基于脚步诱发结构振动的人员特征身份识别研究

PDF(3485 KB)

振动与冲击 ›› 2022, Vol. 41 ›› Issue (23) : 249-256.

论文

侯兴民，李冉，张玉洁

作者信息 +

Personnel characteristics identification based on foot induced structural vibration

HOU Xingmin, LI Ran, ZHANG Yujie

Author information +

文章历史 +

摘要

身份识别是安防领域一项重要工作，目前生物特征识别方法主要利用静态生理特征，利用脚步振动信号进行身份识别研究相对较少，本文提出利用行走过程中脚步诱发结构振动信号的差异性来识别人员。基于能量阈值法检测脚步事件与非脚步事件，对不同测试人员单一脚步事件在时域、频域方面共16个脚步特征参数进行了对比分析，研究发现可以将不同特征组合下参数差异性作为身份识别的依据。为了验证方法的有效性，采用支持向量机（SVM）作为分类工具，测试人数为10人数据样本500个情况下，选用16个脚步特征参数平均识别率为79.21%；采用皮尔逊相关系数法筛选出彼此不相关的10个脚步特征参数平均识别率为91%，相比于采用16个脚步特征参数平均识别率提高了11.79%；对比了在不同SVM核函数下分类工具对选取的10个脚步特征参数平均识别率的影响，结果采用线性核函数下平均识别率最高达到96%。结果表明，有效的脚步特征参数组合适用于小样本下的身份识别。
关键词：身份识别；脚步振动信号；事件检测；特征提取；支持向量机(SVM)

Abstract

Identity recognition is an important work in the field of security. At present, the methods based on biometrics are mainly static physiological characteristics. However, there are relatively few researches on the use of foot vibration signal for identity recognition. In this paper, we use the difference of vibration signals caused by different people's foot induced structure to identify individuals. Firstly, footstep events and non-footstep events are detected based on energy threshold method, and then 16 footstep characteristic parameters of single footstep events of different testers are compared and analyzed in time domain and frequency domain. it is found that the parameter differences under different feature combinations can be used as the basis for identity identification. In order to verify the effectiveness of the method, support vector machine (SVM) is used as a classification tool. In a quiet test environment, without considering the influence of testers' shoes, the average recognition rates of the three experimental cases are 79.21%, 91% and 96%, respectively. The results show that the effective combination of footstep feature parameters is suitable for identity recognition under small samples.
Key words: Identification; Foot vibration signal; Event detection; Feature extraction; Support vector machines

导出引用

侯兴民，李冉，张玉洁. 基于脚步诱发结构振动的人员特征身份识别研究[J]. 振动与冲击, 2022, 41(23): 249-256

HOU Xingmin, LI Ran, ZHANG Yujie. Personnel characteristics identification based on foot induced structural vibration[J]. Journal of Vibration and Shock, 2022, 41(23): 249-256

