A Rapid Adaptation Approach for Dynamic Air-Writing Recognition Using Wearable Wristbands with Self-Supervised Contrastive Learning DOI Creative Commons

Yunjian Guo,

Kunpeng Li, Wei Yue

et al.

Nano-Micro Letters, Journal Year: 2024, Volume and Issue: 17(1)

Published: Oct. 16, 2024

Wearable wristband systems leverage deep learning to revolutionize hand gesture recognition in daily activities. Unlike existing approaches that often focus on static gestures and require extensive labeled data, the proposed wearable with self-supervised contrastive excels at dynamic motion tracking adapts rapidly across multiple scenarios. It features a four-channel sensing array composed of an ionic hydrogel hierarchical microcone structures ultrathin flexible electrodes, resulting high-sensitivity capacitance output. Through wireless transmission from Wi-Fi module, algorithm learns latent unlabeled signals random wrist movements. Remarkably, only few-shot data are sufficient for fine-tuning model, enabling rapid adaptation various tasks. The system achieves high accuracy 94.9% different scenarios, including prediction eight-direction commands, air-writing all numbers letters. method facilitates smooth transitions between tasks without need modifying structure or undergoing task-specific training. Its utility has been further extended enhance human-machine interaction over digital platforms, such as game controls, calculators, three-language login systems, offering users natural intuitive way communication.

Language: Английский

A flexible wearable sensor based on the multiple interaction and synergistic effect of the hydrogel components with anti-freezing, low swelling for human motion detection and underwater communication DOI

Xingru Duan,

Yongheng Wang,

Tongda Lei

et al.

International Journal of Biological Macromolecules, Journal Year: 2025, Volume and Issue: 295, P. 139713 - 139713

Published: Jan. 9, 2025

Language: Английский

Citations

1
Highly Conductive, Ultratough, and Adhesive Eutectogels with Environmental Tolerance Enabled by Liquid Metal Composites DOI Open Access
Jialin Wang, Kai Zhao, Yanbo Zhao

et al.

Small, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 16, 2025

Eutectogels are recently emerged as promising alternatives to hydrogels owing their good environmental stability derived from deep eutectic solvents (DES). However, construction of competent eutectogels with both high conductivity and mechanical toughness is still difficult achieve yet highly demanded. In this work, new LMNP-PEDOT-CMC-AA (LPCA) prepared using acrylic acid (AA) carboxymethylcellulose sodium (CMC) polymeric networks, liquid metal nanoparticle-poly(3,4-ethylenedioxythiophene) (LMNP-PEDOT) added multifunctional soft fillers. Benefiting the compliant conductive LMNP-PEDOT, LPCA exhibit (0.224 S m-1), electromechanical coupling, stretchability (4228%) exceptional (7.278 MJ m-3). Additionally, abundant hydrogen interactions available functional groups within endow them excellent self-healing adhesion on different substrates. Moreover, demonstrate superior tolerance including anti-freezing anti-drying, thus allowing for long-term reliability. These appealing characteristics enable accurate stable tracking human motions, well information delivery based Morse code. This study opens possibility designing tough enabled by LM composites.

Language: Английский

Citations

0
Semi‐Interpenetrating Network Hydrogel Based on TA@MXene and Hydrogen Bonding Synergy for Pressure Sensing and Handwriting Recognition DOI Open Access
Jinlong Li, Qiuping Wu,

Peng‐Yu Zhuang

et al.

