计算机科学 ›› 2021, Vol. 48 ›› Issue (8): 322-327.doi: 10.11896/jsjkx.200600122

• 人机交互 • 上一篇    下一篇

基于3D卷积神经网络的CSI跨场景手势识别方法

王炽, 常俊   

  1. 云南大学信息学院 昆明650500
  • 收稿日期:2020-06-21 修回日期:2020-10-11 发布日期:2021-08-10
  • 通讯作者: 常俊(changjun@ynu.edu.cn)
  • 基金资助:
    云南省省教育厅科研基金(2019J0007)

CSI Cross-domain Gesture Recognition Method Based on 3D Convolutional Neural Network

WANG Chi, CHANG Jun   

  1. College of Information,Yunnan University,Kunming 650500,China
  • Received:2020-06-21 Revised:2020-10-11 Published:2021-08-10
  • About author:WANG Chi,born in 1995,undergra-duate student.His main research in-terests include wireless perception and so on.(2110791732@qq.com)CHANG Jun,born in 1970,assistant professor,postgraduate's supervisor,is a member of China Computer Federation.His main research interests include intelligent wireless perception and wireless communication.
  • Supported by:
    Research Fund of Yunnan Province Department of Education(2019J0007).

摘要: 手势识别在人机交互中有着广泛的应用前景,近年来随着无线通信与物联网的飞速发展,几乎任何地方都部署了WiFi设备,并涌现了大批关于WiFi信道状态信息(Channel State Information,CSI)的手势识别方法,目前大多数基于CSI手势识别的研究仅针对了已知场景下的手势识别研究,对于未知场景,需要增加未知场景中的新数据进行额外的学习训练,否则识别精度将会大幅下降,限制了其实用性。针对这一问题,提出了一种基于3D卷积神经网络的CSI跨场景手势识别方法,该系统通过提取与场景无关的特征,并结合3D卷积神经网络学习模型来实现跨场景手势识别,在实验中使用网络公开数据集来验证该方法,结果显示该方法对于6个不同动作手势,在已知场景中的平均识别准确率达到了86.50%,在未知场景中的平均识别准确率达到了84.67%,能够实现跨场景的手势识别。

关键词: 3D卷积神经网络, WiFi, 跨场景, 手势识别, 信道状态信息

Abstract: Gesture recognition has important application prospects in human-computer interaction.In recent years,with the rapid development of wireless communication and the Internet of Things,WiFi devices have been deployed almost anywhere,and a large number of gesture recognition methods have appeared on WiFi channel status information.At present,most researches based on CSI gesture recognition only focus on the research of gesture recognition in known domain.For unknown domain,new data in unknown scenes need to be added for additional learning and training,otherwise the recognition accuracy will be greatly reduced,limiting practicality.To address this problem,a CSI cross-domain gesture recognition method based on 3D convolutional neural network is proposed.The method realizes cross-scene gesture recognition by extracting domain-independent features,and combining with the 3D convolutional neural network learning model.In order to verify the method,experiment uses the public dataest.For 6 different gestures,the results show that the method achieves 86.50% recognition accuracy in known domain,and achieves 84.67% recognition accuracy in unknown scenes,it can achieve cross-scene gesture recognition.

Key words: 3D convolutional neural network, Channel state information, Cross-domain, Gesture recognition, WiFi

中图分类号: 

