Pub Date : 2026-10-15Epub Date: 2026-09-03DOI: 10.1109/JSEN.2026.3728242
Alfred Moore;Md Rafi Ur Rahman;Eddie Jacobs;Eric Silberg;Ofer Hadar;Ronald Driggers
The area under the modulation transfer function (MTFA) is a widely used metric for evaluating resolution in optical systems. Various parameters, including diffraction, pixel pitch, and atmospheric turbulence, influence MTFA. The component with the narrowest modulation transfer function (MTF) serves as the limiting parameter and largely determines overall system resolution. One potential limiting factor is platform-induced motion blur, which is especially pronounced in moving platforms and with systems that require long exposure times, such as microbolometers. In this work, we investigate the resolution loss in drone-mounted optical sensors caused by motion blur. We evaluate this by subjecting a drone to various wind speeds and directions within a wind wall and by conducting field testing to determine the effect of linear motion. Finally, we utilize inertial measurements to characterize platform stability, allowing us to isolate the effects of rotation and displacement on overall motion blur.
{"title":"Characterization of Motion Blur in Quadcopter Uncrewed Aerial Vehicles","authors":"Alfred Moore;Md Rafi Ur Rahman;Eddie Jacobs;Eric Silberg;Ofer Hadar;Ronald Driggers","doi":"10.1109/JSEN.2026.3728242","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3728242","url":null,"abstract":"The area under the modulation transfer function (MTFA) is a widely used metric for evaluating resolution in optical systems. Various parameters, including diffraction, pixel pitch, and atmospheric turbulence, influence MTFA. The component with the narrowest modulation transfer function (MTF) serves as the limiting parameter and largely determines overall system resolution. One potential limiting factor is platform-induced motion blur, which is especially pronounced in moving platforms and with systems that require long exposure times, such as microbolometers. In this work, we investigate the resolution loss in drone-mounted optical sensors caused by motion blur. We evaluate this by subjecting a drone to various wind speeds and directions within a wind wall and by conducting field testing to determine the effect of linear motion. Finally, we utilize inertial measurements to characterize platform stability, allowing us to isolate the effects of rotation and displacement on overall motion blur.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 20","pages":"30568-30580"},"PeriodicalIF":4.5,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-10-15Epub Date: 2026-09-01DOI: 10.1109/JSEN.2026.3727864
Yiman Huang;Xinlin Zhang;Xiaotong Zhang
Low-field magnetic resonance imaging (MRI) provides a promising route toward portable and cost-effective imaging, but operation without a radio frequency (RF) shielding room is highly vulnerable to electromagnetic interference (EMI). Existing active EMI elimination methods may require dedicated EMI characterization data or supervised training-data construction using clean MR components and synthetic contaminated-clean pairs. In this work, we propose RACE, a calibration-free self-supervised EMI elimination method for RF shielding-free low-field MRI based on repeated-acquisition consistency. The method exploits the fact that repeated measurements of the same k-space readout have highly consistent underlying MR signals when acquired within a short time interval, whereas their EMI components and random noise may differ. This physical consistency is used to construct a cross-acquisition self-supervised loss, in which the EMI-suppressed signal estimated from one acquisition is cross-validated using the contaminated receive-coil signal and sensing-coil data from the other acquisition. RACE therefore learns directly from EMI-contaminated repeated acquisitions and simultaneously acquired EMI sensing signals, without requiring dedicated EMI characterization data, clean MR labels, or synthetic contaminated-clean training pairs. Reference-based simulation, phantom, and human brain experiments demonstrated that RACE effectively reduced EMI artifacts and improved image quality in RF shielding-free low-field MRI. Compared with External Dynamic InTerference Estimation and Removal (EDITER), RACE achieved higher signal-to-noise ratio (SNR) and noise removal percentage (NRP), while providing performance comparable to Deep learning Direct MR Signal Prediction (Deep-DSP) across the evaluated datasets. These results suggest that repeated-acquisition consistency can serve as an effective self-supervised constraint for practical EMI elimination in low-field MRI.
