[1]段鹏松,周志一,王超,等.WiNet:一种适用于无线感知场景的步态识别模型[J].西安交通大学学报,2020,54(07):187-195.[doi:10.7652/xjtuxb202007022]
 DUAN Pengsong,ZHOU Zhiyi,WANG Chao,et al.WiNet: a Gait Recognition Model Suitable for Wireless Sensing Scene[J].Journal of Xi'an Jiaotong University,2020,54(07):187-195.[doi:10.7652/xjtuxb202007022]
点击复制

WiNet:一种适用于无线感知场景的步态识别模型
分享到:

《西安交通大学学报》[ISSN:0253-987X/CN:61-1069/T]

卷:
54
期数:
2020年第07期
页码:
187-195
栏目:
出版日期:
2020-07-08

文章信息/Info

Title:
WiNet: a Gait Recognition Model Suitable for Wireless Sensing Scene
文章编号:
0253-987X(2020)07-0187-09
作者:
段鹏松1 周志一1 王超1 曹仰杰12 王恩东3
1.郑州大学软件学院, 450000, 郑州; 2.郑州大学汉威物联网研究院, 450000, 郑州; 3.浪潮集团, 250000, 济南
Author(s):
DUAN Pengsong1 ZHOU Zhiyi1 WANG Chao1 CAO Yangjie12 WANG Endong3
1. School of Software Engineering, Zhengzhou University, Zhengzhou 450000, China; 2. Hanwei Institute of Internet of Things, Zhengzhou University, Zhengzhou 450000, China; 3. Inspur Group, Jinan 250000, China
关键词:
步态识别 信道状态信息 频率能量图 卷积神经网络
Keywords:
gait recognition channel state information frequency energy map convolutional neural network
分类号:
TP391
DOI:
10.7652/xjtuxb202007022
文献标志码:
A
摘要:
针对现有基于Wi-Fi信号感知步态识别研究存在的特征获取不足、多人场景下单目标识别准确率低的问题,提出了一种基于频率能量图的步态识别模型WiNet。在对信道状态信息影响因子分析的基础上,选取其中的振幅数据作为步态识别的基础数据; 采用频率能量图对原始采集数据进行有效重构使其能够同时容纳步态行为对子载波内和子载波间扰动而产生的有效特征,步态特征的个体辨识度得到较大增强; 将频率能量图作为卷积神经网络模型的输入矩阵,经过多组卷积、正则和激活操作,再使用Softmax方法进行分类,得到步态行为对应的个体身份,实现了Wi-Fi环境下高准确率的多人场景单目标步态识别。与全连接神经网络、循环神经网络及基于卷积神经网络的步态识别模型进行了比较,结果表明:WiNet在40人场景实验中识别准确率达到98.5%,识别准确率得到明显提升; 在典型强/弱多径效应环境及5种人体状态的对比实验中,WiNet均能达到92%以上的识别准确率,具有良好的识别效果和鲁棒性。
Abstract:
Aiming at the problems of insufficient acquisition of features and low recognition accuracy of single target for multi-person scene in gait recognition from Wi-Fi signal perception, a new gait recognition model WiNet is proposed. Depending on the channel state information impact factor analysis, the amplitude data is chosen as the basic data for gait recognition, and a mechanism named frequency energy map is adopted to reconstruct the raw data effectively in WiNet. The advantage of WiNet lies in the capacity of extracting effective features generated by the gait behavior on the inter-subcarrier and intra-subcarrier signals at the same time, which greatly improves the individual recognition in gait recognition. The frequency energy map is used as the input matrix of the convolutional neural network model. After multiple groups convolution, regularization and activation operations, the softmax method is used in classification, and the individual identity corresponding to the gait behavior is obtained. The results show that, compared with other similar gait recognition models which are based on fully connected neural network, recurrent neural network and convolutional neural network, WiNet has a recognition accuracy of 98.5% for 40-person scene experiment, which is significantly improved. Besides, the recognition accuracy of the WiNet is still more than 92% in the contrast experiment of typical strong/weak multipath effect environments and five-person states, which shows that WiNet performs well and robustly.

参考文献/References:

