Sensors

26 pages, 1509 KB

Open AccessFeature PaperArticle

Quick Response Code Verification Using Anti-Counterfeiting Pattern and Multi-Feature Fusion Network

by Ke Sun, Zhongyuan Guo and Hong Zheng

Sensors 2026, 26(10), 3067; https://doi.org/10.3390/s26103067 - 12 May 2026

Quick response codes are widely used as anti-counterfeiting labels in the field of product packaging, but they are easily illegally copied. Thus, this paper introduces a quick response code verification method that combines an anti-counterfeiting pattern with a deep feature fusion network. Firstly, [...] Read more.

Quick response codes are widely used as anti-counterfeiting labels in the field of product packaging, but they are easily illegally copied. Thus, this paper introduces a quick response code verification method that combines an anti-counterfeiting pattern with a deep feature fusion network. Firstly, a specialized anti-counterfeiting quick response code is designed, composed of a standard quick response code and an anti-counterfeiting pattern, which is essentially a fine-grained random texture distribution sensitive to copying. Next, the anti-counterfeiting patterns are overlapped and divided into blocks during the data processing, which effectively expands the data volume and avoids the interference of pattern content on the authenticity identification. Then, a convolutional self-learning preprocessing layer is employed to initially learn the feature information that represents the difference between authenticity and forgery. Finally, a multi-feature fusion convolutional neural network is proposed to identity the authenticity of anti-counterfeiting patterns. The proposed network comprises two branches, facilitating multi-scale feature extraction and fusion. The effectiveness of the proposed approach is evaluated on a self-constructed quick response code dataset, and the experimental results demonstrate that the proposed approach outperforms traditional knowledge engineering methods and similar deep learning methods. Full article

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

12 pages, 1460 KB

Open AccessArticle

Novel Smartphone Paper Sensor for One Health: Monitoring Free Chlorine in Water and Exhaled Breath Condensate

by Caterina Cambrea, Robert Josue Rodriguez Arias, Riccardo Desiderio, Faisal Nazir, Maria Maddalena Calabretta and Elisa Michelini

Sensors 2026, 26(10), 3066; https://doi.org/10.3390/s26103066 - 12 May 2026

Abstract

Disinfection is essential to ensure safe drinking water and hygienic conditions in environmental, industrial, and clinical settings. However, conventional methods for monitoring free residual chlorine are often laboratory-based and not suited for decentralized analysis. Here, we report a novel paper-based colorimetric biosensing platform [...] Read more.

Disinfection is essential to ensure safe drinking water and hygienic conditions in environmental, industrial, and clinical settings. However, conventional methods for monitoring free residual chlorine are often laboratory-based and not suited for decentralized analysis. Here, we report a novel paper-based colorimetric biosensing platform that translates the ISO 7393-2 standard, a method based on the reaction of chlorine with N,N-diethyl-p-phenylenediamine (DPD), into a portable and user-friendly format. The proposed device integrates the DPD chemistry within a paper architecture, enabling reagent-free operation at the point of need. The sensor provides a rapid visual readout that is detectable by the naked eye, while quantitative analysis is achieved within 3 min through smartphone-based image acquisition. This work constitutes the first implementation of the ISO standard in a portable paper-based format suitable for both environmental and clinical matrices. The sensor provided a detection limit of 12 μM for sodium hypochlorite and was successfully validated in real samples, including bottled water and exhaled breath condensate, with satisfactory recoveries. Furthermore, the stability of the paper-based sensor was assessed under storage conditions of 4 °C and room temperature (23 °C), demonstrating excellent performance over 30 days in both cases, indicating that refrigeration is not required for maintaining sensor performance. Full article

(This article belongs to the Topic Functional Materials for Chemical Sensing and Environmental Sustainability: Trends, Concepts and Applications)

► Show Figures

Graphical abstract

19 pages, 953 KB

Open AccessArticle

Subject-Wise Depression Screening from Eight-Channel Resting-State EEG Using Asymmetry-Aware Spectral Features and Connectivity Ablation

by Hassan Ugail, Newton Howard, Ali Ahmed Elmahmudi and Zied Mnasri

Sensors 2026, 26(10), 3065; https://doi.org/10.3390/s26103065 - 12 May 2026

Abstract

Major depressive disorder remains difficult to diagnose objectively, as routine assessment is still largely dependent on clinical interview and rating scales. Resting-state electroencephalography (EEG) is an attractive complementary modality because it is non-invasive, low-cost, and compatible with wearable sensing, but many reported EEG [...] Read more.

