Classification of Five Ambulatory Activities Regarding Stair and Incline Walking Using Smart Shoes

In this paper, we propose a novel method to classify five ambulatory activities, i.e., level ground, incline descent, incline ascent, stair descent, and stair ascent walking using smart shoes which contain eight plantar-pressure sensors on each shoe. Pressure data are collected using an insole-based monitoring system regarding the walking activities conducted by participants at their self-imposed “normal” speed. We present three new features based on an analysis of step patterns to characterize the ambulatory activities and utilize a k-nearest neighbor algorithm to classify the activities from the created features. In experimental results, we obtain walking activity-recognition error rates of 2.16% at the sixth walking step. Furthermore, a proposed method outperforms two reference methods in terms of F1-score and overall accuracy rate.

[1]  Wenyao Xu,et al.  Smart Insole: A Wearable Sensor Device for Unobtrusive Gait Monitoring in Daily Life , 2016, IEEE Transactions on Industrial Informatics.

[2]  Paola Catalfamo,et al.  Gait Event Detection on Level Ground and Incline Walking Using a Rate Gyroscope , 2010, Sensors.

[3]  Jeen-Shing Wang,et al.  Walking Pattern Classification and Walking Distance Estimation Algorithms Using Gait Phase Information , 2012, IEEE Transactions on Biomedical Engineering.

[4]  Edward Sazonov,et al.  Development of SmartStep: An insole-based physical activity monitor , 2013, 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[5]  Kim L. Bennell,et al.  Insole effects on impact loading during walking , 2011, Ergonomics.

[6]  Gu-Min Jeong,et al.  Classification of Three Types of Walking Activities Regarding Stairs Using Plantar Pressure Sensors , 2017, IEEE Sensors Journal.

[7]  Gu-Min Jeong,et al.  Stride Counting in Human Walking and Walking Distance Estimation Using Insole Sensors , 2016, Sensors.

[8]  Chih-Jen Lin,et al.  A comparison of methods for multiclass support vector machines , 2002, IEEE Trans. Neural Networks.

[9]  Paul J. M. Havinga,et al.  Activity Recognition Using Inertial Sensing for Healthcare, Wellbeing and Sports Applications: A Survey , 2010, ARCS Workshops.

[10]  Yasushi Makihara,et al.  Similar gait action recognition using an inertial sensor , 2015, Pattern Recognit..

[11]  Nigel H. Lovell,et al.  Implementation of a real-time human movement classifier using a triaxial accelerometer for ambulatory monitoring , 2006, IEEE Transactions on Information Technology in Biomedicine.

[12]  Stefan Poslad,et al.  Improved Use of Foot Force Sensors and Mobile Phone GPS for Mobility Activity Recognition , 2014, IEEE Sensors Journal.

[13]  P. Veltink,et al.  Ambulatory measurement of ground reaction forces , 2005 .

[14]  Sarah Jane Delany.  k-Nearest Neighbour Classifiers , 2007 .

[15]  J. Ross Quinlan,et al.  Induction of Decision Trees , 1986, Machine Learning.

[16]  Nigel H. Lovell,et al.  Can Triaxial Accelerometry Accurately Recognize Inclined Walking Terrains? , 2010, IEEE Transactions on Biomedical Engineering.

[17]  Matjaz B. Juric,et al.  Inertial Sensor-Based Gait Recognition: A Review , 2015, Sensors.

[18]  Peng Wang,et al.  Machine learning in bioinformatics: A brief survey and recommendations for practitioners , 2006, Comput. Biol. Medicine.

[19]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[20]  Yangsheng Xu,et al.  Gait pattern classification with integrated shoes , 2009, 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[21]  Oishee Mazumder,et al.  Ambulatory activity classification with dendogram-based support vector machine: Application in lower-limb active exoskeleton. , 2016, Gait & posture.

[22]  Björn Eskofier,et al.  Classification of surfaces and inclinations during outdoor running using shoe-mounted inertial sensors , 2012, Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012).

[23]  Edward Sazonov,et al.  Automatic Recognition of Activities of Daily Living Utilizing Insole-Based and Wrist-Worn Wearable Sensors , 2018, IEEE Journal of Biomedical and Health Informatics.

[24]  George C. Runger,et al.  Bias of Importance Measures for Multi-valued Attributes and Solutions , 2011, ICANN.

[25]  G. Cawley,et al.  Efficient approximate leave-one-out cross-validation for kernel logistic regression , 2008, Machine Learning.