一种基于决策树的行人运动模态识别方法A Methodology of Pedestrian Motion Mode Recognition Based on Decision Tree
计洁;黄劲松;
摘要(Abstract):
微机电系统(micro-electro-mechanical system,MEMS)技术的快速发展及其在移动智能终端中的日益普及,促进了人们对移动智能终端高精度定位算法的研究,尤其是利用智能手机进行无缝定位。然而,目前多数行人导航系统都以单一姿态为前提进行定位推算,并未考虑行人在使用手机过程中的多模态问题。因此,利用手机内置MEMS传感器的原始数据,基于决策树模型对行人运动模态进行了识别,判别模态包括绝对静止、通话步行、摆臂步行以及裤兜步行4种。首先,采用巴特沃斯低通滤波器对原始数据进行了数据预处理,并计算其模值;其次,提取特征值,包括信号能量、方差、均值、最大值和最小值,结合k折交叉验证(kfold cross-validation)法,采用决策树进行了模态识别,并进行验证。实验结果表明,本文模型的整体分类准确率可达97.23%,可为行人导航系统提供较为准确、可靠的模态信息以辅助定位。
关键词(KeyWords): 模态识别;MEMS传感器;k折交叉验证;决策树
基金项目(Foundation): 国家重点研发计划(2016YFB0501803)
作者(Authors): 计洁;黄劲松;
DOI: 10.14188/j.2095-6045.2020143
参考文献(References):
- [1]张鹏,赵齐乐,李由,等.基于PDR、WiFi指纹识别、磁场匹配组合的室内行人导航定位[J].测绘地理信息,2016,41(3):29-32
- [2] Chen Y,Shen C. Performance Analysis of Smartphone-Sensor Behavior for Human Activity Recognition[J]. IEEE Access,2017,5:3 095-3 110
- [3] Li Y,Xue D,Forrister E,et al. Human Activity Classification Based on Dynamic Time Warping of an On-Body Creeping Wave Signal[J]. IEEE Transactions on Antennas&Propagation,2016,64(11):4901-4 905
- [4] Uddin M T,Billah M M,Hossain M F. Random forests Based Recognition of Human Activities and Postural Transitions on Smartphone[C]. International Conference on Informatics. IEEE,Mataram,Indonesia,2016
- [5] Cheng W C,Jhan D M. Triaxial Accelerometer-Based Fall Detection Method Using a Self-Constructing Cascade-AdaBoost-SVM Classifier[J]. IEEE Journal of Biomedical and Health Informatics, 2013, 17(2):411-419
- [6] Goumas S K,Zervakis M E,Stavrakakis G S. Classification of Washing Machines Vibration Signals Using Discrete Wavelet Analysis for Feature Extraction[J].IEEE Transactions on Instrumentation&Measurement,2002,51(3):497-508
- [7] Susi M,Borio D,Lachapelle G. Accelerometer Signal Features and Classification Algorithms for Positioning Applications[C]. International Technical Meeting of the Institute of Navigation,Portland,USA,2011
- [8]刘宇,余跃,路永乐,等.基于多传感器时域特征的实时人体行为识别[J].中国惯性技术学报,2017,25(4):455-460
- [9] Susi M,Renaudin V,Lachapelle G. Motion Mode Recognition and Step Detection Algorithms for Mobile Phone Users[J]. Sensors(Basel,Switzerland),2013,13(2):1 539-1 562
- [10]吴佳莲,罗啊玲,丁鼎昊,等.基于视频的行人检测跟踪与统计算法研究[J].测绘地理信息,2019,44(3):88-93
- [11]顾芷宁,吉福龙,呙维,等.利用CNN实现室内人行/车行定位方法切换下的行为状态识别[J].测绘地理信息,2020,45(3):70-74
- [12] Mathie M J. Monitoring and Interpreting Human Movement Patterns Using a Triaxial Accelerometer[D].Sydney,Australia:University of New South Wales,2003
- [13] Olshen R,Breiman L,Friedman J, et al. Classification and Regression Trees[M]. New York:Chapmanand Hall,1984