[1] CONNOR P, ROSS A. Biometric recognition by gait: a survey of modalities and features [J]. Computer Vision and Image Understanding, 2018, 167: 1-27.
[2] 贲晛烨, 徐森, 王科俊. 行人步态的特征表达及识别综述 [J]. 模式识别与人工智能, 2012, 25(1): 71-81.
BEN Xianye, XU Sen, WANG Kejun. Review on pedestrian gait feature expression and recognition [J]. Pattern Recognition and Artificial Intelligence, 2012, 25(1): 71-81.
[3] WU Z F, HUANG Y Z, WANG L, et al. A comprehensive study on cross-view gait based human identification with deep CNNs [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(2): 209-226.
[4] 贲晛烨, 张鹏, 孙成立, 等. 均值张量成分分析及其在步态识别中的应用 [J]. 西安交通大学学报, 2015, 49(12): 40-46.
BEN Xianye, ZHANG Peng, SUN Chengli, et al. A mean tensor component analysis and its application in gait recognition [J]. Journal of Xi’an Jiaotong University, 2015, 49(12): 40-46.
[5] ZHANG Y T, PAN G, JIA K, et al. Accelerometer-based gait recognition by sparse representation of signature points with clusters [J]. IEEE Transactions on Cybernetics, 2015, 45(9): 1864-1875.
[6] NAIMI I A, WONG C B, MOORE P, et al. Multimodal approach for non-tagged indoor identification and tracking using smart floor and pyroelectric infrared sensors [J]. International Journal of Computational Science and Engineering, 2017, 14(1): 1-15
[7] Wi-Fi Alliance. Wi-Fi in 2019 [EB/OL].(2019-02-21)[2020-01-10]. https: ∥www.wi-fi.org/news-events/newsroom/wi-fi-in-2019.
[8] WANG J, ZHAO Y N, FAN X X, et al. Device-free identification using intrinsic CSI features [J]. IEEE Transactions on Vehicular Technology, 2018, 67(9): 8571-8581.
[9] ABDELNASSER H, HARRAS K, YOUSSEF M. A ubiquitous WiFi-based fine-grained gesture recognition system [J]. IEEE Transactions on Mobile Computing, 2019, 18(11): 2474-2487.
[10] SHI C, LIU J, LIU H B, et al. Smart user authentication through actuation of daily activities leveraging WiFi-enabled IoT [C]∥Proceedings of the 18th ACM International Symposium on Mobile Ad Hoc Networking and Computing(MobiHoc). New York, USA: ACM, 2017: 1-10.
[11] WANG W, LIU A X, SHAHZAD M. Gait recognition using WiFi signals [C]∥Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing. New York, USA: ACM, 2016: 363-373.
[12] ZENG Y Z, PATHAK P H, MOHAPATRA P. WiWho: WiFi-based person identification in smart spaces [C]∥Proceedings of the 15th International Conference on Information Processing in Sensor Networks. Piscataway, NJ, USA: IEEE, 2016: 7460727
[13] ZHANG J, WEI B, HU W, et al. WiFi-ID: human identification using WiFi signal [C]∥Proceedings of the 12th Annual International Conference on Distributed Computing in Sensor Systems(DCOSS). Piscataway, NJ, USA: IEEE, 2016: 75-82.
[14] XIN T, GUO B, WANG Z, et al. FreeSense: indoor human identification with WiFi signals [C]∥Proceedings of the 2016 IEEE Global Communications Conference(GLOBECOM). Piscataway, NJ, USA: IEEE 2016: 7841847.
[15] 余星达, 陈文杰, 王鼎, 等. 非接触式身份识别的深度学习算法 [J]. 西安交通大学学报, 2019, 53(4): 122-127.
YU Xingda, CHEN Wenjie, WANG Ding, et al. A deep learning algorithm for contactless human identification [J]. Journal of Xi’an Jiaotong University, 2019, 53(4): 122-127.
[16] OHARA K, MAEKAWA T, MATSUSHITA Y. Detecting state changes of indoor everyday objects using Wi-Fi channel state information [J]. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 2017, 1(3): 88.
[17] POKKUNURU A, JAKKALA K, BHUYAN A, et al. Neuralwave: gait-based user identification through commodity WiFi and deep learning [C]∥Proceedings of the 44th Annual Conference of the IEEE Industrial Electronics Society(IECON). Piscataway, NJ, USA: IEEE, 2018: 758-765.
[18] WEI B, HU W, YANG M R, et al. Radio-based device-free activity recognition with radio frequency interference [C]∥Proceedings of the 14th International Conference on Information Processing in Sensor Networks. New York, USA: ACM, 2015: 154164.
[19] ALAM M R, BENNAMOUN M, TOGNERI R, et al. A joint deep Boltzmann machine(JDBM)model for person identification using mobile phone data [J]. IEEE Transactions on Multimedia, 2017, 19(2): 317-326.
[20] CHEN Z H, ZHANG L, JIANG C Y, et al. WiFi CSI based passive human activity recognition using attention based BLSTM [J]. IEEE Transactions on Mobile Computing, 2019, 18(11): 2714-2724.
[21] ALI K, LIU A X, WANG W, et al. Recognizing keystrokes using WiFi devices [J]. IEEE Journal on Selected Areas in Communications, 2017, 35(5): 1175-1190.
[22] LIU Z J, LIU X L, ZHANG J W, et al. Opportunities and challenges of wireless human sensing for the smart IoT world: a survey [J]. IEEE Network, 2019, 33(5): 104-110.
[23] 鲁勇, 吕绍和, 王晓东, 等. 基于WiFi信号的人体行为感知技术研究综述 [J]. 计算机学报, 2019, 42(2): 1-21.
LU Yong, Lü Shaohe, WANG Xiaodong, et al. A survey on WiFi based human behavior analysis technology [J]. Chinese Journal of Computers, 2019, 42(2): 1-21.
[24] MOU L C, GHAMISI P, ZHU X X. Deep recurrent neural networks for hyperspectral image classification [J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(7): 3639-3655.
[25] SHELHAMER E, LONG J, DARRELL T. Fully convolutional networks for semantic segmentation [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4): 640-651

相似文献/References:

[1]陈实,高有行.一种轮廓变化图像小波矩的步态识别[J].西安交通大学学报,2009,43(01):090.[doi:10.7652/xjtuxb200901020]
 Chen Shi,Gao Youxing.Gait Recognition with Wavelet Moments of Silhouette Change Images[J].Journal of Xi'an Jiaotong University,2009,43(07):090.[doi:10.7652/xjtuxb200901020]
[2]贲晛烨,张鹏,孙成立,等.均值张量成分分析及其在步态识别中的应用[J].西安交通大学学报,2015,49(12):040.[doi:10.7652/xjtuxb201512007]
 BEN Xianye,ZHANG Peng,SUN Chengli,et al.A Mean Tensor Component Analysis and Its Application in Gait Recognition[J].Journal of Xi'an Jiaotong University,2015,49(07):040.[doi:10.7652/xjtuxb201512007]

备注/Memo

备注/Memo:
收稿日期: 2020-02-01。作者简介: 段鹏松(1983—)男,在职博士生,讲师; 曹仰杰(通信作者),男,副教授,硕士生导师。基金项目: 国家自然科学基金资助项目(61972092)。
更新日期/Last Update: 2020-07-10