mode multi-core fiber (FM-MCF). A seven-core six-mode fiber is tapered with an optimized taper ratio, enabling bending sensing through power monitoring. When the tapered FM-MCF bends, coupling occurs between the central core and [...] Read more.

We propose a vector bending sensor based on a tapered few-mode multi-core fiber (FM-MCF). A seven-core six-mode fiber is tapered with an optimized taper ratio, enabling bending sensing through power monitoring. When the tapered FM-MCF bends, coupling occurs between the central core and side cores in the tapered region. By monitoring the power of all cores and employing a power differential method, the bending direction and curvature can be reconstructed. The results show that within a curvature range of 2.5 m⁻¹ to 10 m⁻¹, the sensitivity of the ratio of the side core’s power to the middle core’s power with respect to curvature is not less than 0.14/m⁻¹. A deep fully connected feedforward neural network (DNN) is used to demodulate all power information and predict the bending shape of the optical fiber. The algorithm predicts the bending radius and rotation angle with mean absolute errors less than 0.038 m and 3.087°, respectively. This method is expected to achieve low-cost, high-sensitivity bending measurement applications with vector direction perception, providing an effective solution for scenarios with small curvatures that are challenging to detect using conventional sensing methods. Full article

(This article belongs to the Section Optical Sensors)

► Show Figures

Figure 1

23 pages, 16288 KB

Open AccessArticle

End-Edge-Cloud Collaborative Monitoring System with an Intelligent Multi-Parameter Sensor for Impact Anomaly Detection in GIL Pipelines

by Qi Li, Kun Zeng, Yaojun Zhou, Xiongyao Xie and Genji Tang

Sensors 2026, 26(2), 606; https://doi.org/10.3390/s26020606 - 16 Jan 2026

Abstract

Gas-insulated transmission lines (GILs) are increasingly deployed in dense urban power networks, where complex construction activities may introduce external mechanical impacts and pose risks to pipeline structural integrity. However, existing GIL monitoring approaches mainly emphasize electrical and gas-state parameters, while lightweight solutions capable [...] Read more.

Gas-insulated transmission lines (GILs) are increasingly deployed in dense urban power networks, where complex construction activities may introduce external mechanical impacts and pose risks to pipeline structural integrity. However, existing GIL monitoring approaches mainly emphasize electrical and gas-state parameters, while lightweight solutions capable of rapidly detecting and localizing impact-induced structural anomalies remain limited. To address this gap, this paper proposes an intelligent end-edge-cloud monitoring system for impact anomaly detection in GIL pipelines. Numerical simulations are first conducted to analyze the dynamic response characteristics of the pipeline under impacts of varying magnitudes, orientations, and locations, revealing the relationship between impact scenarios and vibration mode evolution. An end-tier multi-parameter intelligent sensor is then developed, integrating triaxial acceleration and angular velocity measurement with embedded lightweight computing. Laboratory impact experiments are performed to acquire sensor data, which are used to train and validate a multi-class extreme gradient boosting (XGBoost) model deployed at the edge tier for accurate impact-location identification. Results show that, even with a single sensor positioned at the pipeline midpoint, fusing acceleration and angular velocity features enables reliable discrimination of impact regions. Finally, a lightweight cloud platform is implemented for visualizing structural responses and environmental parameters with downsampled edge-side data. The proposed system achieves rapid sensor-level anomaly detection, precise edge-level localization, and unified cloud-level monitoring, offering a low-cost and easily deployable solution for GIL structural health assessment. Full article

(This article belongs to the Section Industrial Sensors)

► Show Figures

Figure 1

13 pages, 1929 KB

Open AccessArticle

Scanning Electrochemical Microscopy of Nystatin-Treated Yeast Used for Biofuel Cells

by Katazyna Blazevic, Antanas Zinovicius, Juste Rozene, Tomas Mockaitis, Ingrida Bruzaite, Laisvidas Striska, Evaldas Balciunas, Arunas Ramanavicius, Almira Ramanaviciene and Inga Morkvenaite

Sensors 2026, 26(2), 605; https://doi.org/10.3390/s26020605 - 16 Jan 2026

Abstract

Biofuel cells (BFCs) generate electricity by converting chemical energy into electrical energy using biological systems. Saccharomyces cerevisiae (yeast) is an attractive biocatalyst for BFCs due to its robustness, low cost, and metabolic versatility; however, electron transfer from the intracellular reactions to the electrode [...] Read more.