{"title":"Calibration-Free Self-Supervised Electromagnetic Interference Elimination for RF Shielding-Free Low-Field MRI Using Repeated-Acquisition Consistency","authors":"Yiman Huang;Xinlin Zhang;Xiaotong Zhang","doi":"10.1109/JSEN.2026.3727864","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3727864","url":null,"abstract":"Low-field magnetic resonance imaging (MRI) provides a promising route toward portable and cost-effective imaging, but operation without a radio frequency (RF) shielding room is highly vulnerable to electromagnetic interference (EMI). Existing active EMI elimination methods may require dedicated EMI characterization data or supervised training-data construction using clean MR components and synthetic contaminated-clean pairs. In this work, we propose RACE, a calibration-free self-supervised EMI elimination method for RF shielding-free low-field MRI based on repeated-acquisition consistency. The method exploits the fact that repeated measurements of the same k-space readout have highly consistent underlying MR signals when acquired within a short time interval, whereas their EMI components and random noise may differ. This physical consistency is used to construct a cross-acquisition self-supervised loss, in which the EMI-suppressed signal estimated from one acquisition is cross-validated using the contaminated receive-coil signal and sensing-coil data from the other acquisition. RACE therefore learns directly from EMI-contaminated repeated acquisitions and simultaneously acquired EMI sensing signals, without requiring dedicated EMI characterization data, clean MR labels, or synthetic contaminated-clean training pairs. Reference-based simulation, phantom, and human brain experiments demonstrated that RACE effectively reduced EMI artifacts and improved image quality in RF shielding-free low-field MRI. Compared with External Dynamic InTerference Estimation and Removal (EDITER), RACE achieved higher signal-to-noise ratio (SNR) and noise removal percentage (NRP), while providing performance comparable to Deep learning Direct MR Signal Prediction (Deep-DSP) across the evaluated datasets. These results suggest that repeated-acquisition consistency can serve as an effective self-supervised constraint for practical EMI elimination in low-field MRI.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 20","pages":"30921-30931"},"PeriodicalIF":4.5,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176015","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Visual simultaneous localization and mapping (SLAM) is a popular positioning method for robots, which has been successfully applied to a variety of tasks such as grasping and unmanned inspection. However, most existing visual SLAM methods suffer from low accuracy, poor robustness, and positioning failure in challenging environments such as texture-less or dim scenes. To address these challenges, we present EvBrainSLAM in this study, a novel brain-inspired visual SLAM that integrates event camera and inertial measurement unit (IMU) for robust pose estimation. EvBrainSLAM converts asynchronous event streams into self-motion and local-view cues that can be effectively integrated by a continuous attractor neural network (CANN)-based navigation framework. Specifically, the frontend combines motion-compensated event frames with inertial measurements to estimate local motion through sliding-window nonlinear optimization, while the CANN-based backend performs attractor-based path integration, maintains a continuous pose representation, and suppresses localization drift caused by noisy or degraded measurements. Local-view information derived from event-intensity templates is further used to reactivate previously learned pose cell states and provide loop-closure corrections through a brain-inspired cognitive map, thereby improving global trajectory consistency. By combining the high temporal resolution and wide dynamic range of event cameras with robust attractor-based state estimation, EvBrainSLAM achieves reliable localization under rapid motion, low illumination, and texture-less conditions. Experimental results on nine challenging sequences demonstrate that EvBrainSLAM consistently outperforms the evaluated baseline methods in terms of localization accuracy and robustness.
