基于聚丙烯材料的传感器制备及在动态手势信号检测中的应用研究

随着人机交互技术的不断发展，手部动作成为了一种自然、直观的交 互方式，目前研究中大多主要利用的是静态动作时的信号，忽略了动态动 作时的识别，会导致动作识别鲁棒性差等问题。由于动态手势具有时空复 杂、变化多样、易受环境干扰的特点，使用目前常用的传感器在捕捉动态 手势方面尚显不足，这对传感器提出了更高的要求。所以需要开发一款响 应速度快、灵敏度高、轻量小巧的传感器以适配动态手势的研究。 针对动态手势特点，本文选取多孔聚丙烯材料来制备压电驻极体薄膜， 这种薄膜具有小巧轻量、良好的机械柔性、成本低的特点。本课题中，制 备 的 压 电 驻 极 体 传 感 器 结 合 了 肌 电 电 极 ， 设 计 并 搭 建 了 一 套 肌 电 信 号 （Electromyography，EMG）和力肌图谱信号（Force Myography，FMG） 复合采集系统检测动态手势信号。 本文利用模式识别方法对采集到的 EMG 和 FMG 进行处理，实现动态 手势的精准识别。在数据预处理阶段，对两种信号进行了滤波的初步处理， 然后进一步利用 FastICA 去除了 FMG 信号中的干扰信号。在特征提取阶段， 首先，提出了采用能量检测的办法来进行起始点和终止点的自动检测和活 动段的自动划分；然后进行特征重要度的分析，选取了四个重要程度高的 特征进行下一步的研究。在手势识别阶段，比较了单一模态下线性判别分 析和支持向量机算法两种分类算法的动态手势时的表现，后续在信息层进 行融合，并将融合的手势分类结果与单一模态比较。 结果表明：两种算法解码下，FMG 信号的表现都要优于 EMG 信号； 不论是仅 FMG 信号还是仅 EMG 信号，线性判别分析的表现都优于支持向 量机；结合 EMG-FMG 双模态时能获得更好的动态手势分类效果。

[1] 王美艳. 中国人口形势 、挑战与应对策略[J] . 国家安全研究, 2023(6): 102- 121 .
[2] KUBER P M, RASHEDI E.Alterations in Physical Demands During Virtual/AugmentedReality-Based Tasks:A Systematic Review[J] . Annals of Biomedical Engineering , 2023 , 51 (10) : 1910–1932.
[3] ADELSBERGER R S, CALATRONI A, SHAHNA S.A Novel Piezo-Based Technology forHaptic Feedback for XR[C]// 2023 IEEE Conference on Virtual Reality and 3D User InterfacesAbstracts and Workshops (VRW) . Shanghai, China, 2023: 1015-1016.
[4] GUARESE R, PRETTY E, ZAMBETTA F. XR towards tele-guidance: mixing realities inassistive technologies for blind and visually impaired people[C]// 2023 IEEE Conference onVirtual Reality and 3D User Interfaces Abstracts and Workshops (VRW) . Shanghai , China , 2023: 324-329.
[5] LIN P C, YANKSON B, CHAUHAN V, et al. Building a speech recognition system withprivacy identification information based on Google Voice for social robots[J] . Journal ofSupercomputing , 2022 , 78(12): 15060–15088 . DOI: 10. 1007/ s11227-022-04487-3.
[6] MULFARI D, CARNEVALE L, GALLETTA A, et al.Edge Computing Solutions SupportingVoice Recognition Services for Speakers with Dysarthria[C]// 2023 IEEE/ACM 23rdInternational Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW) . Bangalore, India, 2023: 231-236.
[7] ZHOU Y, LIU Y Y, WANG N, et al. Partial discharge ultrasonic signals pattern recognition intransformer using BSO-SVM based on microfiber coupler sensor[J]. Measurement, 2022, 201:111737. ISSN: 0263-2241.