Major depressive disorder remains difficult to diagnose objectively, as routine assessment is still largely dependent on clinical interview and rating scales. Resting-state electroencephalography (EEG) is an attractive complementary modality because it is non-invasive, low-cost, and compatible with wearable sensing, but many reported EEG classification results are weakened by segment-level leakage and unclear subject identity handling. This study evaluates whether depression can be distinguished from healthy controls using a compact eight-channel resting-state EEG configuration under a strictly leakage-free subject-wise protocol. Using a widely used public EEG dataset, we first corrected a previously overlooked subject-identity ambiguity by constructing a class-aware composite key, yielding 56 valid unique participants. We then applied ten repeated subject-wise holdout splits and compared five compact baselines spanning Extra Trees and a multi-layer perceptron on asymmetry-aware spectral features and three convolutional networks on raw signals, including the EEG-specific EEGNet and ShallowConvNet architectures. Uncertainty was quantified through 95% bootstrap confidence intervals of the mean across repeats. The best model, an Extra Trees classifier using eight-channel spectral and asymmetry features, achieved a mean balanced accuracy of 93.5% with a 95% bootstrap confidence interval of 89.6% to 96.8% and a mean area under the receiver operating characteristic curve of 98.6% with a 95% bootstrap confidence interval of 96.2% to 100.0%. A connectivity ablation showed that inter-channel coherence was informative in isolation but did not improve performance when naively fused with spectral features. A feature-selection ablation did not show evidence that the 90-dimensional spectral representation was dominated by noisy or uninformative dimensions under this evaluation protocol. These results support compact, subject-wise evaluated EEG screening pipelines while highlighting the importance of rigorous leakage control. Full article

(This article belongs to the Section Wearables)

20 pages, 2464 KB

Open AccessArticle

Application of Core–Shell Bimetallic Nanoparticles with Polydopamine-Assisted Nanogap in SERS-Based Lateral Flow Immunoassay of Prolactin

by Kseniya V. Serebrennikova, Nadezhda S. Komova, Anatoly V. Zherdev and Boris B. Dzantiev

Sensors 2026, 26(10), 3064; https://doi.org/10.3390/s26103064 - 12 May 2026

Abstract

The fabrication of SERS nanotags with efficient antibody loading and high signal enhancement remains a challenging task for combining surface-enhanced Raman spectroscopy (SERS) and lateral flow immunoassay (LFIA). In this study, bimetallic Au^DTNB@PDA^DTNB@Ag nanoparticles with a polydopamine (PDA)-based internal [...] Read more.

The fabrication of SERS nanotags with efficient antibody loading and high signal enhancement remains a challenging task for combining surface-enhanced Raman spectroscopy (SERS) and lateral flow immunoassay (LFIA). In this study, bimetallic Au^DTNB@PDA^DTNB@Ag nanoparticles with a polydopamine (PDA)-based internal nanogap were synthesized and functionalized with anti-prolactin monoclonal antibodies to produce SERS nanotags. Here, polydopamine serves both as a spacer providing a nanogap between the core and the shell, and as a reaction layer to capture Raman reporter 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB) within the nanogap. Regimes (conditions, protocols) for conjugating antibodies to Au^DTNB@PDA^DTNB@Ag were selected to preserve both the binding affinity for the target analyte and the Raman activity of the SERS nanotag. The SERS nanotag provides plasmonic absorption for visible colorimetric readout, as well as strong SERS signals for highly sensitive quantitative immunoassay. Measuring the Raman intensities of DTNB in the test zone after performing LFIA made it possible to determine prolactin with a detection limit of 0.2 ng/mL in the working range from 1 to 10 ng/mL. The achieved limit of detection was 10-fold lower than the LFIA coupled with colorimetric readout (4.7 ng/mL). The recoveries of prolactin from spiked serum samples were in the range of 70.2–82.6% with relative standard deviations of 2.3–6.8%. Overall, the Au^DTNB@PDA^DTNB@Ag nanotag demonstrated high stability, Raman activity, and specificity, indicating that the SERS nanotag with PDA-assisted internal nanogap is promising for use in SERS immunoassay of other target analytes. Full article

(This article belongs to the Special Issue Advances in Biosensors Based on Micro/Nanomaterials)

24 pages, 2668 KB

Open AccessArticle

LSTM-Based Estimation of Solar Energy Production Using Meteorological and Environmental Data: Karabük Case Study

by Fatih Gultekin, Muhammet Tahir Guneser and Mehmet Zahid Yildirim

Sensors 2026, 26(10), 3063; https://doi.org/10.3390/s26103063 - 12 May 2026

Abstract

This study proposes a Long Short-Term Memory (LSTM)-based deep learning model for short-, medium-, and long-term forecasting of solar energy production. Approximately four years of hourly data from four photovoltaic power plants in Karabük were used. In addition to production data, meteorological and [...] Read more.