{"title":"Brain-Inspired Simultaneous Localization and Mapping via Integrating Event Camera and IMU","authors":"Jiarui Dou;Hui Zhao;Fangwen Yu;Xianlei Long;Yan Li;Chao Chen;Xiangwei Zhu;Fuqiang Gu","doi":"10.1109/JSEN.2026.3728853","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3728853","url":null,"abstract":"Visual simultaneous localization and mapping (SLAM) is a popular positioning method for robots, which has been successfully applied to a variety of tasks such as grasping and unmanned inspection. However, most existing visual SLAM methods suffer from low accuracy, poor robustness, and positioning failure in challenging environments such as texture-less or dim scenes. To address these challenges, we present EvBrainSLAM in this study, a novel brain-inspired visual SLAM that integrates event camera and inertial measurement unit (IMU) for robust pose estimation. EvBrainSLAM converts asynchronous event streams into self-motion and local-view cues that can be effectively integrated by a continuous attractor neural network (CANN)-based navigation framework. Specifically, the frontend combines motion-compensated event frames with inertial measurements to estimate local motion through sliding-window nonlinear optimization, while the CANN-based backend performs attractor-based path integration, maintains a continuous pose representation, and suppresses localization drift caused by noisy or degraded measurements. Local-view information derived from event-intensity templates is further used to reactivate previously learned pose cell states and provide loop-closure corrections through a brain-inspired cognitive map, thereby improving global trajectory consistency. By combining the high temporal resolution and wide dynamic range of event cameras with robust attractor-based state estimation, EvBrainSLAM achieves reliable localization under rapid motion, low illumination, and texture-less conditions. Experimental results on nine challenging sequences demonstrate that EvBrainSLAM consistently outperforms the evaluated baseline methods in terms of localization accuracy and robustness.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 20","pages":"31664-31676"},"PeriodicalIF":4.5,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176020","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-10-15Epub Date: 2026-09-03DOI: 10.1109/JSEN.2026.3728523
Zijie Gu;Xukai Ding;Feng Yang;Liye Zhao;Hongsheng Li
The phase delay in the readout and actuation electronics introduces loop coupling and angle-dependent output ripple in whole-angle (WA) microhemispherical resonator gyroscopes ($mu $ HRGs). This article presents an asymmetric phase-delay model and an online compensation method based on alternating virtual rotation. The readout and actuation delays are described by loop-equivalent average-delay and X/Y mismatch terms. These terms clarify how phase errors propagate into the quadrature control loop and angular-rate output. Under forward and reverse virtual rotations, the equivalent average delay and quadrature-loop mismatch are extracted from the DC and cosine harmonic components of the quadrature control voltage without additional hardware. A single-chain time-division readout-and-actuation architecture is used to suppress hardware-induced interchannel mismatch, and the estimated average delay is compensated by an equivalent readout-side phase rotation. The simulations verify the extraction formula under injected phase mismatch and additional loop-equivalent errors. Experiments show that the estimated equivalent average phase delay is reduced from ${3}.{1941}^{circ } $ to ${0}.{0008}^{circ } $ after compensation. Phase-only compensation reduces the overall rms angular-rate fluctuation from ${0}.{1684}~text {deg/ {s}}$ to ${0}.{1350}~text {deg/ {s}}$ . With complete compensation, the overall rms fluctuation is further reduced to ${0}.{0164}~text {deg/ {s}}$ . The results validate the proposed model and demonstrate its applicability to online phase-delay compensation in WA $mu $ HRGs.
{"title":"Asymmetric Phase-Delay Modeling and Online Compensation for Whole-Angle Microhemispherical Resonator Gyroscopes","authors":"Zijie Gu;Xukai Ding;Feng Yang;Liye Zhao;Hongsheng Li","doi":"10.1109/JSEN.2026.3728523","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3728523","url":null,"abstract":"The phase delay in the readout and actuation electronics introduces loop coupling and angle-dependent output ripple in whole-angle (WA) microhemispherical resonator gyroscopes (<inline-formula> <tex-math>$mu $ </tex-math></inline-formula>HRGs). This article presents an asymmetric phase-delay model and an online compensation method based on alternating virtual rotation. The readout and actuation delays are described by loop-equivalent average-delay and X/Y mismatch terms. These terms clarify how phase errors propagate into the quadrature control loop and angular-rate output. Under forward and reverse virtual rotations, the equivalent average delay and quadrature-loop mismatch are extracted from the DC and cosine harmonic components of the quadrature control voltage without additional hardware. A single-chain time-division readout-and-actuation architecture is used to suppress hardware-induced interchannel mismatch, and the estimated average delay is compensated by an equivalent readout-side phase rotation. The simulations verify the extraction formula under injected phase mismatch and additional loop-equivalent errors. Experiments show that the estimated equivalent average phase delay is reduced from <inline-formula> <tex-math>${3}.{1941}^{circ } $ </tex-math></inline-formula> to <inline-formula> <tex-math>${0}.{0008}^{circ } $ </tex-math></inline-formula> after compensation. Phase-only compensation reduces the overall rms angular-rate fluctuation from <inline-formula> <tex-math>${0}.{1684}~text {deg/ {s}}$ </tex-math></inline-formula> to <inline-formula> <tex-math>${0}.{1350}~text {deg/ {s}}$ </tex-math></inline-formula>. With complete compensation, the overall rms fluctuation is further reduced to <inline-formula> <tex-math>${0}.{0164}~text {deg/ {s}}$ </tex-math></inline-formula>. The results validate the proposed model and demonstrate its applicability to online phase-delay compensation in WA <inline-formula> <tex-math>$mu $ </tex-math></inline-formula>HRGs.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 20","pages":"30205-30216"},"PeriodicalIF":4.5,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176081","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-10-15Epub Date: 2026-09-03DOI: 10.1109/JSEN.2026.3728692