[8] ZHANG Z. Partial Discharge Pattern Recognition Based on a Multifrequency F–P SensingArray, AOK Time–Frequency Representation, and Deep Learning[J]. IEEE Transactions onDielectrics and Electrical Insulation, 2022, 29(5): 1701-1710.
[9] CAI S, LU Z, CHEN B, et al. Dynamic Gesture Recognition of A-Mode Ultrasonic Basedon the DTW Algorithm[J]. IEEE Sensors Journal, 2022, 22(18): 17924-17931.
[10] OLSEN C D, HAMRICK W C, LEWIS S R, et al. Wrist EMG Improves GestureClassification for Stroke Patients[C]// 2023 International Conference on RehabilitationRobotics (ICORR) . Singapore , Singapore , 2023: 1- 6.
[11] BAI Y, LI X, ZHENG C, et al. Liquid Metal Flexible EMG Gel Electrodes for GestureRecognition[J]. Biosensors, 2023, 13(692).
[12] CHEN C, YU Y, SHENG X, et al. Real-Time Hand Gesture Recognition by Decoding MotorUnit Discharges Across Multiple Motor Tasks From Surface Electromyography[J]. IEEETransactions on Biomedical Engineering ,2023, 70(7): 2058-2068.
[13] REHMAN M U, SHAH B K, HAQ I U, et al. A Wearable Force Myography-Based Armbandfor Recognition of Upper Limb Gestures[J]. Sensors, 2023, 23(9357).
[14] REHMAN M U, SHAH K, HAQ I U, et al. A Force Myography based HMI for Classificationof Upper Extremity Gestures[C]// 2022 2nd International Conference on Artificial Intelligence(ICAI). Islamabad, Pakistan , 2022: 100-104.
[15] ZOU P. Wearable Iontronic FMG for Classification of Muscular Locomotion[J]. IEEE Journalof Biomedical and Health Informatics, 2022, 26(7): 2854-2863.
[16] REINSCHMIDT E, VOGT C, MAGNO M. Realtime Hand-Gesture Recognition Based onNovel Charge Variation Sensor and IMU[C]// 2022 IEEE Sensors. Dallas, TX, USA, 2022: 1-4.
[17] ZHANG D.Fine-Grained and Real-Time Gesture Recognition by Using IMU Sensors[J]. IEEETransactions on Mobile Computing, 202, 22(4): 2177-2189.
[18] WANG C.Intake Gesture Detection With IMU Sensor in Free-Living Environments: TheEffects of Measuring Two-Hand Intake and Down-Sampling[C]// 2023 IEEE 19thInternational Conference on Body Sensor Networks (BSN). Boston, MA, USA, 2023: 1-4.
[19] LING Y,CHEN X,RUAN Y,et al. Comparative Study of Gesture Recognition Based onAccelerometer and Photoplethysmography Sensor for Gesture Interactions in WearableDevices[J]. IEEE Sensors Journal, 2021, 21(15): 17107- 17117.
[20] ZHAO T, LIU J, WANG Y, et al. Towards Low-Cost Sign Language Gesture RecognitionLeveraging Wearables[J]. IEEE Transactions on Mobile Computing, 2021, 20(4): 1685-701.
[21] LI D, KANG P, ZHU K,et al. Feasibility of Wearable PPG for Simultaneous Hand Gestureand Force Level Classification[J]. IEEE Sensors Journal, 2023, 23(6): 6008- 6017.
[22] QI J, MA L, CUI Z, et al. Computer vision-based hand gesture recognition for human-robotinteraction: a review[J]. Complex & Intelligent Systems, 2024, 10: 1581–1606.
[23] CÓRDOVA J C, FLORES C, ANDREU-PEREZ J. EMGTFNet: Fuzzy Vision Transformer toDecode Upperlimb sEMG Signals for Hand Gestures Recognition[C]//2023 IEEE InternationalConference on Fuzzy Systems (FUZZ), 2023: 1- 6.
[24] TAN C K, LIM K M, LEE C P, et al. SDViT: Stacking of Distilled Vision Transformers forHand Gesture Recognition[J]. Applied Sciences, 2023, 13:12204 .