This study proposes a Long Short-Term Memory (LSTM)-based deep learning model for short-, medium-, and long-term forecasting of solar energy production. Approximately four years of hourly data from four photovoltaic power plants in Karabük were used. In addition to production data, meteorological and environmental variables were included through a multivariate forecasting approach. The model was tested under three scenarios at different time scales. Performance was evaluated using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and coefficient of determination (R²) metrics. Results showed high prediction accuracy, particularly with seasonal data, where R² values exceeded 0.90 in most cases. In forecasts based on monthly data, performance was more variable, and the shorter data window limited the model’s learning capacity. Long-term analyses indicated that the model successfully captured overall production trends and achieved high accuracy across all Photovoltaic (PV) systems. The findings also revealed that incorporating meteorological and environmental variables significantly improved prediction performance. In particular, air pollution parameters were effective in long-term production forecasting. Overall, the study demonstrates that Long Short-Term Memory (LSTM)-based models are reliable and effective tools for solar energy forecasting, with strong potential for energy planning and smart grid applications. Full article

(This article belongs to the Section Environmental Sensing)

19 pages, 1062 KB

Open AccessArticle

Explainable and Subject-Independent VO₂ Estimation Using a Single IMU: A Lightweight Ensemble Framework Under LOSO Validation

by Vidyarani K. Rajashekaraiah, Viswanath Talasila, Rashmi Alva, Prem Venkatesan, Ravi Prasad K. Jagannath and Gurusiddappa R. Prashanth

Sensors 2026, 26(10), 3062; https://doi.org/10.3390/s26103062 - 12 May 2026

Abstract

Continuous estimation of oxygen uptake (VO₂) using wearable inertial sensors offers a practical alternative to laboratory-based metabolic testing but remains challenging due to the indirect relationship between kinematics and physiological demand. This study presents a lightweight two-stage pipeline for simultaneous heel-strike [...] Read more.

Continuous estimation of oxygen uptake (VO₂) using wearable inertial sensors offers a practical alternative to laboratory-based metabolic testing but remains challenging due to the indirect relationship between kinematics and physiological demand. This study presents a lightweight two-stage pipeline for simultaneous heel-strike (HS) detection and VO₂ estimation using a single calf-mounted IMU. In Stage 1, an Extreme Learning Machine (ELM) + Random Forest (RF) ensemble achieves the highest HS detection F1-score (0.818) under leave-one-subject-out (LOSO) validation, outperforming a temporal convolutional network (TCN) deep learning baseline (F1 = 0.674), which exhibited higher variability across subjects. In Stage 2, kinematic and gait-derived features from 30 s windows are used to estimate normalized VO₂ via RF and ensemble regression under LOSO cross-validation across 24 participants. The RF model achieves a median R² of 0.687 using predicted HS (Pred-HS) events and 0.679 using ground-truth (GT) annotations, with the ensemble showing similar performance (median R² ≈ 0.675–0.691). No statistically significant difference was observed between GT-HS and Pred-HS conditions (p > 0.05). SHAP analysis identifies accelerometer variability (acc_std) and gyroscope-derived features as dominant predictors, with demographic variables contributing minimally. Overall, the results suggest that VO₂ estimation may be achieved using automatically detected gait events without manual annotation. The proposed pipeline is computationally efficient and indicates feasibility under controlled conditions, subject to further validation. Full article

(This article belongs to the Section Biomedical Sensors)

29 pages, 15467 KB

Open AccessArticle

DB-LIO: Database-Driven LiDAR–Inertial Odometry for Memory-Bounded Persistent Mapping

by Hun-Hee Kim, Ho-Hyun Kang, Dong-Hee Noh and Hea-Min Lee

Sensors 2026, 26(10), 3061; https://doi.org/10.3390/s26103061 - 12 May 2026

Abstract

This paper proposes DB-LIO (database-driven LiDAR-inertial odometry), a simultaneous localization and mapping (SLAM) system that addresses memory scalability challenges in extended autonomous operation. Existing LiDAR-SLAM systems accumulate keyframe history in memory, leading to

O (N)

growth and out-of-memory failures during extended [...] Read more.

This paper proposes DB-LIO (database-driven LiDAR-inertial odometry), a simultaneous localization and mapping (SLAM) system that addresses memory scalability challenges in extended autonomous operation. Existing LiDAR-SLAM systems accumulate keyframe history in memory, leading to

O (N)

growth and out-of-memory failures during extended operation. To overcome this limitation, DB-LIO introduces three core design elements. First, it proposes a spatially indexed keyframe management scheme that persistently stores keyframes in SQLite with R-Tree spatial indexing, enabling

O (log N + k)

spatial queries that tightly couple cache eviction with factor-graph optimization requirements—a design that ensures every keyframe potentially involved in the next optimization cycle resides in cache. Second, it presents a four-level memory bounding architecture—SLAM-engine keyframe trimming with transparent on-demand reloading, a DB-level least recently used (LRU) cache with a spatial active window, Scan Context descriptor pool bounding, and iSAM2 sliding window compaction with a sparse global anchor graph—that collectively bounds the dominant memory consumers to

O (C)

. Third, the DB-based persistent storage enables a localization mode that can reload previously built maps—including full point clouds, six-degree-of-freedom poses, timestamps, and inter-keyframe relationships—and perform pose estimation using the stored map, which is particularly valuable for agricultural robots and other autonomous systems requiring map reuse. Experiments on a custom orchard dataset demonstrate an 81.9% reduction in memory usage compared with that of the in-memory baseline (2888 MB → 524 MB), while preserving equivalent trajectory accuracy (absolute trajectory error (ATE) root mean square error (RMSE) 0.305 ± 0.001 m vs. 0.296 m). Validation on the KITTI odometry benchmark confirms that the proposed localization mode generalizes across different LiDAR types (Livox Mid360, Velodyne HDL-64E) and environments (orchard, urban driving). Full article