Biofuel cells (BFCs) generate electricity by converting chemical energy into electrical energy using biological systems. Saccharomyces cerevisiae (yeast) is an attractive biocatalyst for BFCs due to its robustness, low cost, and metabolic versatility; however, electron transfer from the intracellular reactions to the electrode is limited by the cell membrane. Nystatin is an antifungal antibiotic that increases the permeability of fungal membranes. We hypothesized that sub-lethal nystatin treatment could enhance mediator-assisted electron transfer without compromising cell viability. In this work, yeast was treated with nystatin during cultivation at concentrations of up to 6 µg/mL and combined with a dual-mediator system consisting of a lipophilic mediator (9,10-phenanthrenequinone, PQ) and a hydrophilic mediator (potassium ferricyanide). Scanning electrochemical microscopy revealed that the dual-mediator system increased local current responses by approximately fivefold compared to a single mediator (from ~11 pA to ~59 pA), and that nystatin-treated yeast exhibited higher local electrochemical activity than untreated yeast (maximum currents of ~0.476 nA versus ~0.303 nA). Microbial fuel cell measurements showed that nystatin treatment increased the maximum power density from approximately 0.58 mW/m² to approximately 0.62 mW/m² under identical conditions. Nystatin concentrations between 4 and 5 µg/mL maintain yeast viability at near-control levels, while higher concentrations cause a decrease in viability. These results demonstrate that controlled, sub-lethal membrane permeabilization combined with a dual-mediator strategy can enhance electron transfer in yeast-based biofuel cells. Full article

(This article belongs to the Special Issue Feature Papers in Biosensors Section 2025)

► Show Figures

Figure 1

20 pages, 4891 KB

Open AccessArticle

Active Inference Modeling of Socially Shared Cognition in Virtual Reality

by Yoshiko Arima and Mahiro Okada

Sensors 2026, 26(2), 604; https://doi.org/10.3390/s26020604 - 16 Jan 2026

Abstract

This study proposes a process model for sharing ambiguous category concepts in virtual reality (VR) using an active inference framework. The model executes a dual-layer Bayesian update after observing both self and partner actions and predicts actions that minimize free energy. To incorporate [...] Read more.

This study proposes a process model for sharing ambiguous category concepts in virtual reality (VR) using an active inference framework. The model executes a dual-layer Bayesian update after observing both self and partner actions and predicts actions that minimize free energy. To incorporate agreement-seeking with others into active inference, we added disagreement in category judgments as a risk term in the free energy, weighted by gaze synchrony measured using Dynamic Time Warping (DTW), which is assumed to reflect joint attention. To validate the model, an object classification task in VR including ambiguous items was created. The experiment was conducted first under a bot avatar condition, in which ambiguous category judgments were always incorrect, and then under a human–human pair condition. This design allowed verification of the collaborative learning process by which human pairs reached agreement from the same degree of ambiguity. Analysis of experimental data from 14 participants showed that the model achieved high prediction accuracy for observed values as learning progressed. Introducing gaze synchrony weighting (

γ_{0} \geq 0.5

) further improved prediction accuracy, yielding optimal performance. This approach provides a new framework for modeling socially shared cognition using active inference in human–robot interaction contexts. Full article

(This article belongs to the Special Issue Advances in Human–Robot Interactions Through Multimodal Sensing and Virtual Reality)

► Show Figures

Figure 1

20 pages, 3081 KB

Open AccessArticle

Fractional-Order Bioimpedance Modelling for Early Detection of Tissue Freezing in Cryogenic and Thermal Medical Applications

by Noelia Vaquero-Gallardo, Herminio Martínez-García and Oliver Millán-Blasco

Sensors 2026, 26(2), 603; https://doi.org/10.3390/s26020603 - 15 Jan 2026

Abstract

Cryotherapy and radiofrequency (RF) treatments modulate tissue temperature to induce therapeutic effects; however, improper application can result in thermal injury. Traditional temperature-based monitoring methods rely on multiple thermal sensors whose accuracy strongly depends on their number and spatial positioning, often failing to detect [...] Read more.