[25] DACCOLTI D, CLEMENTE F, MANNINI A, et al. Online Classification of Transient EMGPatterns for the Control of the Wrist and Hand in a Transradial Prosthesis[J]. IEEE Roboticsand Automation Letters, 2023, 8(2): 1045- 1052.
[26] BARRY D T, LEONARD J A, GITTER A J, et al. Acoustic myography as a control signal foran externally powered prosthesis[J]. Arch Phys Med Rehabil , 1986 , 67: 267-269 .
[27] XIAO Y, LIU T, HAN Y, et al. Realtime Recognition of Dynamic Hand Gestures in PracticalApplications[J]. ACM Trans. Multimedia Comput. Commun. Appl, 2024, 20(2): Article 50.
[28] YANG X, CHEN K, WAN H, et al. An Approach to Dynamic Gesture Recognition Based onInstantaneous Posture[C]// 2021 IEEE 7th International Conference on Virtual Reality (ICVR). Foshan, China , 2021: 90-95.
[29] WANG G, QIU W, LIU Y, et al. Damage detection for structural health monitoring usingultra- sensitive flexible piezoelectret sensors[J]. Structural Health Monitoring , 2023, 22(4):2800-2812 .
[30] REHMAN M U, SHAH K, HAQ I U, et al. Assessment of Low-Density Force MyographyArmband for Classification of Upper Limb Gestures[J]. Sensors, 2023, 23(9): 2716.
[31] REHMAN M U, SHAH K, HAQ I U, et al. A Wearable Force Myography-Based Armband forRecognition of Upper Limb Gestures[J]. Sensors, 2023, 23(16): 9357.
[32] MOQADAM S B, ASHEGHABADI A S, XU J. A Novel Hybrid Approach to PatternRecognition of Finger Movements and Grasping Gestures in Upper Limb Amputees[J]. IEEESensors Journal, 2022, 22(3): 2591-2602.
[33] LI N, YANG D, JIANG L, et al. Combined use of FSR sensor array and SVM classifier forfinger motion recognition based on pressure distribution map[J]. J Bionic Eng, 2012, 9(1): 39–47.
[34] XIAO Z G, MENON C. Towards the development of a wearable feedback system formonitoring the activities of the upper-extremities[J]. Xiao Menon J NeuroEng Rehabil, 2014, 11: 13.
[35] XIAO Z G, ELNADY A M, MENON C. Control an exoskeleton for forearm rotation usingFMG[C]//Proceedings of the 5th IEEE RAS/EMBS international conference on biomedicalrobotics and biomechatronics. 2014: 591- 596.
[36] 孙翰轩, 陈盛华, 徐策, 等. 具有多孔双微结构层的柔性电容式压力传感器[J/OL]. 微纳电子技术:1-9.
[37] GODIYAL A K, VERMA H K, KHANNA N, et al. A Force Myography-Based System forGait Event Detection in Overground and Ramp Walking[J]. IEEE Transactions onInstrumentation and Measurement, 201 , 67(10): 2314-2323.
[38] TAN J W, YI H. Application of Forearm FMG signals in Closed Loop Modality- matchedSensory Feedback Stimulation[J]. J Bionic Eng, 2020, 17(3): 899- 908.
[39] HELLARA H. Classification of Dynamic Hand Gestures using Multi SensorsCombinations[C]// 2022 IEEE 9th International Conference on Computational Intelligence andVirtual Environments for Measurement Systems and Applications (CIVEMSA). Chemnitz, Germany, 2022: 1-5.
[40] SESSLER G M, HILLENBRAND J. Electromechanical response of cellular electret films[C]//10th International Symposium on Electrets (ISE 10). Proceedings (Cat. No. 99 CH36256). IEEE, 1999: 261-264.
[41] ERHARD D P, LOVERA D. Recent advances in the improvement of polymer electretfilms[C]// Complex macromolecular systems II. 2010: 155-207.