(This article belongs to the Special Issue Robotic Systems for Future Farming)

39 pages, 2681 KB

Open AccessArticle

Digital Twin-Enabled Waveform Optimization for VHF Radio Communication Systems

by Chenzhe Zhong, Bo Liu, Wei Zhu, Binnian Wang, Yifan Tan and Xiangchen Wang

Sensors 2026, 26(10), 3060; https://doi.org/10.3390/s26103060 - 12 May 2026

Abstract

Very High Frequency (VHF) radio communication systems face significant challenges in modern electromagnetic environments, including spectrum congestion, dynamic interference, and varying channel conditions. Existing adaptive approaches rely on static rule-based switching or single-cycle optimization, which cannot accumulate operational experience across decision cycles. This [...] Read more.

Very High Frequency (VHF) radio communication systems face significant challenges in modern electromagnetic environments, including spectrum congestion, dynamic interference, and varying channel conditions. Existing adaptive approaches rely on static rule-based switching or single-cycle optimization, which cannot accumulate operational experience across decision cycles. This paper proposes a digital twin-enabled online learning framework (DT-MAB) for adaptive waveform selection in tactical VHF communication. The framework employs a contextual multi-armed bandit algorithm (Lin-UCB) that continuously learns the mapping from channel conditions to optimal configurations, with the digital twin serving as a virtual exploration sandbox that screens candidate configurations before physical deployment—preventing link disruptions during exploratory actions. An expanded configuration space of 63 candidates (7 waveforms × 3 MAC protocols × 3 power levels) is constructed, and a hierarchical performance evaluation model combining voice quality, bit error rate, communication delay, and transmission range is developed using the Analytic Hierarchy Process (AHP) as the reward function for online learning. Experimental results across 10 random seeds demonstrate that DT-MAB achieves the lowest mean cumulative regret, reducing regret by 29% relative to MAB without a digital twin and by 16.5% relative to PSO-based optimization on average. Ablation experiments confirm that removing virtual exploration increases performance drop events by 49% (from 250 ± 79 to 373 ± 6), demonstrating that the digital twin is a functionally indispensable component of the online learning architecture. Full article

(This article belongs to the Section Communications)

25 pages, 5460 KB

Open AccessArticle

Development of Channelized K/V Band Dicke Microwave Radiometer Based on SDR

by Zhenzhen Liang, Wei Guo, Caiyun Wang, Peng Liu and Shijie Yang

Sensors 2026, 26(10), 3059; https://doi.org/10.3390/s26103059 - 12 May 2026

Abstract

With the rapid development of software-defined radio (SDR) technology, a digital, software-reconfigurable, and flexible solution is provided for microwave radiometers, particularly suitable for atmospheric water vapor and oxygen detection with wideband, multi-channel requirements, significantly improving system efficiency. Meanwhile, digitization helps improve channel consistency [...] Read more.

With the rapid development of software-defined radio (SDR) technology, a digital, software-reconfigurable, and flexible solution is provided for microwave radiometers, particularly suitable for atmospheric water vapor and oxygen detection with wideband, multi-channel requirements, significantly improving system efficiency. Meanwhile, digitization helps improve channel consistency and address nonlinearity issues, while the digital zero-balancing mechanism implemented through adaptive integration is more suitable for digital platforms. This paper proposes a digital Dicke-type radiometer system based on an SDR platform, using Xilinx RFSoC XCZU47DR (AMD, San Jose, CA, USA) as the core hardware to achieve single-chip integration of RF signal sampling, digital local oscillator generation, and signal processing. The system implements a 46-channel channelized receiver (23 channels each for K-band and V-band) on an FPGA using a polyphase filter bank. The prototype filters achieve 70 dB stopband attenuation and 0.5 dB passband ripple, with each polyphase branch requiring only 25 coefficients, significantly reducing hardware resource consumption. An adaptive integration method is proposed, where an adaptive switch controller dynamically adjusts the hot source injection time ratio by calculating the power difference between adjacent integration periods, enabling the Dicke zero-balancing mechanism to operate entirely in the digital domain. Furthermore, a complete hardware transfer model is established for three signal branches (antenna, hot source, and matched load), and full-chain calibration of all 46 channels is performed using a liquid nitrogen cold source, with calibration reliability verified through blackbody measurements. Experimental results demonstrate brightness temperature consistency better than 0.7 K, with a sensitivity of less than 0.15 K for the K-band and less than 0.21 K for the V-band at 1 s integration time. Full article

(This article belongs to the Section Electronic Sensors)

24 pages, 8408 KB

Open AccessArticle

An Improved YOLOv9 Based Object Detection with Attention Mechanism for Personal Protective Equipment

by Geunho Lee, Jieun Lee, Tae-yong Kim and Jongpil Jeong

Sensors 2026, 26(10), 3058; https://doi.org/10.3390/s26103058 - 12 May 2026

Abstract

Industrial sites pose numerous hazards where unexpected accidents can occur at any time, and personal protective equipment (PPE) is a primary safeguard for worker safety. In this study, PPE specifically refers to safety helmets, safety shoes, and safety gloves. Manual verification of PPE [...] Read more.