  • TN99
[1]GKIOXARI G,GIRSHICK R,DOLLÁR P,et al.Detecting and Recognizing Human-Object Interactions[C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition.2018:8359-8367.
[2]XU C,PATHAK P H,MOHAPATRA P.Finger-writing with smartwatch:A case for finger and hand gesture recognition using smartwatch[C]//The 16th International Workshop on Mobile Computing Systems and Applications.2015:9-14.
[3]SHEN S,WANG H,CHOUDHURY R R.I am a Smartwatch and I can Track my User's Arm[C]//Proceedings of the 14th Annual International Conference on Mobile Systems,Applications,and Services.2016:85-96.
[4]NANDAKUMAR R,TAKAKUWA A,KOHNO T,et al.Covertband:Activity Information Leakage Using Music[J].ACM Journals,2017,1(3):1-24.
[5]FILHO J D P R,SILVA F J S,COUTINHO L R,et al.MHARS:A mobile system for human activity recognition and inference of health situations in ambient assisted living[J].Journal of Applied Computing Research,2016,5(1):44-58.
[6]WANG Y,LIU J,CHEN Y Y,et al.E-eyes:Device-free Location-oriented Activity Identification Using Fine-grained WiFi Signatures[C]//Proceedings of the 20th Annual International Conference on Mobile Computing and Networking.2014:617-628.
[7]WANG W,LIU A X,SHAHZAD M,et al.Device-Free Human Activity Recognition Using Commercial WiFi Devices [J].IEEE Journal on Selected Areas in Communications,2017,35(5):1118-1131.
[8]TIAN Z,WANG J,YANG X,et al.WiCatch:A WiFi BasedHand Gesture Recognition System [J].IEEE Access.,2018,6:16911-16923.
[9]VIRMANI A,SHAHZAD M.Position and Orientation Agnostic Gesture Recognition Using WiFi[C]//Proceedings of the 15th Annual International Conference on Mobile Systems,Applications,and Services.2017:252-264.
[10]ZHANG J,TANG Z Y,LI M,et al.CrossSense:Towards Cross-Site and Large-Scale WiFi Sensing[C]//Proceedings of the 24th Annual International Conference on Mobile Computing and Networking.2018:305-320.
[11]JIANG W J,MIAO C L,MA F L,et al.Towards Environment Independent Device Free Human Activity Recognition[C]//Proceedings of the 24th Annual International Conference on Mobile Computing and Networking.2018:289-304.
[12]ZHENG Y,ZHANG Y,QIAN K,et al.Zero-Effort Cross-Domain Gesture Recognition with WiFi[C]//Proceedings of the 17th Annual International Conference on Mobile Systems,Applications,and Services.2019:313-325.
[13]WANG W,LIU A X,SHAHZAD M,et al.Understanding and Modeling of WiFi Signal Based Human Activity Recognition[C]//Proceedings of the 21st Annual International Conference on Mobile Computing and Networking.2015:65-76.
[14]QIAN K,WU C S,YANG Z,et al.Widar:Decimeter-Level Passive Tracking via Velocity Monitoring with Commodity WiFi[C]//Proceedings of the 18th ACM International Symposium on Mobile Ad Hoc Networking and Computing.2017:1-10.
[1] 周楚霖, 陈敬东, 黄凡.
基于无迹粒子滤波的WiFi-PDR融合室内定位技术
WiFi-PDR Fusion Indoor Positioning Technology Based on Unscented Particle Filter
计算机科学, 2022, 49(6A): 606-611. https://doi.org/10.11896/jsjkx.210700108
[2] 卿朝进, 杜艳红, 叶青, 杨娜, 张岷涛.
存在CSI估计错误的增强型ELM叠加CSI反馈方法
Enhanced ELM-based Superimposed CSI Feedback Method with CSI Estimation Errors
计算机科学, 2022, 49(6A): 632-638. https://doi.org/10.11896/jsjkx.210800036
[3] 刘亮, 蒲浩洋.
基于LSTM的多维度特征手势实时识别
Real-time LSTM-based Multi-dimensional Features Gesture Recognition
计算机科学, 2021, 48(8): 328-333. https://doi.org/10.11896/jsjkx.210300079
[4] 尤凌, 管张均.
一种低复杂度的水声OFDM通信系统子载波分配算法
Low-complexity Subcarrier Allocation Algorithm for Underwater OFDM Acoustic CommunicationSystems
计算机科学, 2021, 48(6A): 387-391. https://doi.org/10.11896/jsjkx.201100064
[5] 王颖颖, 常俊, 武浩, 周详, 彭予.
基于WiFi-CSI的入侵检测方法
Intrusion Detection Method Based on WiFi-CSI
计算机科学, 2021, 48(6): 343-348. https://doi.org/10.11896/jsjkx.200700006
[6] 刘肖, 袁冠, 张艳梅, 闫秋艳, 王志晓.
基于自适应多分类器融合的手势识别
Hand Gesture Recognition Based on Self-adaptive Multi-classifiers Fusion
计算机科学, 2020, 47(7): 103-110. https://doi.org/10.11896/jsjkx.200100073
[7] 景雨, 祁瑞华, 刘建鑫, 刘朝霞.
基于改进多尺度深度卷积网络的手势识别算法
Gesture Recognition Algorithm Based on Improved Multiscale Deep Convolutional Neural Network
计算机科学, 2020, 47(6): 180-183. https://doi.org/10.11896/jsjkx.200200030
[8] 田春元, 余江, 常俊, 王彦舜.
NWI:基于CSI的非视距信号识别方法
NWI:CSI Based Non-line-of-sight Signal Recognition Method
计算机科学, 2020, 47(11): 327-332. https://doi.org/10.11896/jsjkx.190900019
[9] 贾玉福, 李明磊, 刘文平, 胡胜红, 蒋洪波.
一种基于WiFi相异度的群组感知分析方法
Group Perception Analysis Method Based on WiFi Dissimilarity
计算机科学, 2020, 47(10): 63-68. https://doi.org/10.11896/jsjkx.200600014
[10] 韩笑, 张晶, 李月龙.
基于手势几何分布特征的手势识别
Gesture Recognition Based on Hand Geometric Distribution Feature
计算机科学, 2019, 46(6A): 246-249.
[11] 李愚, 柴国钟, 卢纯福, 唐智川.
基于增量自适应学习的在线肌电手势识别
On-line sEMG Hand Gesture Recognition Based on Incremental Adaptive Learning
计算机科学, 2019, 46(4): 274-279. https://doi.org/10.11896/j.issn.1002-137X.2019.04.043
[12] 李晓薇, 余江, 常俊, 杨锦朋, 冉亚鑫.
一种基于CSI的非合作式人体行为识别方法
Non-cooperative Human Behavior Recognition Method Based on CSI
计算机科学, 2019, 46(12): 266-271. https://doi.org/10.11896/jsjkx.190200349
[13] 刘佳慧, 王昱洁, 雷艺.
基于LSTM的CSI手势识别方法
CSI Gesture Recognition Method Based on LSTM
计算机科学, 2019, 46(11A): 283-288.
[14] 宋一凡, 张鹏, 刘立波.
基于视觉手势识别的人机交互系统
Human-machine Interaction System with Vision-based Gesture Recognition
计算机科学, 2019, 46(11A): 570-574.
[15] 丁亚三, 郭斌, 辛通, 王沛, 王柱, 於志文.
WiCount:一种基于WiFi-CSI的人数识别方法
WiCount:A Crowd Counting Method Based on WiFi Channel State Information
计算机科学, 2019, 46(11): 297-303. https://doi.org/10.11896/jsjkx.191100506C
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!