[42] SESSLER G M Electrets: recent developments[J]. Journal of Electrostatics, 2001, 51:137- 145.
[43] NEUGSCHWANDTNER G S, SCHWÖDIAUER R, BAUER-GOGONEA S, et al. Piezo-andpyroelectricity of a polymer-foam space-charge electret[J]. Journal of Applied Physics, 2001, 89(8): 4503-4511.
[44] RAHMATI A H, YANG S, BAUER S, et al. Nonlinear bending deformation of soft electretsand prospects for engineering flexoelectricity and transverse (d31)piezoelectricity[J]. SoftMatter, 2019, 15(1): 127-148.
[45] BAUER S. Piezo-, pyro-and ferroelectrets: soft transducer materials for electromechanicalenergy conversion[J]. IEEE Transactions on Dielectrics and Electrical Insulation, 2006, 13(5):953-962.
[46] TAJITSU Y. Piezoelectric properties of ferroelectret[J]. Ferroelectrics, 2011, 415(1): 57-66.
[47] LEKKALA J, PORAMO R, NYHOLM K, et al. EMF force sensor— a flexible and sensitiveelectret film for physiological applications[J]. Medical and Biological Engineering andComputing, 1996, 34(Suppl 1): 67- 68.
[48] 游 琼 , 张 晓 青 . 聚 丙 烯 压 电 驻 极 体 薄 膜 声 电 传 感 器 的 性 能 [J]. 压 电 与 声 光 , 2013 ,35(06):849- 852 .
[49] GERHARD-MULTHAUPT R. Less can be more. Holes in polymers lead to a new paradigmof piezoelectric materials for electret transducers[J]. IEEE Transactions on Dielectrics andElectrical Insulation, 2002, 9(5): 850-859.
[50] WEGENER M, BAUER S. Microstorms in cellular polymers: A route to soft piezoelectrictransducer materials with engineered macroscopic dipoles[J]. ChemPhysChem, 2005, 6(6):1014-1025.
[51] 张添乐, 黄曦, 郑凯等. 极化电压对聚丙烯压电驻极体膜压电性能的影响[J].物理学报,2014,63(15):389-395.
[52] CHU Y, ZHONG J, LIU H, et al. Human pulse diagnosis for medical assessments using awearable piezoelectret sensing system[J]. Advanced Functional Materials, 2018, 28(40):1803413.
[53] ZHANG L, CHEN Q, HUANG X, et al. Fiber-based electret nanogenerator withasemisupported structure for wearable electronics[J]. ACS Applied Materials & Interfaces, 2021, 13(39): 46840-46847.
[54] WANG S, YANG J, LU P, et al. Speech Communication System Based on PiezoelectricElectret Mechanical Antenna[J]. Applied Sciences, 2023, 13(4): 2332.
[55] WANG X, CHEN P, WU M, et al. A Dynamic Gesture Recognition Algorithm based onFeature Fusion from RGB-D Sensor[C]// 2022 IEEE International Conference onMechatronics and Automation (ICMA). IEEE, 2022: 1040-1045.
[56] CAI S, LU Z, CHEN B, et al. Dynamic gesture recognition of A-mode ultrasonic based on theDTW algorithm[J]. IEEE Sensors Journal, 2022 , 22(18): 17924-17931.
[57] ZHANG Y, WANG F. HandFormer: A Dynamic Hand Gesture Recognition Method Based onAttention Mechanism[J]. Applied Sciences, 2023, 13(7): 4558.
[58] KE A, HUANG J, CHEN L, et al. An ultra-sensitive modular hybrid EMG–FMG sensor withfloating electrodes[J]. Sensors, 2020, 20(17): 4775.
[59] LIU C, ATITALLAH B B, RAMALINGAME R, et al.A Hybrid Measurement System forHand Signs Recognition based on EMG-FMG Measurements[C]// 2022 IEEE 9th InternationalConference on Computational Intelligence and Virtual Environments for MeasurementSystems and Applications (CIVEMSA). IEEE, 2022: 1-6.