Industrial sites pose numerous hazards where unexpected accidents can occur at any time, and personal protective equipment (PPE) is a primary safeguard for worker safety. In this study, PPE specifically refers to safety helmets, safety shoes, and safety gloves. Manual verification of PPE usage is infeasible in environments with many workers, motivating automated detection. This study proposes a method that integrates the Convolutional Block Attention Module (CBAM) exclusively into the training-only auxiliary reversible branch of YOLOv9’s Programmable Gradient Information (PGI) architecture. The proposed CBAMLinear module enhances gradient information during training while introducing zero additional computational overhead at inference, as the entire auxiliary branch is removed. The proposed YOLOv9 with CBAMLinear achieved consistent mAP@0.5:0.95 gains of 0.005–0.007 over the baseline for the three larger model variants, while maintaining identical inference-time parameters and FLOPs. In industrial safety, even modest performance gains can directly contribute to accident prevention by reducing false positives and false negatives, making this approach well suited for real-time safety management systems in industrial settings. Full article

(This article belongs to the Special Issue Feature Papers in Sensing and Imaging 2025&2026)

29 pages, 5543 KB

Open AccessArticle

PriorNav: Prior Knowledge Enhanced Zero-Shot Goal Navigation via Multi-Step Iterative Reasoning

by Wen Liu, Xuanshun Zhuang, Lei Ma and Zhongliang Deng

Sensors 2026, 26(10), 3057; https://doi.org/10.3390/s26103057 - 12 May 2026

Abstract

Zero-shot goal navigation requires an agent to locate targets in unseen environments based on object categories, reference images, or text descriptions, placing high demands on scene understanding and reasoning. Existing methods mainly rely on online observations, modality similarity, or heuristic graph matching, and [...] Read more.

Zero-shot goal navigation requires an agent to locate targets in unseen environments based on object categories, reference images, or text descriptions, placing high demands on scene understanding and reasoning. Existing methods mainly rely on online observations, modality similarity, or heuristic graph matching, and therefore still struggle with complex target search due to limited use of external knowledge and weak multi-step reasoning. We propose PriorNav, a prior-knowledge-enhanced framework for zero-shot goal navigation. PriorNav learns a unified retrievable knowledge space from semantic, instance, and relational knowledge, maintains a knowledge-enhanced scene graph by fusing retrieved priors with online observations, and performs progressive decision-making through multi-step iterative reasoning across exploration, verification, and approach stages. Experiments on Object-Goal, Image-Instance Goal, and Text-Goal navigation show that PriorNav improves the success rate over the baseline by 3.5%, 13.3%, and 3.5%, respectively, while also outperforming the strongest training-free baselines on all three tasks. Ablation studies further verify the effectiveness of multi-level prior knowledge, scene-graph enhancement, and iterative reasoning. These results show that combining prior knowledge with explicit reasoning is a promising direction for improving zero-shot goal navigation. Full article

(This article belongs to the Section Navigation and Positioning)

22 pages, 3939 KB

Open AccessArticle

Semantic Segmentation Method for Sparse Point Clouds Based on Straight Flow Completion and Multi-Feature Fusion

by Tong Zheng, Zhiyuan Meng, Chongchong Yu, Tao Xie and Yewang Xu

Sensors 2026, 26(10), 3056; https://doi.org/10.3390/s26103056 - 12 May 2026

Abstract

Point cloud semantic segmentation is a vital task in 3D computer vision. However, the inherent sparsity of point clouds complicates the segmentation process. In contexts such as autonomous driving, moving objects frequently exhibit motion blur, which adversely affects semantic segmentation performance. These challenges [...] Read more.

Point cloud semantic segmentation is a vital task in 3D computer vision. However, the inherent sparsity of point clouds complicates the segmentation process. In contexts such as autonomous driving, moving objects frequently exhibit motion blur, which adversely affects semantic segmentation performance. These challenges hinder the practical application of point cloud semantic segmentation. To address these issues, this paper presents a novel semantic segmentation method that integrates sparse point cloud completion with multi-feature fusion. Specifically, the study emphasizes the development of efficient strategies for constructing and training point cloud completion models, aiming to expedite model parameter training while maximizing completion accuracy. Additionally, a semantic segmentation model is introduced that combines motion feature-enhanced instance features with semantic features, thereby enhancing adaptability to moving objects. Moreover, point cloud completion and semantic segmentation are linked in an end-to-end pipeline, facilitating accurate semantic segmentation of sparse point clouds in dynamic environments. During the experimental phase, publicly available Lidar point cloud datasets, including SemanticKITTI and the millimeter-wave radar dataset RADIal, are utilized to evaluate the proposed method against classical approaches in terms of point cloud completion performance and semantic segmentation effectiveness, thereby demonstrating the reliability of the proposed method. Full article