Cryotherapy and radiofrequency (RF) treatments modulate tissue temperature to induce therapeutic effects; however, improper application can result in thermal injury. Traditional temperature-based monitoring methods rely on multiple thermal sensors whose accuracy strongly depends on their number and spatial positioning, often failing to detect early tissue crystallization. This study introduces a fractional order bioimpedance modelling framework for the early detection of tissue freezing during cryogenic and thermal medical treatments, with the feasibility and effectiveness of this approach having been reported in our prior publications. While bioimpedance spectroscopy itself is a well-est. The corresponablished technique in biomedical engineering, its novel application to predict and identify premature freezing events provides a new pathway for safe and efficient energy-based therapies. Fractional-order models derived from the Cole family accurately reproduce the complex electrical behavior of biological tissues using fewer parameters than classical integer-order models, thus reducing both hardware requirements and computational cost. Experimental impedance data from human abdominal, gluteal, and femoral regions were modelled to extract fractional parameters that serve as sensitive indicators of phase-transition onset. The results demonstrate that the proposed approach enables real-time identification of freezing-induced electrical transitions, offering a physiologically grounded alternative to conventional temperature-based monitoring. Furthermore, the fractional order bioimpedance method exhibits high reproducibility and selectivity, and its analytical figures of merit, including the limits of detection and quantification, support its use for reliable real-time tissue monitoring and early injury detection. Overall, the proposed fractional order bioimpedance framework enhances both safety and control precision in cryogenic and thermal medical applications. Full article

(This article belongs to the Special Issue Feature Papers in Biosensors Section 2025)

► Show Figures

Figure 1

13 pages, 4033 KB

Open AccessArticle

A Low-Sidelobe Fully Metallic Ridge Gap Waveguide Antenna Array for W-Band Applications

by Huixia Jiang, Lili Sheng, Pengsheng Nie, Yu Feng, Jinfang Wen, Jianbo Ji and Weiping Cao

Sensors 2026, 26(2), 602; https://doi.org/10.3390/s26020602 - 15 Jan 2026

Abstract

To address the critical demand for high-gain, low-sidelobe, and high-efficiency antennas in W-band arrays, this work presents a low-sidelobe all-metal array antenna based on ridge gap waveguide technology. The design employs a three-layer contactless metal structure, integrating a stepped-ridge feeding network with Taylor [...] Read more.

To address the critical demand for high-gain, low-sidelobe, and high-efficiency antennas in W-band arrays, this work presents a low-sidelobe all-metal array antenna based on ridge gap waveguide technology. The design employs a three-layer contactless metal structure, integrating a stepped-ridge feeding network with Taylor amplitude distribution and a higher-order mode resonant cavity. This integration enables efficient power distribution and low-loss transmission while eliminating the need for conventional welding or bonding processes. Measurement results indicate that the antenna exhibits a reflection coefficient below −10 dB across the 92.5–103.5 GHz. The in-band gain exceeds 25.8 dBi with less than 1 dB fluctuation, and the radiation efficiency surpasses 78%. Specifically, the sidelobe levels in both E- and H-planes remain below −17.5 dB, reaching under −19.5 dB at 94 GHz, while cross-polarization is better than −30 dB. The proposed antenna demonstrates high gain, low sidelobe, and high efficiency, showing promising potential for applications in millimeter-wave radar, imaging, and 6G communication systems. Full article

(This article belongs to the Topic Advanced Array Antenna Design and Signal Processing Techniques)

► Show Figures

Figure 1

22 pages, 20327 KB

Open AccessArticle

PSgANet: Polar Sequence-Guided Attention Network for Edge-Related Defect Classification in Contact Lenses

by Sung-Hoon Kim, In Joo and Kwan-Hee Yoo

Sensors 2026, 26(2), 601; https://doi.org/10.3390/s26020601 - 15 Jan 2026

Abstract

The integration of artificial intelligence (AI) into industrial processes is a promising method for enhancing operational efficiency and quality control. In particular, contact lens manufacturing requires specialized artificial intelligence technologies owing to stringent safety requirements. This study introduces a novel approach that employs [...] Read more.

