Walk detection and step counting on unconstrained smartphones

Smartphone pedometry offers the possibility of ubiquitous health monitoring, context awareness and indoor location tracking through Pedestrian Dead Reckoning (PDR) systems. However, there is currently no detailed understanding of how well pedometry works when applied to smartphones in typical, unconstrained use. This paper evaluates common walk detection (WD) and step counting (SC) algorithms applied to smartphone sensor data. Using a large dataset (27 people, 130 walks, 6 smartphone placements) optimal algorithm parameters are provided and applied to the data. The results favour the use of standard deviation thresholding (WD) and windowed peak detection (SC) with error rates of less than 3%. Of the six different placements, only the back trouser pocket is found to degrade the step counting performance significantly, resulting in undercounting for many algorithms.

[1]  Juha Röning,et al.  Recognizing Human Activities User-independently on Smartphones Based on Accelerometer Data , 2012, Int. J. Interact. Multim. Artif. Intell..

[2]  P. Barralon,et al.  Walk Detection With a Kinematic Sensor: Frequency and Wavelet Comparison , 2006, 2006 International Conference of the IEEE Engineering in Medicine and Biology Society.

[3]  M. Friedman.  A Comparison of Alternative Tests of Significance for the Problem of $m$ Rankings , 1940 .

[4]  Dong-Hwan Hwang,et al.  A Step, Stride and Heading Determination for the Pedestrian Navigation System , 2004 .

[5]  C. Torrence,et al.  A Practical Guide to Wavelet Analysis. , 1998 .

[6]  E. Ambikairajah,et al.  Time-Frequency Based Features for Classification of Walking Patterns , 2007, 2007 15th International Conference on Digital Signal Processing.

[7]  Stephane Beauregard,et al.  A Helmet-Mounted Pedestrian Dead Reckoning System , 2006 .

[8]  Jian-Jiun Ding,et al.  Real time accelerometer-based gait recognition using adaptive windowed wavelet transforms , 2012, 2012 IEEE Asia Pacific Conference on Circuits and Systems.

[9]  Gary M. Weiss,et al.  Activity recognition using cell phone accelerometers , 2011, SKDD.

[10]  Shanshan Chen,et al.  Unsupervised activity clustering to estimate energy expenditure with a single body sensor , 2013, 2013 IEEE International Conference on Body Sensor Networks.

[11]  Henk L. Muller,et al.  Personal position measurement using dead reckoning , 2003, Seventh IEEE International Symposium on Wearable Computers, 2003. Proceedings..

[12]  Thilo Pfau,et al.  A hidden Markov model-based stride segmentation technique applied to equine inertial sensor trunk movement data. , 2008, Journal of biomechanics.

[13]  Gaetano Borriello,et al.  Validated caloric expenditure estimation using a single body-worn sensor , 2009, UbiComp.

[14]  Alistair Shilton,et al.  Automatic detection of different walking conditions using inertial sensor data , 2012, The 2012 International Joint Conference on Neural Networks (IJCNN).

[15]  Reinhold Haux,et al.  A performance comparison of accelerometry-based step detection algorithms on a large, non-laboratory sample of healthy and mobility-impaired persons , 2008, 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[16]  Janez Demsar,et al.  Statistical Comparisons of Classifiers over Multiple Data Sets , 2006, J. Mach. Learn. Res..

[17]  Tatsuo Nakajima,et al.  Feature Selection and Activity Recognition from Wearable Sensors , 2006, UCS.

[18]  Liu Ming,et al.  Identification of Individual Walking Patterns Using Gait Acceleration , 2007, 2007 1st International Conference on Bioinformatics and Biomedical Engineering.

[19]  M. N. Nyan,et al.  Classification of gait patterns in the time-frequency domain. , 2006, Journal of biomechanics.

[20]  Mikko Lindholm,et al.  Identifying people from gait pattern with accelerometers , 2005, SPIE Defense + Commercial Sensing.

[21]  S. P. Lloyd,et al.  Least squares quantization in PCM , 1982, IEEE Trans. Inf. Theory.

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

[23]  Michael L. Littman,et al.  Activity Recognition from Accelerometer Data , 2005, AAAI.

[24]  Sung-Bae Cho,et al.  Online Gesture Recognition for User Interface on Accelerometer Built-in Mobile Phones , 2010, ICONIP.

[25]  Bernt Schiele,et al.  Analyzing features for activity recognition , 2005, sOc-EUSAI '05.

[26]  Christopher M. Bishop,et al.  Pattern Recognition and Machine Learning (Information Science and Statistics) , 2006 .

[27]  Kenji Mase,et al.  Activity and Location Recognition Using Wearable Sensors , 2002, IEEE Pervasive Comput..

[28]  Andreas Krause,et al.  Unsupervised, dynamic identification of physiological and activity context in wearable computing , 2003, Seventh IEEE International Symposium on Wearable Computers, 2003. Proceedings..