(This article belongs to the Section Sensing and Imaging)

► Show Figures

Figure 1

19 pages, 1085 KB

Open AccessArticle

DSMUNet: A Lightweight Model for Road Crack Segmentation

by Yunqing Liu, Xu Du and Chunting Zuo

Sensors 2026, 26(10), 3055; https://doi.org/10.3390/s26103055 - 12 May 2026

Abstract

As the most common safety hazard among road surface distresses, pavement cracks require fine-grained segmentation to support subsequent maintenance decision-making. However, existing methods find it difficult to cope with complex environmental interference and multi-morphology cracks while maintaining low computational cost. To address the [...] Read more.

As the most common safety hazard among road surface distresses, pavement cracks require fine-grained segmentation to support subsequent maintenance decision-making. However, existing methods find it difficult to cope with complex environmental interference and multi-morphology cracks while maintaining low computational cost. To address the above issues, this study proposes a lightweight crack segmentation model, DSMUNet. Based on the U-shaped encoder–decoder framework, depthwise separable convolution is adopted to replace traditional convolution to reduce computational cost, and the SGE spatial group enhancement mechanism is introduced at the encoder side to emphasize crack features and suppress interference from non-crack textures. In addition, a new multi-scale feature fusion module is proposed to improve the overall connectivity of cracks during the feature fusion stage, thereby further improving model performance on the basis of model lightweighting. The experimental results show that DSMUNet contains only 0.55 M parameters and requires 21.708 GFLOPs, with an average inference latency of 5.42 ms, while achieving Dice/mIoU scores of 78.10%/69.05% on the private dataset and 87.12%/79.57% on the Crack500 dataset, respectively. These results demonstrate the advantages of this study in terms of resource consumption as well as its excellent segmentation performance, providing a more efficient implementation scheme for pavement crack segmentation in resource-constrained environments. Full article

(This article belongs to the Special Issue Advanced Sensing Technologies for Structural Health Monitoring and Nondestructive Testing)

35 pages, 13709 KB

Open AccessArticle

Detection of Microplastics and Heavy Metals Using Electronic Tongues and Machine Learning

by Luis Angel Peña, Juan P. Hoyos-Sanchez, Juan Daniel Sarmiento, Mónica V. Sandoval Rincón and Diego A. Tibaduiza

Sensors 2026, 26(10), 3054; https://doi.org/10.3390/s26103054 - 12 May 2026

Abstract

Water resources face a significant environmental challenge: pollution from microplastics (MP) and heavy metals (HM). These elements pose a dual threat to ecosystems and public health. Microplastics, defined as particles smaller than 5 mm, are of anthropogenic origin, resulting from the degradation of [...] Read more.

Water resources face a significant environmental challenge: pollution from microplastics (MP) and heavy metals (HM). These elements pose a dual threat to ecosystems and public health. Microplastics, defined as particles smaller than 5 mm, are of anthropogenic origin, resulting from the degradation of plastics by environmental factors such as solar radiation and friction with the surrounding environment, as well as from their addition to cosmetic and textile products. These materials have been widely detected in drinking water and everyday foods. Heavy metals, high-density elements

(> 5 {g / cm}^{3})

, while naturally present in the Earth’s crust, are also generated in large quantities through human activity. Their toxicological risk lies in their ability to accumulate and efficiently move through the trophic chain. Due to the risks to public health and the impacts these pose to ecosystems, it is necessary to continue seeking solutions that enable their monitoring and detection. As a contribution, this work presents a methodology for detecting microplastics and heavy metals in seawater using different machine learning models and an electronic tongue coupled to a sensor network. Two different types of heavy metals, primarily zinc (Zn) and cadmium (Cd), as well as microplastic particles composed of expanded polystyrene (EPS), were detected under controlled conditions simulating different types of water. Atomic absorption spectroscopy (AAS) confirmed the concentrations of the heavy metals studied, supporting machine-learning classification of contaminated waters. Microplastics exhibited strong metal adsorption, influenced by the physicochemical properties of the water. Overall, AUC values above

90 %

were obtained for seven different models, demonstrating the reliability of the electronic tongue in conjunction with classical machine learning techniques for detecting these elements. Full article

(This article belongs to the Section Industrial Sensors)

18 pages, 7333 KB

Open AccessArticle

Aerosol Jet Printed Ion-Selective Electrodes for Potassium Detection

by Giorgia Polidori, Emilio Sardini and Mauro Serpelloni

Sensors 2026, 26(10), 3053; https://doi.org/10.3390/s26103053 - 12 May 2026

Abstract

This work evaluates a potassium ion-selective electrode (K-ISE) fabricated using Aerosol Jet Printing (AJP) and compares its performance with that of a commercial K⁺-selective electrode (KION). Both sensors exhibit near-Nernstian behavior, with average sensitivities of 57.91 ± 5.07 mV/dec for the [...] Read more.