Andrea Petroni;Gaetano Scarano;Roberto Cusani;Mauro Biagi
Indoor positioning systems represent a fundamental tool for enhancing efficiency, safety, and user experience in a wide range of applications requiring location awareness. Specifically, privacy concerns and the need for unobtrusive monitoring in sensitive environments, such as eldercare or child safety, motivate the development of nonvisual indoor tracking systems. This article proposes an indoor infrared (IR) tomographic framework that estimates and tracks human presence by exploiting signal occlusions between spatially distributed light-emitting diode (LED)–photodetector (PD) links. By modeling IR shadowing as a function of voxel-level occupancy, the system reconstructs 3-D spatial maps without relying on cameras or identity-revealing sensors. A regularized least-squares approach is then employed to obtain robust static occupancy estimates. Furthermore, the proposed framework is extended to dynamic scenarios, where Kalman filtering is employed to achieve reliable tracking of subjects from the reconstructed spatial centroids. Numerical results demonstrate subdecimeter localization accuracy, scalability with sensor density, and improved robustness with respect to representative indoor positioning benchmarks. Moreover, a small-scale hardware prototype experimentally validates the core static tomographic reconstruction stage using multiple 3-D objects with different geometries.
{"title":"Tomography-Inspired Infrared Indoor Localization and Tracking Based on Optical Shadow Imaging","authors":"Andrea Petroni;Gaetano Scarano;Roberto Cusani;Mauro Biagi","doi":"10.1109/JSEN.2026.3728692","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3728692","url":null,"abstract":"Indoor positioning systems represent a fundamental tool for enhancing efficiency, safety, and user experience in a wide range of applications requiring location awareness. Specifically, privacy concerns and the need for unobtrusive monitoring in sensitive environments, such as eldercare or child safety, motivate the development of nonvisual indoor tracking systems. This article proposes an indoor infrared (IR) tomographic framework that estimates and tracks human presence by exploiting signal occlusions between spatially distributed light-emitting diode (LED)–photodetector (PD) links. By modeling IR shadowing as a function of voxel-level occupancy, the system reconstructs 3-D spatial maps without relying on cameras or identity-revealing sensors. A regularized least-squares approach is then employed to obtain robust static occupancy estimates. Furthermore, the proposed framework is extended to dynamic scenarios, where Kalman filtering is employed to achieve reliable tracking of subjects from the reconstructed spatial centroids. Numerical results demonstrate subdecimeter localization accuracy, scalability with sensor density, and improved robustness with respect to representative indoor positioning benchmarks. Moreover, a small-scale hardware prototype experimentally validates the core static tomographic reconstruction stage using multiple 3-D objects with different geometries.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 20","pages":"30482-30494"},"PeriodicalIF":4.5,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176115","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-10-15Epub Date: 2026-08-24DOI: 10.1109/JSEN.2026.3724932
Lilia Tightiz;L. Minh Dang;Sanjeevikumar Padmanaban;Hyosik Yang
In the era of smart grids, the use of digital technologies has introduced unprecedented cybersecurity challenges, particularly in safeguarding against false data injection attacks (FDIAs) that may potentially manipulate power system measurements and disrupt grid operation. This article proposes an informer-based FDIA detection framework with multiobjective optimization, a nondominated sorting genetic algorithm II (NSGA-II), and adversarial defense techniques to maximize the attack detection rate (ADR), minimize the false positive rate (FPR), and reduce computational overhead. Leveraging ProbSparse self-attention, the informer efficiently captures long-range temporal dependencies with low complexity, making it well-suited for real-time anomaly detection in power-system time series. Experimental evaluation on IEEE 5-bus, 14-bus, and 118-bus systems demonstrates the robustness and scalability of the proposed approach. Feature selection optimization, evaluated relative to the same Informer detector trained without feature selection, increased the ADR by 12.5 percentage points and reduced the FPR by 9 percentage points, confirming that redundant features should be omitted. Specifically, Bayesian hyperparameter optimization reduces informer training time from 188.7 to 145.2 s (22.0%), while subsequent NSGA-II-based multiobjective optimization further improves ADR from 87.3% to 95.1% under identical training conditions. The proposed defense mechanism is robust against fast gradient sign method (FGSM), basic iterative method (BIM), and momentum iterative method (MIM) attacks, and it outperforms a conventional autoencoder-based detector. Moreover, the proposed detector consistently outperforms state-of-the-art transformer time-series baselines (TimesNet, PatchTST, and FEDformer) under the same experimental protocol. Real-time feasibility experiments further show that the informer achieves an inference latency of under 6 ms per sample and a maximum throughput of 2100 samples per second, supporting deployment in large-scale smart grid monitoring.
