| 121 | 0 | 5 |
| 下载次数 | 被引频次 | 阅读次数 |
传统人机交互手势识别系统多依赖接触式传感或复杂硬件,存在交互体验受限、功耗较高等问题。非接触式摩擦纳米发电机(NC-TENG)凭借其无接触响应、低功耗、结构简单的优势,成为人机交互领域手势识别传感方案的理想选择。设计了一种基于NC-TENG的人体手势识别系统,旨在通过捕捉人体手部动作的机械能实现动作分类与人机交互。系统选用电负性差异较大的铝和聚四氟乙烯(PTFE)制备NC-TENG,搭配多通道信号采集电路与PyQt5上位机软件,实现了感应电压信号的稳定采集与实时监测。针对5种动态手势的识别任务,构建基于ResNte18模型的深度残差网络模型,并采用时频联合预处理与数据增强策略,有效提升了模型的泛化能力与分类精度。在此基础上,基于Pygame框架开发了贪吃蛇游戏交互系统,通过将传感器信号实时解析为控制指令,实现了手势对游戏的无接触操控。实验结果表明,该系统对5种手势的识别准确率达96.4%,且交互响应实时,为运动康复训练及新型人机交互应用提供了低成本、高互动性的解决方案。
Abstract：Traditional human-computer interaction gesture recognition systems mostly rely on contactbased sensors or complex hardware, which suffer from limited interaction experience and high power consumption. Non-contact triboelectric nanogenerators(NC-TENGs) have emerged as an ideal sensing solution for gesture recognition in the field of human-computer interaction due to their advantages of noncontact response, low power consumption, and simple structure. A human gesture recognition system based on an NC-TENG was designed to realize motion classification and human-computer interaction by capturing the mechanical energy of human hand movements. Aluminum and polytetrafluoroethylene(PTFE) with a large electronegativity difference were selected to fabricate the NC-TENG, which was combined with a multi-channel signal acquisition circuit and a PyQt5 upper computer software to achieve stable acquisition and real-time monitoring of the induced voltage signals. For the recognition task of five types of dynamic gestures, a deep residual network model based on ResNte18 was constructed, and a time-frequency joint preprocessing and data augmentation strategy was adopted, which effectively improved the generalization ability and classification accuracy of the model. On this basis, a Snake game interaction system was developed based on the Pygame framework, and non-contact control of the game by gestures was realized by real-time parsing of sensor signals into control commands. Experimental results show that the system achieves a recognition accuracy of 96.4% for the five types of gestures with real-time interactive response, providing a low-cost and highly interactive solution for motor rehabilitation training and new human-computer interaction applications.
[1]ABDELMAKSOUD M, NABIL E, FARAG I, et al.A novel neural network method for face recognition with a single sample per person[J].IEEE Access, 2020, 8:102212-102221.
[2]ZHAO J, YAN S C, FENG J S.Towards age-invariant face recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(1):474-487.
[3]LI Y, WANG Z, LI Y, et al.Design of face recognition system based on CNN[J].Journal of Physics:Conference Series, 2020, 1601(5):052011.
[4]ANAND V, KANHANGAD V.Pore Nte:CNN-based pore descriptor for high-resolution fingerprint recognition[J].IEEE Sensors Journal, 2020, 20(16):9305-9313.
[5]BALDI P, CHAUVIN Y.Neural networks for fingerprint recognition[J].Neural Computation, 1993, 5(3):402-418.
[6]ZHAO G Q, YANG J, CHEN J, et al.Keystroke dynamics identification based on triboelectric nanogenerator for intelligent keyboard using deep learning method[J].Advanced Materials Technologies, 2019, 4(1):1800167.
[7]WANG Z L, CHEN J, LIN L.Progress in triboelectric nanogenerators as a new energy technology and self-powered sensors[J].Energy&Environmental Science, 2015, 8(8):2250-2282.
[8]ZHANG P, PAN W M, LI Z H, et al.Deep learning-assisted triboelectric sensor for complex gesture recognition[J].ACS Omega, 2025, 10(9):9381-9389.
[9]LIN H B, HE M H, JING Q S, et al.Angle-shaped triboelectric nanogenerator for harvesting environmental wind energy[J].Nano Energy, 2019, 56:269-276.
[10]WANG S W, BI M Z, CAO Z Y, et al.Linear freestanding electret generator for harvesting swinging motion energy:optimization and experiment[J].Nano Energy, 2019, 65:104013.