This work evaluates a potassium ion-selective electrode (K-ISE) fabricated using Aerosol Jet Printing (AJP) and compares its performance with that of a commercial K⁺-selective electrode (KION). Both sensors exhibit near-Nernstian behavior, with average sensitivities of 57.91 ± 5.07 mV/dec for the AJP device and 57.28 ± 5.07 mV/dec for the commercial electrode, confirming a near-Nernstian K⁺ response over the tested concentration range. Single-interferent response tests demonstrate that AJP-printed electrodes provide a more stable and less sensitive response to sodium interference (24.37 ± 1.39 mV/dec) compared to KION (33.95 ± 8.95 mV/dec), while showing comparable NH₄⁺ response and a slightly higher response toward urea than KION. Morphological analysis (OM and SEM) reveals that AJP enables smoother, more homogeneous films and improved control over the transducer/membrane interface. Unlike previous studies, this work presents a direct experimental comparison between AJP-fabricated and commercial ISEs under controlled interference conditions relevant to agricultural and environmental matrices. Although the AJP sensors exhibited near-Nernstian behavior and fast response times, their reproducibility was lower than that of the commercial electrodes (RSD = 30.12% vs. 18.45%), indicating that further optimization of the printing and membrane deposition processes is required. Full article

(This article belongs to the Special Issue Metrology, Sensors, and Instrumentation for Industry 4.0 and IoT)

► Show Figures

Graphical abstract

16 pages, 2850 KB

Open AccessArticle

Synthetic Spectrogram Augmentation via Semi-Supervised WGAN-GP for Acoustic Industrial Quality Inspection of Turbine Housings Under Extreme Data Scarcity

by Ander Gracia Moisés, Óscar Del Barrio Farran, David Martinez García and María Puy Zudaire Latienda

Sensors 2026, 26(10), 3052; https://doi.org/10.3390/s26103052 - 12 May 2026

Abstract

Impact-based acoustic inspection provides a rapid non-destructive approach for screening metallic components by analyzing the sound radiated after a controlled mechanical excitation. However, the limited availability of labeled data from defective parts remains a major challenge for deploying deep learning classifiers in production. [...] Read more.

Impact-based acoustic inspection provides a rapid non-destructive approach for screening metallic components by analyzing the sound radiated after a controlled mechanical excitation. However, the limited availability of labeled data from defective parts remains a major challenge for deploying deep learning classifiers in production. This paper proposes a complete pipeline that converts raw impact-response audio recordings into magnitude log-spectrogram images and trains a semi-supervised Wasserstein GAN with gradient penalty (SS-WGAN-GP) designed to operate under extreme data scarcity. The architecture couples a shared convolutional backbone with two output heads: a Wasserstein critic for unsupervised discrimination between real and generated samples, and a binary classification head for supervised quality labeling, jointly optimized through a combined loss that balances Wasserstein distance, gradient penalty, and cross-entropy. A key property of the design is that the generator acts as a source of synthetic training samples, producing progressively more realistic spectrograms as training advances. These samples, in turn, enrich the feature representations learned by the shared backbone and improve the performance of the classification head. The classification head of the trained discriminator is deployed directly as the quality classifier, without requiring external data or post hoc retraining. Full article

(This article belongs to the Special Issue Acoustic Sensing for Condition Monitoring)

17 pages, 23400 KB

Open AccessArticle

Prediction of Center-of-Mass Kinematics of Sensopro Exercises with Neural Network Models

by Heinz Hegi, Michael Single, Tobias Nef and Ralf Kredel

Sensors 2026, 26(10), 3051; https://doi.org/10.3390/s26103051 - 12 May 2026

Abstract

Monitoring center-of-mass is crucial for assessing postural control, but field measurements are often impractical or cost-prohibitive. This study investigates the feasibility of predicting center-of-mass kinematics from the motion of an unstable base—the Sensopro Luna—using deep learning, eliminating the need for wearable sensors. We [...] Read more.

Monitoring center-of-mass is crucial for assessing postural control, but field measurements are often impractical or cost-prohibitive. This study investigates the feasibility of predicting center-of-mass kinematics from the motion of an unstable base—the Sensopro Luna—using deep learning, eliminating the need for wearable sensors. We conducted a cross-sectional study in which 64 participants were recorded performing three coordination exercises (Single-Leg Stance, Stepping, and Waves). Marker-based motion capture and auxiliary inertial sensors were used to record reference and tape kinematics. The model inputs consisted of IMU- and motion-capture-derived tape segment orientations, IMU accelerations and angular velocities, and algorithmic estimates of the lowest tape positions. Nine axis-specific exercise models were developed using a hybrid Encoder–LSTM–Decoder architecture and compared against linear regression baselines. Our results indicate that the deep learning models successfully predicted horizontal center-of-mass displacements (DNN Mean Absolute Errors of 16.1–23.7 mm for X-axis and 4.4–31.3 mm for Y-axis) and exhibited descriptively lower errors than linear models in mean absolute error and signal morphology. However, vertical predictions were less reliable, likely due to the physical constraints inherent to the kinematics of the unstable base. Error analysis revealed that prediction accuracy was highest within common postural ranges, but decreased for extreme displacements. These findings provide a proof-of-concept for wearable-free postural monitoring, particularly for movement along the mediolateral and sagittal axes. Such a system could facilitate automated, cost-effective postural feedback and performance tracking in rehabilitation and fitness environments, supporting autonomous coordination training without the practical constraints of traditional measurement systems. Full article