{"title":"Enhancing Smart Grid Cybersecurity: Multiobjective Optimization of Informer-Based FDIA Detection With Adversarial Defense Mechanisms","authors":"Lilia Tightiz;L. Minh Dang;Sanjeevikumar Padmanaban;Hyosik Yang","doi":"10.1109/JSEN.2026.3724932","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3724932","url":null,"abstract":"In the era of smart grids, the use of digital technologies has introduced unprecedented cybersecurity challenges, particularly in safeguarding against false data injection attacks (FDIAs) that may potentially manipulate power system measurements and disrupt grid operation. This article proposes an informer-based FDIA detection framework with multiobjective optimization, a nondominated sorting genetic algorithm II (NSGA-II), and adversarial defense techniques to maximize the attack detection rate (ADR), minimize the false positive rate (FPR), and reduce computational overhead. Leveraging ProbSparse self-attention, the informer efficiently captures long-range temporal dependencies with low complexity, making it well-suited for real-time anomaly detection in power-system time series. Experimental evaluation on IEEE 5-bus, 14-bus, and 118-bus systems demonstrates the robustness and scalability of the proposed approach. Feature selection optimization, evaluated relative to the same Informer detector trained without feature selection, increased the ADR by 12.5 percentage points and reduced the FPR by 9 percentage points, confirming that redundant features should be omitted. Specifically, Bayesian hyperparameter optimization reduces informer training time from 188.7 to 145.2 s (22.0%), while subsequent NSGA-II-based multiobjective optimization further improves ADR from 87.3% to 95.1% under identical training conditions. The proposed defense mechanism is robust against fast gradient sign method (FGSM), basic iterative method (BIM), and momentum iterative method (MIM) attacks, and it outperforms a conventional autoencoder-based detector. Moreover, the proposed detector consistently outperforms state-of-the-art transformer time-series baselines (TimesNet, PatchTST, and FEDformer) under the same experimental protocol. Real-time feasibility experiments further show that the informer achieves an inference latency of under 6 ms per sample and a maximum throughput of 2100 samples per second, supporting deployment in large-scale smart grid monitoring.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 20","pages":"30746-30764"},"PeriodicalIF":4.5,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176146","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-10-15Epub Date: 2026-09-04DOI: 10.1109/JSEN.2026.3729077
Laura Duarte;Michele Polito;Laura Gastaldi;Stefano Pastorelli;Pedro Neto
To establish a reliable human–robot collaboration (HRC), the collaborative robot needs to be aware of the environment and the human actions within it. In particular, the need to understand upper body movements and hand gestures performed by humans has led to the development of multiple modalities of motion capture (MoCap) systems. However, current MoCap systems exhibit significant limitations, including vulnerability to occlusion, reduced accuracy during rapid movements, and high system latency. This work proposes a tracking method, inspired by the continuously adaptive mean-shift (CAM-Shift) tracking algorithm, to track hand movements through the use of an event camera. A detection algorithm is implemented to allow the system to recover when the tracking algorithm fails due to occlusions. The proposed tracker is benchmarked against two state-of-the-art event-based tracking approaches, achieving comparable tracking accuracy while maintaining a median processing time of 0.1 ms per event packet. Tracking and detection experiments demonstrate that the proposed method can effectively handle occlusions by switching between tracking and detection modes. In online experiments, the tracking error was quantified using the Hausdorff distance and the average minimum distance, yielding values of 11.54 pixels (1.14 cm) and 2.32 pixels (0.23 cm), respectively.