The integration of artificial intelligence (AI) into industrial processes is a promising method for enhancing operational efficiency and quality control. In particular, contact lens manufacturing requires specialized artificial intelligence technologies owing to stringent safety requirements. This study introduces a novel approach that employs polar coordinate transformation and a customized deep learning model, the Polar Sequence-guided Attention Network (PSgANet), to improve the accuracy of defect detection in the rim-connected zone (RCZ) of contact lenses. PSgANet is specifically designed to process polar coordinate-transformed image data by integrating sequence learning and attention mechanisms to maximise the capability for detecting and classifying defective patterns. This model converts irregularities along the edges of contact lenses into linear arrays via polar coordinate transformation, enabling a clearer and more consistent identification of defective regions. To achieve this, we applied sequence learning architectures such as GRU, LSTM, and Transformer within PSgANet and compared their performances with those of conventional models, including GoogleNetv4, EfficientNet, and Vision Transformer. The experimental results demonstrated that the PSgANet models outperformed the existing CNN-based models. In particular, the LSTM-based PSgANet achieved the highest accuracy and balanced precision and recall metrics, showing up to a 7.75% improvement in accuracy compared with the traditional GoogleNetv4 model. These results suggest that the proposed method is an effective tool for detecting and classifying defects within the RCZ during contact lens manufacturing processes. Full article

(This article belongs to the Section Sensing and Imaging)

► Show Figures

Figure 1

17 pages, 1776 KB

Open AccessArticle

Multi-Scale Adaptive Light Stripe Center Extraction for Line-Structured Light Vision Based Online Wheelset Measurement

by Saisai Liu, Qixin He, Wenjie Fu, Boshi Du and Qibo Feng

Sensors 2026, 26(2), 600; https://doi.org/10.3390/s26020600 - 15 Jan 2026

Abstract

The extraction of the light stripe center is a pivotal step in line-structured light vision measurement. This paper addresses a key challenge in the online measurement of train wheel treads, where the diverse and complex profile characteristics of the tread surface lead to [...] Read more.

The extraction of the light stripe center is a pivotal step in line-structured light vision measurement. This paper addresses a key challenge in the online measurement of train wheel treads, where the diverse and complex profile characteristics of the tread surface lead to uneven gray-level distribution and varying width features in the stripe image, ultimately degrading the accuracy of center extraction. To solve this problem, a region-adaptive multiscale method for light stripe center extraction is proposed. First, potential light stripe regions are identified and enhanced based on the gray-gradient features of the image, enabling precise segmentation. Subsequently, by normalizing the feature responses under Gaussian kernels with different scales, the locally optimal scale parameter (σ) is determined adaptively for each stripe region. Sub-pixel center extraction is then performed using the Hessian matrix corresponding to this optimal σ. Experimental results demonstrate that under on-site conditions featuring uneven wheel surface reflectivity, the proposed method can reliably extract light stripe centers with high stability. It achieves a repeatability of 0.10 mm, with mean measurement errors of 0.12 mm for flange height and 0.10 mm for flange thickness, thereby enhancing both stability and accuracy in industrial measurement environments. The repeatability and reproducibility of the method were further validated through repeated testing of multiple wheels. Full article

(This article belongs to the Special Issue Intelligent Sensors and Signal Processing in Industry)

► Show Figures

Figure 1

24 pages, 2116 KB

Open AccessArticle

Optical Design of a Large-Angle Spectral Confocal Sensor for Liquid Surface Tension Measurement

by Lingling Wu, Tingting Yang, Fang Wang, Qian Wang, Fei Xi and Jinsong Lv

Sensors 2026, 26(2), 599; https://doi.org/10.3390/s26020599 - 15 Jan 2026

Abstract

The surface tension of a liquid droplet can be determined by fitting its actual profiles using the Young–Laplace equation, effectively reducing the measurement of surface tension to an accurate determination of the droplet’s profiles. Spectral confocal sensors are high-precision, interference-resistant, non-contact measurement systems [...] Read more.