[60] JIANG S, GAO Q, LIU H, et al. A novel, co-located EMG-FMG-sensing wearable armbandfor hand gesture recognition[J]. Sensors and Actuators A: Physical, 2020, 301: 111738.
[61] FANG P, PENG Y, LIN W H, et al. Wrist pulse recording with a wearablepiezoelectret compound sensing system and its applications in health monitoring[J]. IEEESensors Journal, 2021, 21(18): 20921-20930.
[62] MO X, ZHOU H, LI W, et al. Piezoelectrets for wearable energy harvesters and sensors[J]. Nano Energy, 2019, 65: 104033.
[63] MADDAH H A. Polypropylene as a promising plastic: A review[J]. Am. J. Polym. Sci, 2016, 6(1) : 1-11.
[64] MO X, ZHOU H, LI W, et al. Piezoelectrets for wearable energy harvesters and sensors[J]. Nano Energy, 2019, 65: 104033.
[65] MA X, ZHANG X, FANG P. Flexible film-transducers based on polypropylene piezoelectrets:Fabrication, properties, and applications in wearable devices[J]. Sensors and Actuators A:Physical, 2017, 256: 35-42 .
[66] CHEN L, CAO J, LI G, et al. Property assessment and application exploration for layeredpolytetrafluoroethylene piezoelectrets[J]. IEEE Sensors Journal, 2019, 19(23): 11262-11271.
[67] FANG P, PENG Y, LIN W H, et al. Wrist pulse recording with a wearable piezoresistor- piezoelectret compound sensing system and its applications in health monitoring[J]. IEEESensors Journal, 2021, 21(18): 20921-20930.
[68] RAPIN M, BRAUN F, ADLER A, et al. Wearable sensors for frequency-multiplexed EIT andmultilead ECG data acquisition[J]. IEEE Transactions on Biomedical Engineering, 2018,66(3):810-820.
[69] BELYEA A, ENGLEHART K, SCHEME E. FMG versus EMG: A comparison of usability forreal-time pattern recognition based control[J]. IEEE Transactions on Biomedical Engineering, 2019, 66(11): 3098-3104.
[70] KE A, HUANG J, CHEN L, et al. An ultra-sensitive modular hybrid EMG– FMG sensor withfloating electrodes[J]. Sensors, 2020, 20(17): 4775.
[71] ZHANG D, XIONG A, ZHAO X, et al. PCA and LDA for EMG-based control of bionicmechanical hand[C]// 2012 IEEE international conference on information and automation. IEEE, 2012: 960-965.
[72] CHU J U, MOON I, LEE Y J, et al. A supervised feature-projection-based real-time EMGpattern recognition for multifunction myoelectric hand control[J]. IEEE/ASME Transactionson mechatronics, 2007, 12(3): 282-290.
[73] SUN Q, ZHANG X, LI H, et al. A fault-tolerant algorithm to enhance generalization of EMG- based pattern recognition for lower limb movement[C]// 2020 10th Institute of Electrical andElectronics Engineers International Conference on Cyber Technology in Automation, Control, and Intelligent Systems (CYBER). IEEE, 2020: 332-337.
[74] YU Y, CHEN X, CAO S, et al. Exploration of Chinese sign language recognition usingwearable sensors based on deep belief net[J]. IEEE journal of biomedical and healthinformatics, 2019, 24(5): 1310-1320.
[75] ZHANG H, XIAO Z, WANG J, et al. A novel IoT-perceptive human activity recognition(HAR) approach using multihead convolutional attention[J]. IEEE Internet of Things Journal, 2019, 7(2): 1072-1080.
[76] DU C, ZHANG L, SUN X, et al. Enhanced Multi-Channel Feature Synthesis for Hand GestureRecognition Based on CNN With a Channel and Spatial Attention Mechanism. IEEE Access, Vol. 8 (2020), 144610-144620[J]. 2020.