{"title":"Real-Time Human Hand Tracking and Detection From Event Camera Data","authors":"Laura Duarte;Michele Polito;Laura Gastaldi;Stefano Pastorelli;Pedro Neto","doi":"10.1109/JSEN.2026.3729077","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3729077","url":null,"abstract":"To establish a reliable human–robot collaboration (HRC), the collaborative robot needs to be aware of the environment and the human actions within it. In particular, the need to understand upper body movements and hand gestures performed by humans has led to the development of multiple modalities of motion capture (MoCap) systems. However, current MoCap systems exhibit significant limitations, including vulnerability to occlusion, reduced accuracy during rapid movements, and high system latency. This work proposes a tracking method, inspired by the continuously adaptive mean-shift (CAM-Shift) tracking algorithm, to track hand movements through the use of an event camera. A detection algorithm is implemented to allow the system to recover when the tracking algorithm fails due to occlusions. The proposed tracker is benchmarked against two state-of-the-art event-based tracking approaches, achieving comparable tracking accuracy while maintaining a median processing time of 0.1 ms per event packet. Tracking and detection experiments demonstrate that the proposed method can effectively handle occlusions by switching between tracking and detection modes. In online experiments, the tracking error was quantified using the Hausdorff distance and the average minimum distance, yielding values of 11.54 pixels (1.14 cm) and 2.32 pixels (0.23 cm), respectively.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 20","pages":"31608-31617"},"PeriodicalIF":4.5,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176147","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-10-15Epub Date: 2026-09-02DOI: 10.1109/JSEN.2026.3728294
Soumya Kanti Raj;Aritra Acharyya;Rajashree Das;Angsuman Sarkar;Amitava Bandyopadhyay
A flexible intensity-modulated vector bend sensor with a single TiO${}_{2} sim $ gel-wax nanocomposite (NC)-based core and two multiphotodetector-array-assisted radiation-loss measurement is presented. The sensor employs a rectangular macro-optical waveguide with silicone-rubber (SR) cladding, a 570-nm source, a main photodetector (MD) for bend-magnitude estimation, and two auxiliary photodetector arrays for bend-direction discrimination through differential radiation-loss monitoring. An analytical model describing guided-power attenuation and bend-induced radiation loss is developed and experimentally validated. The MD exhibits bend sensitivity exceeding 0.50 dB$cdot $ m within the curvature range of 1.33–10.0 m${}^{-{1}}$ . Reliability analysis up to 10 000 repeated bending cycles reveals a maximum photocurrent deviation of only 3.10%, with nearly zero hysteresis. Mean repeatability and reproducibility errors remain within 0.12%–0.29% and 0.42%–0.59%, respectively. Thermal characterization over $- 10~^{circ }$ C to $70~^{circ }$ C shows mean photocurrent deviations below 8%, whereas over 30%–80% ambient relative humidity, the sensor demonstrates negligible influence owing to airtight encapsulation. Proof-of-concept soft-robotic deformation monitoring over a 1–100-cm bend-radius range shows excellent agreement with independent 2-D laser-profiler measurements, while dynamic characterization yields response and recovery times of 0.195 and 0.218 s, respectively, and an effective −3-dB bandwidth of 1.740 Hz. Continuous cyclic measurements further demonstrate dynamic bend-radius tracking and correct discrimination of the bending directions.