Journal of Polymer Science, Journal Year: 2025, Volume and Issue: unknown

Published: March 6, 2025

ABSTRACT To meet the diverse demands of smart electronic devices in contemporary convenience‐driven settings, development multifunctional sensors is crucial. However, producing reliable remains a significant challenge, limiting their widespread adoption across various application domains. In this study, novel hydrogel synthesized by incorporating antioxidant‐functionalized TA@MXene into polyacrylamide/carboxymethyl cellulose semi‐interpenetrating network via hydrogen bonding synergy. The exhibits remarkable properties, including outstanding stretchability (1377%), high toughness (60% compression), and strong adhesion (9.64 kPa). Additionally, it fulfills stringent requirements for wearable demonstrating excellent antifreeze performance (−20°C), broad sensing range (0%–600%), stability over 500 cycles. also shows superior monitoring human motion facial expressions, alongside exceptional pressure capabilities, accurately detecting weights from 5 to 200 g with sensitivity up 5.4 kPa −1 . Its durability validated through 100‐cycle tests. well‐suited applications, such as controlling light bulb functionality handwriting recognition, achieving recognition accuracy 93% when paired deep learning models. These results highlight hydrogel's potential applications electronics devices.

Language: Английский

Citations

0
Carbon-based flexible strain sensors: Recent advances and performance insights in human motion detection DOI Creative Commons

Ahmed I.J. Alqaderi,

Narayanan Ramakrishnan

Chemical Engineering Journal, Journal Year: 2025, Volume and Issue: unknown, P. 162609 - 162609

Published: April 1, 2025

Language: Английский

Citations

0
Synergistic Enhancement of Hole–Bridge Structure and Molecular‐Crowding Effect in Multifunctional Eutectic Hydrogel Strain/Pressure Sensor for Personal Rehabilitation Training DOI
Wenwu Wang, Zeyu Ma,

Zilu Hu

et al.

Advanced Functional Materials, Journal Year: 2025, Volume and Issue: unknown

Published: April 27, 2025

Abstract Given the electrical signal transduction capability and excellent biocompatibility, conductive hydrogels are regarded as ideal candidates for high‐performance strain/pressure sensors applied in personalized medicine. However, there challenges concurrent attainment of flexible hydrogel‐based with remarkable conductivity, sensitivity, reliable stability. Herein, a synergistic strategy based on hole–bridge structure molecular‐crowding effect is proposed to fabricate multifunctional sensor. As‐prepared eutectic hydrogel displays comprehensive performances impressive conductivity (2.81 S m −1 ), boosted mechanical robustness (a tensile strength 2.95 MPa), environmental tolerance (≈79.8% water retention at 50 °C 20 days; frost resistance = −45.3 °C). Notably, hydrogel‐derived stretchable sensor effective antibacterial ability exhibits enhanced sensitivity (gauge factor 4.49) across wide linear range, supporting monitoring joint movement electrocardiographic signals, along on‐demand photothermal treatment. As demonstration, employment efficiently conveying information high‐fidelity handwriting recognition investigated assistance machine learning. This innovative holds high promise future applications wearable‐smart devices integrated wireless transmission modules, exhibiting great potential personal rehabilitation training healthcare monitoring.

Language: Английский

Citations

0
A Rapid Adaptation Approach for Dynamic Air-Writing Recognition Using Wearable Wristbands with Self-Supervised Contrastive Learning DOI Creative Commons

Yunjian Guo,

Kunpeng Li, Wei Yue

et al.

Nano-Micro Letters, Journal Year: 2024, Volume and Issue: 17(1)

Published: Oct. 16, 2024

Wearable wristband systems leverage deep learning to revolutionize hand gesture recognition in daily activities. Unlike existing approaches that often focus on static gestures and require extensive labeled data, the proposed wearable with self-supervised contrastive excels at dynamic motion tracking adapts rapidly across multiple scenarios. It features a four-channel sensing array composed of an ionic hydrogel hierarchical microcone structures ultrathin flexible electrodes, resulting high-sensitivity capacitance output. Through wireless transmission from Wi-Fi module, algorithm learns latent unlabeled signals random wrist movements. Remarkably, only few-shot data are sufficient for fine-tuning model, enabling rapid adaptation various tasks. The system achieves high accuracy 94.9% different scenarios, including prediction eight-direction commands, air-writing all numbers letters. method facilitates smooth transitions between tasks without need modifying structure or undergoing task-specific training. Its utility has been further extended enhance human-machine interaction over digital platforms, such as game controls, calculators, three-language login systems, offering users natural intuitive way communication.

Language: Английский

Citations

3