参考文献

[1] 余瑶，郭建敏，王晅 . 基于局部频谱特征与贝叶斯决策的脚步声识别［J］. 计算机应用与软件，2015, 32(12): 136-139.
YU Yao, GUO Jianmin, WANG Xuan. Footstep recognition based on local spectrogram characteristics and Bayesian decision [J]. Computer Applications and Software, 2015, 32(12): 136-139.
[2] Nikolaos V Boulgouris, Dimitrios Hatzinakos, and Konstantinos N Plataniotis. 2005. Gait recognition: a challenging signal processing technology for biometric identification. signal processing magazine, IEEE 22, 6 (2005), 78–90.
[3] I. Bisio, A. Delfino, F. Lavagetto, and A. Sciarrone, “Enabling IoT for in-home rehabilitation: Accelerometer signals classification methods for activity and movement recognition,” IEEE Internet Things J., vol. 4,no. 1, pp. 135–146, Feb 2017.
[4] S. Pan, N. Wang, Y . Qian, I. V elibeyoglu, H. Y . Noh, and P . Zhang,“Indoor person identification through footstep induced structural vibration,” in Proc. of the 16th Int. Workshop on Mobile Computing Systems and Applications. New Y ork, NY , USA: ACM, 2015, pp. 81–86.
[5] X. Jin, S. Sarkar, A. Ray, S. Gupta, and T. Damarla, “Target detection and classification using seismic and pir sensors,” IEEE Sensors J.vol. 12, no. 6, pp. 1709–1718, June 2012.
[6] S. Anchal, B. Mukhopadhyay, and S. Kar, “Predicting gender from footfalls using a seismic sensor,” in Int. Conf. Commun. Syst. and Netw.(COMSNETS), Jan 2017, pp. 47–54.
[7] Y .Hori, T. Ando, and A. Fukuda, “Initial evaluation of personal
identification system using walking footstep sound of one step,” in
The 17th Int’l Conf on Software Engineering Research and Practice,
2019, pp. 70–76.
[8] S. Yasuhiro, T. Takasuka, and H. Yasukawa, “Personal identification using footstep detection,” pp. 43–47, Nov 2004.
[9] 张瑞兴，王晅，余瑶 . 基于脚步声的身份识别［J］. 计算机应用与软件，2014, 31(1): 162-164.
ZHANG Ruixing, WANG Xuan, YU Yao. Personal identiﬁ-cation based on footstep [J]. Computer Applications and Soft-ware, 2014, 31(1): 162-164.
[10] S. Mouring, “Dynamic response of floor systems to building occupant activities,” Ph.D. dissertation, The Johns Hopkins University, MD, 1992.
[11] 余腾,胡伍生,吴杰,李海锋,乔燕.基于小波阈值去噪与EMD分解方法提取润扬大桥振动信息[J].振动与冲击,2019,38(12):264-270.
YU Teng, HU Wusheng, WU Jie, et al. Extraction of Runyang bridge vibration information based on a fusionmethod of wavelet threshold denoising and EMD decomposition [J]. Journal of Vibration and Shock, 2019, 38(12): 264-270．
[12] T. Akazawa, “A technique for automatic detection of onset time of p-ands-phases in strong motion records,” in Proc. World Conf. on Earthquake Engineering, V ancouver, B.C., Canada, 2004.
[13] G. P . Succi, D. Clapp, R. Gampert, and G. Prado, “Footstep detection and tracking,” in Aerospace/Defense Sensing, Simulation, and Controls.Int. Society for Optics and Photonics, 2001, pp. 22–29.
[14] J. Akram and D. Eaton, “Adaptive microseismic event detection and automatic time picking,” in CSEG Annual Convention, vol. 15, 2012.
[15] Ekimov Alexander and Sabatier James M. Vibration and sound signatures of human footsteps in buildings.[J]. The Journal of the Acoustical Society of America, 2006, 120(2) : 762-8.
[16]徐庆阳,李爱群,张志强,丁幼亮.考虑人体舒适度的大跨度悬挂结构振动控制研究[J].振动与冲击,2008,(04):139-142+176.
XU Qingyang, LI Aiqun, ZHANG Zhiqiang, et al. Vibration control of a long-span suspended structure considering human comfort [J]. Journal of Vibration and Shock, 2008, 27(4): 139-142．
[17] 窦希杰,王世博,谢洋,等. 基于IMF能量矩和SVM的煤矸识别[J].振动与冲击,2020,39(24):39-45.
DOU Xijie, WANG Shibo, XIE Yang, el at. Coal and gangue identification based on IMF energy moment and SVM[J]. Journal of Vibrationand Shock, 2020, 39(24): 39-45.
[18] Clemente Jose et al. Smart Seismic Sensing for Indoor Fall Detection, Location, and Notification.[J]. IEEE journal of biomedical and health informatics, 2020, 24(2) : 524-532.
[19] 赵志宏,杨绍普,申永军.基于独立分量分析与相关系数的机械故障特征提取[J]. 振动与冲击,2013, 32(06): 67-72.
ZHAO Zhihong, YANG Shaopu, SHEN Yongjun. Machineryfault feature extraction based on independent componentanalysis and correlation coefficient [J]. Journal of Vibrationand Shock, 2013, 32(6): 67－72．