{"title":"Vector Bend Sensing Using Single-Core Optical Waveguide and Dual Photodetector-Array-Based Radiation-Loss Measurement","authors":"Soumya Kanti Raj;Aritra Acharyya;Rajashree Das;Angsuman Sarkar;Amitava Bandyopadhyay","doi":"10.1109/JSEN.2026.3728294","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3728294","url":null,"abstract":"A flexible intensity-modulated vector bend sensor with a single TiO<inline-formula> <tex-math>${}_{2} sim $ </tex-math></inline-formula>gel-wax nanocomposite (NC)-based core and two multiphotodetector-array-assisted radiation-loss measurement is presented. The sensor employs a rectangular macro-optical waveguide with silicone-rubber (SR) cladding, a 570-nm source, a main photodetector (MD) for bend-magnitude estimation, and two auxiliary photodetector arrays for bend-direction discrimination through differential radiation-loss monitoring. An analytical model describing guided-power attenuation and bend-induced radiation loss is developed and experimentally validated. The MD exhibits bend sensitivity exceeding 0.50 dB<inline-formula> <tex-math>$cdot $ </tex-math></inline-formula>m within the curvature range of 1.33–10.0 m<inline-formula> <tex-math>${}^{-{1}}$ </tex-math></inline-formula>. Reliability analysis up to 10 000 repeated bending cycles reveals a maximum photocurrent deviation of only 3.10%, with nearly zero hysteresis. Mean repeatability and reproducibility errors remain within 0.12%–0.29% and 0.42%–0.59%, respectively. Thermal characterization over <inline-formula> <tex-math>$- 10~^{circ }$ </tex-math></inline-formula>C to <inline-formula> <tex-math>$70~^{circ }$ </tex-math></inline-formula>C shows mean photocurrent deviations below 8%, whereas over 30%–80% ambient relative humidity, the sensor demonstrates negligible influence owing to airtight encapsulation. Proof-of-concept soft-robotic deformation monitoring over a 1–100-cm bend-radius range shows excellent agreement with independent 2-D laser-profiler measurements, while dynamic characterization yields response and recovery times of 0.195 and 0.218 s, respectively, and an effective −3-dB bandwidth of 1.740 Hz. Continuous cyclic measurements further demonstrate dynamic bend-radius tracking and correct discrimination of the bending directions.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 20","pages":"30463-30472"},"PeriodicalIF":4.5,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176165","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-10-15Epub Date: 2026-09-11DOI: 10.1109/JSEN.2026.3730834
Yuxin Liu;Wen Li;Dongyan Wei;Ge Shen
Accurate compensation of magnetic interference generated by moving platforms is essential for reliable geomagnetic navigation. Conventional Tolles–Lawson (T–L) methods require dedicated high-altitude maneuvers under a clean magnetic background and bandpass filtering, which introduce systematic errors and are impractical for platforms lacking maneuverability. To address these limitations, this article proposes a hierarchical compensation framework that synergistically combines physical model calibration with data-driven refinement. First, a Helmholtz coil generates a known true field, enabling direct extraction of platform interference without filtering. An Adaptive transient-aliasing-outlier (ATAO) suppression algorithm automatically removes coil transient effects, performs antialiasing decimation, and eliminates outliers. A truth-guided ridge regression (TGRR) estimator exploits the known true field to select the optimal regularization parameter deterministically, yielding an unbiased and stable T–L parameters. The remaining nonlinear residuals—which the linear T–L model cannot capture—are then compensated by a deep neural network (DNN) with wavelet decomposition and an attention mechanism. Experimental results on an uncrewed ground vehicle demonstrate that coil-based calibration improves parameter accuracy by approximately 48.0% compared to conventional methods. The complete hierarchical method achieves final compensation errors of $16.55~pm ~1.69$ nT (Route 1, training), and $20.24~pm ~1.12$ nT, $3.62~pm ~0.29$ nT, and $40.13~pm ~3.13$ nT (Routes 2–4, testing), respectively. These results verify the performance and generalization ability of the proposed method.
{"title":"Compensation of Platform Magnetic Interference via Coil-Calibrated Physical Model and Attention Network","authors":"Yuxin Liu;Wen Li;Dongyan Wei;Ge Shen","doi":"10.1109/JSEN.2026.3730834","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3730834","url":null,"abstract":"Accurate compensation of magnetic interference generated by moving platforms is essential for reliable geomagnetic navigation. Conventional Tolles–Lawson (T–L) methods require dedicated high-altitude maneuvers under a clean magnetic background and bandpass filtering, which introduce systematic errors and are impractical for platforms lacking maneuverability. To address these limitations, this article proposes a hierarchical compensation framework that synergistically combines physical model calibration with data-driven refinement. First, a Helmholtz coil generates a known true field, enabling direct extraction of platform interference without filtering. An Adaptive transient-aliasing-outlier (ATAO) suppression algorithm automatically removes coil transient effects, performs antialiasing decimation, and eliminates outliers. A truth-guided ridge regression (TGRR) estimator exploits the known true field to select the optimal regularization parameter deterministically, yielding an unbiased