[11]YI F, LIN L, NIU S M, et al.Stretchable-rubber-based triboelectric nanogenerator and its application as selfpowered body motion sensors[J].Advanced Functional Materials, 2015, 25(24):3688-3696.
[12]WANG S H, XIE Y N, NIU S M, et al.Freestanding triboelectric-layer-based nanogenerators for harvesting energy from a moving object or human motion in contact and noncontact modes[J].Advanced Materials, 2014, 26(18):2818-2824.
[13]WU P F, ZHAO C X, CUI E D, et al.Advances in magnetic-assisted triboelectric nanogenerators:structures, materials and self-sensing systems[J].International Journal of Extreme Manufacturing, 2024, 6(5):052007.
[14]FU S K, HE W C, TANG Q, et al.An ultrarobust and high-performance rotational hydrodynamic triboelectric nanogenerator enabled by automatic mode switching and charge excitation[J].Advanced Materials, 2022, 34(2):2105882.
[15]LI Q Y, LIU W L, YANG H M, et al.Ultra-stability high-voltage triboelectric nanogenerator designed by ternary dielectric triboelectrification with partial soft-contact and non-contact mode[J].Nano Energy, 2021, 90:106585.
[16]LIN Z M, ZHANG B B, ZOU H Y, et al.Rationally designed rotation triboelectric nanogenerators with much extended lifetime and durability[J].Nano Energy, 2020, 68:104378.
[17]HUANG L B, BAI G X, WONG M C, et al.Magnetic-assisted noncontact triboelectric nanogenerator converting mechanical energy into electricity and light emissions[J].Advanced Materials, 2016, 28(14):2744-2751.
[18]HUANG L B, XU W, BAI G X, et al.Wind energy and blue energy harvesting based on magnetic-assisted noncontact triboelectric nanogenerator[J].Nano Energy, 2016,30:36-42.
[19]LIN Z M, ZHANG B B, XIE Y Y, et al.Elastic-connection and soft-contact triboelectric nanogenerator with superior durability and efficiency[J].Advanced Functional Materials, 2021, 31(40):2105237.
[20]LUO H, DU J Y, YANG P, et al.Human-machine interaction via dual modes of voice and gesture enabled by triboelectric nanogenerator and machine learning[J].ACS Applied Materials&Interfaces, 2023, 15(13):17009-17018.
[21]LU L J, WU J, ZHANG Y J, et al.Noncontact 3D gesture recognition enabled VR human-machine interface via electret-nanofiber-based triboelectric sensor[J].Nano Research, 2025, 18(11):94907924.
[22]吴旭,张中良,程卫东,等.基于电场传感器的人体坐姿监测[J].测控技术, 2021, 40(6):51-56.WU X, ZHANG Z L, CHENG W D, et al.Monitoring human body sitting posture by electric field sensor[J].Measurement&Control Technology, 2021, 40(6):51-56(in Chinese).
[23]WANG Z L.On Maxwell’s displacement current for energy and sensors:the origin of nanogenerators[J].Materials Today, 2017, 20(2):74-82.
[24]XU C, ZI Y L, WANG A C, et al.On the electron-transfer mechanism in the contact-electrification effect[J].Advanced Materials, 2018, 30(15):1706790.
[25]王中林.摩擦纳米发电机[M].北京:科学出版社, 2017.
[26]INNECI T, BADEM H.Detection of corneal ulcer using a genetic algorithm-based image selection and residual neural network[J].Bioengineering, 2023, 10(6):639.
[27]ZHANG X L, WEI K L, KANG X N, et al.Hybrid nonlinear convolution filters for image recognition[J].Applied Intelligence, 2021, 51(2):980-990.
[28]刘珊珊,陈枢茜,孙溢洋,等.基于Pygame的2D微型游戏引擎的开发与应用研究[J].信息与电脑(理论版),2023, 35(12):143-147.LIU S S, CHEN S X, SUN Y Y, et al.Development and application research of 2D micro game engine based on pygame[J].Information&Computer, 2023, 35(12):143-147(in Chinese).
基本信息:
DOI：10.13250/j.cnki.wndz.26060402
中图分类号:TP212.9;TM31
引用信息:
[1]乔早,武靖博,曹自平.基于非接触式摩擦电传感器的人体手势识别系统[J].微纳电子技术().DOI:10.13250/j.cnki.wndz.26060402.
2026-04-21
2026-04-21
2026-04-21