Adaptive sliding window segmentation for physical activity recognition using a single tri-axial accelerometer

Abstract Previous studies on physical activity recognition have utilized various fixed window sizes for signal segmentation targeting specific activities. Naturally, an optimum window size varies depending on the characteristics of activity signals and fixed window size will not produce good segmentation for all activities. This paper presents a novel approach to activity signal segmentation for physical activity recognition. Central to the approach is that the window size is adaptively adjusted according to the probability of the signal belongs to a particular activity to achieve the most effective segmentation. In addition, an activity transition diagram for activity recognition is developed to validate the activity transition and improve recognition accuracy. The adaptive sliding window segmentation algorithm and the role of activity transition diagram are described in the context of physical activity recognition. The approach recognizes not only well defined static and dynamic activities, but also transitional activities. The presented approach has been implemented, evaluated and compared with an existing state-of-the-art approach by using internal and public datasets which contains activity signals of dynamic, static and transitional activities. Results have shown that the proposed adaptive sliding window segmentation achieves overall accuracy of 95.4% in all activities considered in the experiments compared to the existing approach which achieved an overall accuracy of 89.9%. The proposed approach achieved an overall accuracy of 96.5% compared to 91.9% overall accuracy with the existing approach when tested on the public dataset.

[1]  Tim Dallas,et al.  Feature Selection and Activity Recognition System Using a Single Triaxial Accelerometer , 2014, IEEE Transactions on Biomedical Engineering.

[2]  Cagatay Catal,et al.  On the use of ensemble of classifiers for accelerometer-based activity recognition , 2015, Appl. Soft Comput..

[3]  Arkady B. Zaslavsky,et al.  Context Aware Computing for The Internet of Things: A Survey , 2013, IEEE Communications Surveys & Tutorials.

[4]  K. I-K Wang,et al.  Dynamic sliding window method for physical activity recognition using a single tri-axial accelerometer , 2015, 2015 IEEE 10th Conference on Industrial Electronics and Applications (ICIEA).

[5]  Ifeyinwa E. Achumba,et al.  Sensor Data Acquisition and Processing Parameters for Human Activity Classification , 2014, Sensors.

[6]  H.J. Stam,et al.  Automated estimation of initial and terminal contact timing using accelerometers; development and validation in transtibial amputees and controls , 2005, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

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

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

[9]  Christopher J. James,et al.  Wrist-Worn Accelerometer to Detect Postural Transitions and Walking Patterns , 2014 .

[10]  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.

[11]  Wendong Xiao,et al.  Daily Human Physical Activity Recognition Based on Kernel Discriminant Analysis and Extreme Learning Machine , 2015 .

[12]  M. Mancini,et al.  Sit-stand and stand-sit transitions in older adults and patients with Parkinson’s disease: event detection based on motion sensors versus force plates , 2012, Journal of NeuroEngineering and Rehabilitation.

[13]  R. Adaskevicius,et al.  Method for Recognition of the Physical Activity of Human Being Using a Wearable Accelerometer , 2014 .

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

[15]  Jorge Dias,et al.  Trajectory-based human action segmentation , 2015, Pattern Recognit..

[16]  P H Peeters,et al.  Design criteria for an automatic safety-alarm system for elderly. , 2000, Technology and health care : official journal of the European Society for Engineering and Medicine.

[17]  Billur Barshan,et al.  Detecting Falls with Wearable Sensors Using Machine Learning Techniques , 2014, Sensors.

[18]  Shaojun Zhang,et al.  An adaptive time window method for human activity recognition , 2015, CCECE.

[19]  M. Gams,et al.  Dynamic signal segmentation for activity recognition , 2011 .

[20]  Iván Pau,et al.  The Elderly’s Independent Living in Smart Homes: A Characterization of Activities and Sensing Infrastructure Survey to Facilitate Services Development , 2015, Sensors.

[21]  Richard W. Bohannon,et al.  Relationship of knee extension force to independence in sit-to-stand performance in patients receiving acute rehabilitation. , 2003, Physical therapy.

[22]  J. Allum,et al.  Gait event detection using linear accelerometers or angular velocity transducers in able-bodied and spinal-cord injured individuals. , 2006, Gait & posture.

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

[24]  Rémi Ronfard,et al.  A survey of vision-based methods for action representation, segmentation and recognition , 2011, Comput. Vis. Image Underst..

[25]  Ricard Gavaldà,et al.  Learning from Time-Changing Data with Adaptive Windowing , 2007, SDM.

[26]  Héctor Pomares,et al.  Window Size Impact in Human Activity Recognition , 2014, Sensors.

[27]  Héctor Pomares,et al.  Evaluating the effects of signal segmentation on activity recognition , 2014, IWBBIO.

[28]  Silvia Conforto,et al.  Varying behavior of different window sizes on the classification of static and dynamic physical activities from a single accelerometer. , 2015, Medical engineering & physics.

[29]  Eva Negri,et al.  Risk factors for falls in older people in nursing homes and hospitals. A systematic review and meta-analysis. , 2013, Archives of gerontology and geriatrics.

[30]  Hristijan Gjoreski,et al.  Three-layer Activity Recognition Combining Domain Knowledge and Meta-classification , 2013 .

