Optimization of Physical Activity Recognition for Real-Time Wearable Systems: Effect of Window Length, Sampling Frequency and Number of Features

The aim of this study was to develop an optimized physical activity classifier for real-time wearable systems with the focus on reducing the requirements on device power consumption and memory buffer. Classification parameters evaluated in this study were the sampling frequency of the acceleration signal, window length of the classification fragment, and the number of classification features, found with different feature selection methods. For parameter evaluation, a decision tree classifier was created based on the acceleration signals recorded during tests, where 25 healthy test subjects performed various physical activities. Overall average F1-score achieved in this study was about 0.90. Similar F1-scores were achieved with the evaluated window lengths of 5 s (0.92 ± 0.02) and 3 s (0.91 ± 0.02), while classification performance with 1 s were lower (0.87 ± 0.02). Tested sampling frequencies of 50 Hz, 25 Hz, and 13 Hz had similar results with most classified activity types, with an exception of outdoor cycling, where differences were significant. Using forward sequential feature selection enabled the decreasing of the number of features from initial 110 features to about 12 features without lowering the classification performance. The results of this study have been used for developing more efficient real-time physical activity classifiers.

[1]  Xiaohui Peng,et al.  Deep Learning for Sensor-based Activity Recognition: A Survey , 2017, Pattern Recognit. Lett..

[2]  Gert Jervan,et al.  Fall detection and activity recognition system for usage in smart work-wear , 2018, 2018 16th Biennial Baltic Electronics Conference (BEC).

[3]  Vinod Chandran,et al.  Physical Activity Recognition Using Posterior-Adapted Class-Based Fusion of Multiaccelerometer Data , 2017, IEEE Journal of Biomedical and Health Informatics.

[4]  Mei-Po Kwan,et al.  Physical activity classification in free-living conditions using smartphone accelerometer data and exploration of predicted results , 2018, Comput. Environ. Urban Syst..

[5]  Reza Malekian,et al.  Physical Activity Recognition From Smartphone Accelerometer Data for User Context Awareness Sensing , 2017, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[6]  S. Edney,et al.  Users’ experiences of wearable activity trackers: a cross-sectional study , 2017, BMC Public Health.

[7]  Mahbub Hassan,et al.  A Survey of Wearable Devices and Challenges , 2017, IEEE Communications Surveys & Tutorials.

[8]  Tao Dong,et al.  A Review of Wearable Technologies for Elderly Care that Can Accurately Track Indoor Position, Recognize Physical Activities and Monitor Vital Signs in Real Time , 2017, Sensors.

[9]  Ivo Fridolin,et al.  Activity classification for real-time wearable systems: Effect of window length, sampling frequency and number of features on classifier performance , 2016, 2016 IEEE EMBS Conference on Biomedical Engineering and Sciences (IECBES).

[10]  S. Coughlin,et al.  Use of Consumer Wearable Devices to Promote Physical Activity: A Review of Health Intervention Studies , 2016, Journal of environment and health sciences.

[11]  Jungsun Kim,et al.  Energy-Efficient Real-Time Human Activity Recognition on Smart Mobile Devices , 2016, Mob. Inf. Syst..

[12]  R. Furberg,et al.  Systematic review of the validity and reliability of consumer-wearable activity trackers , 2015, International Journal of Behavioral Nutrition and Physical Activity.

[13]  Muhammad Awais,et al.  Physical activity classification meets daily life: Review on existing methodologies and open challenges , 2015, 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[14]  Reynald Hoskinson,et al.  Fitness activity classification by using multiclass support vector machines on head-worn sensors , 2015, 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[15]  Meng Li,et al.  A Random Forest-based ensemble method for activity recognition , 2015, 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[16]  Roberto Sassi,et al.  Parametric estimation of sample entropy for physical activity recognition , 2015, 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[17]  J. Brug,et al.  Apps to promote physical activity among adults: a review and content analysis , 2014, International Journal of Behavioral Nutrition and Physical Activity.

[18]  A Moncada-Torres,et al.  Activity classification based on inertial and barometric pressure sensors at different anatomical locations , 2014, Physiological measurement.

[19]  Bernt Schiele,et al.  A tutorial on human activity recognition using body-worn inertial sensors , 2014, CSUR.

[20]  Accelerometer-based energy expenditure estimation methods and performance comparison , 2013, CSE 2013.

[21]  SHAOPENG LIU,et al.  Computational methods for estimating energy expenditure in human physical activities. , 2012, Medicine and science in sports and exercise.

[22]  Julien Penders,et al.  Energy expenditure estimation using wearable sensors: a new methodology for activity-specific models , 2012, Wireless Health.

[23]  David W. Dunstan,et al.  Too much sitting--a health hazard. , 2012, Diabetes research and clinical practice.

[24]  Vigneshwaran Subbaraju,et al.  Energy-Efficient Continuous Activity Recognition on Mobile Phones: An Activity-Adaptive Approach , 2012, 2012 16th International Symposium on Wearable Computers.

[25]  Billur Barshan,et al.  Comparative study on classifying human activities with miniature inertial and magnetic sensors , 2010, Pattern Recognit..

[26]  David Howard,et al.  A Comparison of Feature Extraction Methods for the Classification of Dynamic Activities From Accelerometer Data , 2009, IEEE Transactions on Biomedical Engineering.

[27]  Emmanuel,et al.  Using machine learning for real-time activity recognition and estimation of energy expenditure , 2008 .

[28]  D. Powers.  Evaluation: From Precision, Recall and F-Factor to ROC, Informedness, Markedness & Correlation , 2008 .

[29]  D. Warburton,et al.  Health benefits of physical activity: the evidence , 2006, Canadian Medical Association Journal.

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

[31]  A. Astrup,et al.  Obesity : Preventing and managing the global epidemic , 2000 .

[32]  K. Rajamani,et al.  An efficient algorithm for sample rate conversion from CD to DAT , 2000, IEEE Signal Processing Letters.

[33]  J. D. Janssen,et al.  A triaxial accelerometer and portable data processing unit for the assessment of daily physical activity , 1997, IEEE Transactions on Biomedical Engineering.