and stable T–L parameters. The remaining nonlinear residuals—which the linear T–L model cannot capture—are then compensated by a deep neural network (DNN) with wavelet decomposition and an attention mechanism. Experimental results on an uncrewed ground vehicle demonstrate that coil-based calibration improves parameter accuracy by approximately 48.0% compared to conventional methods. The complete hierarchical method achieves final compensation errors of <inline-formula> <tex-math>$16.55~pm ~1.69$ </tex-math></inline-formula> nT (Route 1, training), and <inline-formula> <tex-math>$20.24~pm ~1.12$ </tex-math></inline-formula> nT, <inline-formula> <tex-math>$3.62~pm ~0.29$ </tex-math></inline-formula> nT, and <inline-formula> <tex-math>$40.13~pm ~3.13$ </tex-math></inline-formula> nT (Routes 2–4, testing), respectively. These results verify the performance and generalization ability of the proposed method.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 20","pages":"30282-30292"},"PeriodicalIF":4.5,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176196","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-10-15Epub Date: 2026-09-01DOI: 10.1109/JSEN.2026.3727664
V. Lavorgna;M. Pulcinelli;A. Polimadei;R. D'Amato;C. Massaroni;M. A. Caponero;E. Schena;D. Lo Presti
Distributed optical fiber sensing (DOFS) combines the advantages of fiber-optic sensors (FOSs), including flexibility, small size, immunity to electromagnetic interference, and high metrological performance, with the capability to transform a single optical fiber into a continuous sensing element for spatially resolved mechanical measurements. Optical frequency domain reflectometry (OFDR), based on Rayleigh backscattering, enables high-spatial-resolution DOFS measurements, broadening the range of potential sensing applications. However, OFDR-based DOFS remains largely unexplored for biomedical applications, despite the need for sensitive, spatially resolved, and conformable sensing interfaces. This study presents a soft DOFS-based mat (DM) as a large-area interface for physiological monitoring. A single-mode optical fiber was embedded in a flexible silicone matrix and arranged in a serpentine layout to distribute sensing over the mat surface. With a gauge pitch of 2.6 mm, the system provided 2250 sensing sites across the active area at a sampling frequency of 50 Hz. The mat was assessed on six healthy volunteers in a seated nearable configuration on the backrest of a standard office chair. The distributed output enabled 2-D mapping of the mat response, reflecting back–mat mechanical coupling and cardiorespiratory-induced perturbations. Respiratory rate (RR) and heart rate (HR) were therefore estimated and compared with a reference wearable system. The maps revealed physiologically coherent spatial and temporal patterns, while the estimated rates showed good agreement with the reference measurements. These results demonstrate the feasibility of combining large-area distributed sensing, spatial mapping, and quantitative cardiorespiratory monitoring within a DOFS-based soft nearable interface.
{"title":"A Nearable Soft Mat Based on Distributed Optical Fiber Sensing for Physiological Monitoring","authors":"V. Lavorgna;M. Pulcinelli;A. Polimadei;R. D'Amato;C. Massaroni;M. A. Caponero;E. Schena;D. Lo Presti","doi":"10.1109/JSEN.2026.3727664","DOIUrl":"https://doi.org/10.1109/JSEN.2026.3727664","url":null,"abstract":"Distributed optical fiber sensing (DOFS) combines the advantages of fiber-optic sensors (FOSs), including flexibility, small size, immunity to electromagnetic interference, and high metrological performance, with the capability to transform a single optical fiber into a continuous sensing element for spatially resolved mechanical measurements. Optical frequency domain reflectometry (OFDR), based on Rayleigh backscattering, enables high-spatial-resolution DOFS measurements, broadening the range of potential sensing applications. However, OFDR-based DOFS remains largely unexplored for biomedical applications, despite the need for sensitive, spatially resolved, and conformable sensing interfaces. This study presents a soft DOFS-based mat (DM) as a large-area interface for physiological monitoring. A single-mode optical fiber was embedded in a flexible silicone matrix and arranged in a serpentine layout to distribute sensing over the mat surface. With a gauge pitch of 2.6 mm, the system provided 2250 sensing sites across the active area at a sampling frequency of 50 Hz. The mat was assessed on six healthy volunteers in a seated nearable configuration on the backrest of a standard office chair. The distributed output enabled 2-D mapping of the mat response, reflecting back–mat mechanical coupling and cardiorespiratory-induced perturbations. Respiratory rate (RR) and heart rate (HR) were therefore estimated and compared with a reference wearable system. The maps revealed physiologically coherent spatial and temporal patterns, while the estimated rates showed good agreement with the reference measurements. These results demonstrate the feasibility of combining large-area distributed sensing, spatial mapping, and quantitative cardiorespiratory monitoring within a DOFS-based soft nearable interface.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"26 20","pages":"30429-30437"},"PeriodicalIF":4.5,"publicationDate":"2026-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11674370","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149175992","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：