The surface tension of a liquid droplet can be determined by fitting its actual profiles using the Young–Laplace equation, effectively reducing the measurement of surface tension to an accurate determination of the droplet’s profiles. Spectral confocal sensors are high-precision, interference-resistant, non-contact measurement systems for droplet surface profiling, employing a light source together with a dispersive objective lens and a spectrometer to acquire depth-dependent spectral information. The accuracy and stability of surface tension measurements can be effectively enhanced by spectral confocal sensors measuring the droplet surface profile. Although existing spectral confocal sensors have significantly improved measurement range and accuracy, their angular measurement performance remains limited, and deviations may arise at droplet edges with large inclinations or pronounced surface profile variations. This study presents the optical design of a large-angle spectral confocal sensor. By theoretically analyzing the conditions for generating linear axial dispersion in the dispersive objective lens, a front-end dispersive objective lens was designed by combining positive and negative lenses. Based on a Czerny–Turner (C-T) configuration, the back-end spectrometer was designed under the astigmatism-free condition, taking into account both central and edge wavelength effects. Zemax was employed for simulation optimization and tolerance analysis of each optical module. The results show that the designed system achieves an axial dispersion of 1.5 mm over the 430–700 nm wavelength range, with a maximum allowable object angle of ±40° and a theoretical resolution of 3 μm. The proposed spectral confocal sensor maintains high measurement accuracy over a wide angular range, facilitating precise measurement of droplet surface tension at large inclination angles. Full article

(This article belongs to the Section Optical Sensors)

► Show Figures

Graphical abstract

17 pages, 1208 KB

Open AccessArticle

Portable Raspberry Pi Platform for Automated Interpretation of Lateral Flow Strip Tests

by Natalia Nakou, Panagiotis K. Tsikas and Despina P. Kalogianni

Sensors 2026, 26(2), 598; https://doi.org/10.3390/s26020598 - 15 Jan 2026

Abstract

Paper-based rapid tests are widely used in point-of-care diagnostics due to their simplicity and low cost. However, their application in quantitative analysis remains limited. In this work, a nucleic acid lateral flow assay (NALFA) was integrated with an automated image acquisition system built [...] Read more.

Paper-based rapid tests are widely used in point-of-care diagnostics due to their simplicity and low cost. However, their application in quantitative analysis remains limited. In this work, a nucleic acid lateral flow assay (NALFA) was integrated with an automated image acquisition system built on a Raspberry Pi platform for the quantitative detection of SARS-CoV-2 virus, increasing the accuracy of the test compared to subjective visual interpretation. The assay employed blue polystyrene microspheres as reporters, while automated image capturing, image processing and quantification were performed using custom Python software (version 3.12). Signal quantification was achieved by comparing the grayscale intensity of the test line with that of a simultaneously captured negative control strip, allowing correction for illumination and background variability. Calibration curves were used for the training of the algorithm. The system was applied for the analysis of a series of samples with varying DNA concentrations, yielding recoveries between 84 and 108%. The proposed approach integrates a simple biosensor with an accessible computational platform to achieve full low-cost automation. This work introduces the first Raspberry Pi-driven image processing approach for accurate quantification of NALFAs and establishes a foundation for future low-cost, portable diagnostic systems targeting diverse nucleic acids, proteins, and biomarkers. Full article

(This article belongs to the Special Issue Development and Application of Optical Chemical Sensing)

34 pages, 3406 KB

Open AccessArticle

Reconstructing Spatial Localization Error Maps via Physics-Informed Tensor Completion for Passive Sensor Systems

by Zhaohang Zhang, Zhen Huang, Chunzhe Wang and Qiaowen Jiang

Sensors 2026, 26(2), 597; https://doi.org/10.3390/s26020597 - 15 Jan 2026

Abstract

Accurate mapping of localization error distribution is essential for assessing passive sensor systems and guiding sensor placement. However, conventional analytical methods like the Geometrical Dilution of Precision (GDOP) rely on idealized error models, failing to capture the complex, heterogeneous error distributions typical of [...] Read more.

Accurate mapping of localization error distribution is essential for assessing passive sensor systems and guiding sensor placement. However, conventional analytical methods like the Geometrical Dilution of Precision (GDOP) rely on idealized error models, failing to capture the complex, heterogeneous error distributions typical of real-world environments. To overcome this challenge, we propose a novel data-driven framework that reconstructs high-fidelity localization error maps from sparse observations in TDOA-based systems. Specifically, we model the error distribution as a tensor and formulate the reconstruction as a tensor completion problem. A key innovation is our physics-informed regularization strategy, which incorporates prior knowledge from the analytical error covariance matrix into the tensor factorization process. This allows for robust recovery of the complete error map even from highly incomplete data. Experiments on a real-world dataset validate the superiority of our approach, showing an accuracy improvement of at least 27.96% over state-of-the-art methods. Full article