[29]  Varghese S. Jacob,et al.  A study of the classification capabilities of neural networks using unsupervised learning: A comparison withK-means clustering , 1994 .

[30]  Gary M. Weiss,et al.  Cell phone-based biometric identification , 2010, 2010 Fourth IEEE International Conference on Biometrics: Theory, Applications and Systems (BTAS).

[31]  S. Mallat.  A wavelet tour of signal processing , 1998 .

[32]  Diogo R. Ferreira,et al.  Preprocessing techniques for context recognition from accelerometer data , 2010, Personal and Ubiquitous Computing.

[33]  Hylton B Menz,et al.  Accelerometry: a technique for quantifying movement patterns during walking. , 2008, Gait & posture.

[34]  Tapio Seppänen,et al.  Recognizing human motion with multiple acceleration sensors , 2001, 2001 IEEE International Conference on Systems, Man and Cybernetics. e-Systems and e-Man for Cybernetics in Cyberspace (Cat.No.01CH37236).

[35]  Angelo M. Sabatini,et al.  Accelerometry-Based Classification of Human Activities Using Markov Modeling , 2011, Comput. Intell. Neurosci..

[36]  Christoph Busch,et al.  Classification of Acceleration Data for Biometric Gait Recognition on Mobile Devices , 2011, BIOSIG.

[37]  Andrew J. Viterbi,et al.  Error bounds for convolutional codes and an asymptotically optimum decoding algorithm , 1967, IEEE Trans. Inf. Theory.

[38]  Kenichi Yamazaki,et al.  Gait analyzer based on a cell phone with a single three-axis accelerometer , 2006, Mobile HCI.

[39]  Anshul Rai,et al.  Zee: zero-effort crowdsourcing for indoor localization , 2012, Mobicom '12.

[40]  Einar Snekkenes,et al.  Towards understanding the uniqueness of gait biometric , 2008, 2008 8th IEEE International Conference on Automatic Face & Gesture Recognition.

[41]  Kôiti Hasida,et al.  Rotation invariant feature extraction from 3-D acceleration signals , 2011, 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[42]  R. Iman,et al.  Approximations of the critical region of the fbietkan statistic , 1980 .

[43]  Angelo M. Sabatini,et al.  A hidden Markov model-based technique for gait segmentation using a foot-mounted gyroscope , 2011, 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[44]  Joseph B. Kruskall,et al.  The Symmetric Time-Warping Problem : From Continuous to Discrete , 1983 .

[45]  Zhenyu He,et al.  Weightlessness feature — a novel feature for single tri-axial accelerometer based activity recognition , 2008, 2008 19th International Conference on Pattern Recognition.

[46]  Yasushi Makihara,et al.  Phase Registration of a Single Quasi-Periodic Signal Using Self Dynamic Time Warping , 2010, ACCV.

[47]  Kenneth Meijer,et al.  Activity identification using body-mounted sensors—a review of classification techniques , 2009, Physiological measurement.

[48]  Robert Harle,et al.  Pedestrian localisation for indoor environments , 2008, UbiComp.

[49]  Christoph Busch,et al.  Benchmarking the performance of SVMs and HMMs for accelerometer-based biometric gait recognition , 2011, 2011 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT).

[50]  Jelena Kovacevic,et al.  Wavelets and Subband Coding , 2013, Prentice Hall Signal Processing Series.

[51]  Nasser M. Nasrabadi,et al.  Pattern Recognition and Machine Learning , 2006, Technometrics.

[52]  Steffen Leonhardt,et al.  Automatic Step Detection in the Accelerometer Signal , 2007, BSN.

[53]  Waltenegus Dargie,et al.  Analysis of Time and Frequency Domain Features of Accelerometer Measurements , 2009, 2009 Proceedings of 18th International Conference on Computer Communications and Networks.

[54]  E. Ambikairajah,et al.  An Adapted Gaussian Mixture Model Approach to Accelerometry-Based Movement Classification Using Time-Domain Features , 2006, 2006 International Conference of the IEEE Engineering in Medicine and Biology Society.

[55]  Neal Lesh,et al.  Indoor navigation using a diverse set of cheap, wearable sensors , 1999, Digest of Papers. Third International Symposium on Wearable Computers.

[56]  Anshul Kumar,et al.  Strap-down Pedestrian Dead-Reckoning system , 2011, 2011 International Conference on Indoor Positioning and Indoor Navigation.

[57]  Robert Harle,et al.  A Survey of Indoor Inertial Positioning Systems for Pedestrians , 2013, IEEE Communications Surveys & Tutorials.

[58]  Feng Zhao,et al.  A reliable and accurate indoor localization method using phone inertial sensors , 2012, UbiComp.

[59]  Ling Bao,et al.  Activity Recognition from User-Annotated Acceleration Data , 2004, Pervasive.