[31]  Shuangquan Wang,et al.  b-COELM: A fast, lightweight and accurate activity recognition model for mini-wearable devices , 2014, Pervasive Mob. Comput..

[32]  Tae-Seong Kim,et al.  A Triaxial Accelerometer-Based Physical-Activity Recognition via Augmented-Signal Features and a Hierarchical Recognizer , 2010, IEEE Transactions on Information Technology in Biomedicine.

[33]  K. Aminian,et al.  Temporal feature estimation during walking using miniature accelerometers: an analysis of gait improvement after hip arthroplasty , 1999, Medical & Biological Engineering & Computing.

[34]  Lei Gao,et al.  Evaluation of accelerometer based multi-sensor versus single-sensor activity recognition systems. , 2014, Medical engineering & physics.

[35]  Rafael Morales Bueno,et al.  Learning in Environments with Unknown Dynamics: Towards more Robust Concept Learners , 2007, J. Mach. Learn. Res..

[36]  Paul Müller,et al.  Ambient Intelligence in Assisted Living: Enable Elderly People to Handle Future Interfaces , 2007, HCI.

[37]  Clemens Becker,et al.  Epidemiology of falls in residential aged care: analysis of more than 70,000 falls from residents of bavarian nursing homes. , 2012, Journal of the American Medical Directors Association.

[38]  Bin Liu,et al.  Human daily activity recognition by fusing accelerometer and multi-lead ECG data , 2013, 2013 IEEE International Conference on Signal Processing, Communication and Computing (ICSPCC 2013).

[39]  Nicholas Wickström,et al.  A Symbol-Based Approach to Gait Analysis From Acceleration Signals: Identification and Detection of Gait Events and a New Measure of Gait Symmetry , 2010, IEEE Transactions on Information Technology in Biomedicine.

[40]  Ilkka Korhonen,et al.  Detection of Daily Activities and Sports With Wearable Sensors in Controlled and Uncontrolled Conditions , 2008, IEEE Transactions on Information Technology in Biomedicine.

[41]  L. Nyberg,et al.  Patient falls in stroke rehabilitation. A challenge to rehabilitation strategies. , 1995, Stroke.

[42]  Jung-Hsien Chiang,et al.  Pattern analysis in daily physical activity data for personal health management , 2014, Pervasive Mob. Comput..

[43]  Michelle McKinney Ageing and Life Course , 2011 .

[44]  Kamiar Aminian,et al.  Spatio-temporal parameters of gait measured by an ambulatory system using miniature gyroscopes. , 2002, Journal of biomechanics.

[45]  Jesse Hoey,et al.  Sensor-Based Activity Recognition , 2012, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[46]  Kamiar Aminian,et al.  Measurement of stand-sit and sit-stand transitions using a miniature gyroscope and its application in fall risk evaluation in the elderly , 2002, IEEE Transactions on Biomedical Engineering.

[47]  Minoru Yoshizawa,et al.  Parameter exploration for response time reduction in accelerometer-based activity recognition , 2013, UbiComp.

[48]  Davide Anguita,et al.  Transition-Aware Human Activity Recognition Using Smartphones , 2016, Neurocomputing.

[49]  T Togawa,et al.  Classification of waist-acceleration signals in a continuous walking record. , 2000, Medical engineering & physics.

[50]  B. Unver,et al.  ABility to rise independently from a chair during 6-month follow-up after unilateral and bilateral total knee replacement. , 2005, Journal of rehabilitation medicine.

[51]  Yuting Zhang,et al.  Continuous functional activity monitoring based on wearable tri-axial accelerometer and gyroscope , 2011, 2011 5th International Conference on Pervasive Computing Technologies for Healthcare (PervasiveHealth) and Workshops.

[52]  Zhenyu He,et al.  Activity recognition from accelerometer signals based on Wavelet-AR model , 2010, 2010 IEEE International Conference on Progress in Informatics and Computing.

[53]  Sajal K. Das,et al.  Multimodal Wearable Sensing for Fine-Grained Activity Recognition in Healthcare , 2015, IEEE Internet Computing.

[54]  Weng-Keen Wong,et al.  Machine learning for activity recognition: hip versus wrist data , 2014, Physiological measurement.

[55]  G. Pyka,et al.  Effect of muscle strength and movement speed on the biomechanics of rising from a chair in healthy elderly and young women. , 1998, Gait & posture.

[56]  A. C. Rencher Methods of multivariate analysis , 1995 .

[57]  LarsNyberg,et al.  Patient Falls in Stroke Rehabilitation , 1995 .

[58]  Luca Benini,et al.  Wearable assistant for load monitoring: recognition of on—body load placement from gait alterations , 2010, 2010 4th International Conference on Pervasive Computing Technologies for Healthcare.

[59]  Baozhi Chen,et al.  Research challenges in computation, communication, and context awareness for ubiquitous healthcare , 2012, IEEE Communications Magazine.

[60]  Miguel A. Labrador,et al.  A Survey on Human Activity Recognition using Wearable Sensors , 2013, IEEE Communications Surveys & Tutorials.

[61]  Guang-Zhong Yang,et al.  Sensor Positioning for Activity Recognition Using Wearable Accelerometers , 2011, IEEE Transactions on Biomedical Circuits and Systems.