(This article belongs to the Special Issue Multi-Agent Sensors Systems and Their Applications)

► Show Figures

Figure 1

22 pages, 8300 KB

Open AccessArticle

Sign2Story: A Multimodal Framework for Near-Real-Time Hand Gestures via Smartphone Sensors to AI-Generated Audio-Comics

by Gul Faraz, Lei Jing and Xiang Li

Sensors 2026, 26(2), 596; https://doi.org/10.3390/s26020596 - 15 Jan 2026

Abstract

This study presents a multimodal framework that uses smartphone motion sensors and generative AI to create audio comics from live news headlines. The system operates without direct touch or voice input, instead responding to simple hand-wave gestures. The system demonstrates potential as an [...] Read more.

This study presents a multimodal framework that uses smartphone motion sensors and generative AI to create audio comics from live news headlines. The system operates without direct touch or voice input, instead responding to simple hand-wave gestures. The system demonstrates potential as an alternative input method, which may benefit users who find traditional touch or voice interaction challenging. In the experiments, we investigated the generation of comics on based on the latest tech-related news headlines using Really Simple Syndication (RSS) on a simple hand wave gesture. The proposed framework demonstrates extensibility beyond comic generation, as various other tasks utilizing large language models and multimodal AI could be integrated by mapping them to different hand gestures. Our experiments with open-source models like LLaMA, LLaVA, Gemma, and Qwen revealed that LLaVA delivers superior results in generating panel-aligned stories compared to Qwen3-VL, both in terms of inference speed and output quality, relative to the source image. These large language models (LLMs) collectively contribute imaginative and conversational narrative elements that enhance diversity in storytelling within the comic format. Additionally, we implement an AI-in-the-loop mechanism to iteratively improve output quality without human intervention. Finally, AI-generated audio narration is incorporated into the comics to create an immersive, multimodal reading experience. Full article

(This article belongs to the Special Issue Body Area Networks: Intelligence, Sensing and Communication)

► Show Figures

Figure 1

22 pages, 7304 KB

Open AccessArticle

Adaptive Trajectory-Constrained Heading Estimation for Tractor GNSS/SINS Integrated Navigation

by Shupeng Hu, Song Chen, Lihui Wang, Zhijun Meng, Weiqiang Fu, Yaxin Ren, Cunjun Li and Hao Wang

Sensors 2026, 26(2), 595; https://doi.org/10.3390/s26020595 - 15 Jan 2026

Abstract

Accurate heading estimation is crucial for the autonomous navigation of small-to-medium tractors. While dual-antenna GNSS systems offer precision, they face installation and safety challenges. Single-antenna GNSS integrated with a low-cost Strapdown Inertial Navigation System (SINS) presents a more adaptable solution but suffers from [...] Read more.

Accurate heading estimation is crucial for the autonomous navigation of small-to-medium tractors. While dual-antenna GNSS systems offer precision, they face installation and safety challenges. Single-antenna GNSS integrated with a low-cost Strapdown Inertial Navigation System (SINS) presents a more adaptable solution but suffers from slow convergence and low accuracy of heading estimation in low-speed farmland operations. This study proposes an adaptive trajectory-constrained heading estimation method. A sliding-window adaptive extended Kalman filter (SWAEKF) was developed, incorporating a heading constraint model that utilizes the GNSS-derived trajectory angle. An enhanced Sage–Husa algorithm was employed for the adaptive estimation of the trajectory angle measurement variance. Furthermore, a covariance initialization strategy based on the variance of trajectory angle increments was implemented to accelerate convergence. Field tests demonstrated that the proposed method achieved rapid heading convergence (less than 10 s for straight lines and 14 s for curves) and high accuracy (RMS heading error below 0.15° for straight-line tracking and 0.25° for curved paths). Compared to a conventional adaptive EKF, the SWAEKF improved accuracy by 23% and reduced convergence time by 62%. The proposed algorithm effectively enhances the performance of GNSS/SINS integrated navigation for tractors in low-dynamic environments, meeting the requirements for autonomous navigation systems. Full article

(This article belongs to the Section Navigation and Positioning)

► Show Figures

Figure 1

Journal Description

Sensors

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Topical Collections

Further Information

Guidelines

MDPI Initiatives

Follow MDPI