(This article belongs to the Special Issue Biomechanical Analysis of Motion and Postural Control: Sensor-Based Classical and AI Methods)

27 pages, 23708 KB

Open AccessArticle

SGMR-LPR: A Semantic-Guided Network Robust to Movable Objects for LiDAR-Based Place Recognition

by Weizhong Jiang, Zhipeng Xiao, Lilin Qian, Erke Shang, Dawei Zhao, Qi Zhu and Liang Xiao

Sensors 2026, 26(10), 3050; https://doi.org/10.3390/s26103050 - 12 May 2026

Abstract

Robust LiDAR point cloud processing in dynamic outdoor environments, where movable objects such as vehicles and pedestrians introduce significant structural uncertainty, remains a key challenge for remote sensing and autonomous systems. This work addresses LiDAR-based place recognition (LPR), a critical component for loop [...] Read more.

Robust LiDAR point cloud processing in dynamic outdoor environments, where movable objects such as vehicles and pedestrians introduce significant structural uncertainty, remains a key challenge for remote sensing and autonomous systems. This work addresses LiDAR-based place recognition (LPR), a critical component for loop closure and re-localization that is highly susceptible to such dynamics. While semantic information is beneficial, existing methods often require external segmentation models at inference or lack explicit mechanisms to suppress movable objects under uncertain predictions. To address these limitations, we propose SGMR-LPR, an end-to-end semantic-guided framework designed to explicitly counteract movable-object interference during feature encoding. Building on the “segmentation-while-describing” paradigm, SGMR-LPR incorporates an internal semantic segmentation branch and two novel modules: a probabilistic movable object masking (PMOM) module, which transforms semantic logits into continuous, uncertainty-aware masks of movable regions; and a movable-suppressed channel–spatial attention (MSCS) module, which uses these masks to adaptively modulate high-level BEV features—suppressing responses from movable-object regions while enhancing stable structural elements. By embedding explicit movable-awareness into feature modulation, SGMR-LPR achieves enhanced robustness without external semantic models at inference. Extensive experiments on multiple benchmarks demonstrate consistent performance gains, particularly in scenes with dense movable objects, advancing reliable point cloud-based scene understanding in dynamic environments. Full article

(This article belongs to the Section Radar Sensors)

► Show Figures

Figure 1

39 pages, 20016 KB

Open AccessReview

Neuromorphic Technologies for Neuroengineering: From Adaptive Stimulation to SNN-Based Inference and Deployable Biointerfaces

by Zhengdi Sun, Anle Mu, Fuxiang Hao and Hang Wang

Sensors 2026, 26(10), 3049; https://doi.org/10.3390/s26103049 - 12 May 2026

Abstract

Neuromorphic technologies are attracting increasing interest in neuroengineering, as they provide an event-driven, spike-based computational framework that is well suited to temporally structured, sparse, and resource-constrained biological systems. Compared with conventional computing pipelines, neuromorphic approaches enable tighter integration of sensing, encoding, inference, feedback, [...] Read more.

Neuromorphic technologies are attracting increasing interest in neuroengineering, as they provide an event-driven, spike-based computational framework that is well suited to temporally structured, sparse, and resource-constrained biological systems. Compared with conventional computing pipelines, neuromorphic approaches enable tighter integration of sensing, encoding, inference, feedback, and actuation under low-power and low-latency conditions. These features make them particularly relevant for wearable, implantable, and other edge-native neuroengineering applications. This review examines neuromorphic neuroengineering from four closely related perspectives: neuromorphic neurostimulation and adaptive actuation; tactile and sensory biointerfaces; spiking neural network (SNN)-based biosignal processing and state decoding; and wearable or implantable neuromorphic platforms. Across these domains, we highlight how neuromorphic systems may facilitate edge-native, closed-loop architectures that operate closer to the body and respond selectively to meaningful state changes. Neurorehabilitation is further discussed as an important translational context, as it involves long-term use, multimodal sensing, adaptive intervention, and substantial real-world deployment constraints. At present, however, the evidence base remains fragmented and is still largely dominated by device demonstrations and proof-of-concept studies rather than robust translational validation. Overall, neuromorphic approaches offer a promising systems-level pathway toward neuroengineering platforms that are not only computationally efficient but also adaptive, deployable, and responsive in real-world settings. Full article

(This article belongs to the Section Biomedical Sensors)

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