A multi-level refinement approach towards the classification of quotidian activities using accelerometer data

Wearable inertial measurement units incorporating accelerometers and gyroscopes are increasingly used for activity analysis and recognition. In this paper an activity classification algorithm is presented which includes a novel multi-step refinement with the aim of improving the classification accuracy of traditional approaches. To do so, after the classification takes place, information is extracted from the confusion matrix to focus the computational efforts on those activities with worse classification performance. It is argued that activities differ diversely from each other, therefore a specific set of features may be informative to classify a specific set of activities, but such informativeness should not necessarily be extended to a different activity set. This approach has shown promising results, achieving important classification accuracy improvements.

[1]  Jin Wang,et al.  Recognizing Human Daily Activities From Accelerometer Signal , 2011 .

[2]  Juan M. Santos-Gago,et al.  How do you sleep? Using off the shelf wrist wearables to estimate sleep quality, sleepiness level, chronotype and sleep regularity indicators , 2018, J. Ambient Intell. Humaniz. Comput..

[3]  Subhas Mukhopadhyay,et al.  Determining Wellness through an Ambient Assisted Living Environment , 2014, IEEE Intelligent Systems.

[4]  Sabine Van Huffel,et al.  Accelerometry-Based Activity Recognition and Assessment in Rheumatic and Musculoskeletal Diseases , 2016, Sensors.

[5]  S. Shankar Sastry,et al.  Physical Activity Monitoring for Assisted Living at Home , 2007, BSN.

[6]  Hitoshi Ogawa,et al.  Human Activity Recognition System Including Smartphone Position , 2014 .

[7]  Hasan Ogul,et al.  Integrating Features for Accelerometer-based Activity Recognition , 2016, EUSPN/ICTH.

[8]  Ahmad Lotfi,et al.  Human Movement Recognition Based on the Stochastic Characterisation of Acceleration Data , 2016, Sensors.

[9]  Duc A. Tran,et al.  The 11th International Conference on Mobile Systems and Pervasive Computing (MobiSPC-2014) A Study on Human Activity Recognition Using Accelerometer Data from Smartphones , 2014 .

[10]  Ben J Smith,et al.  Screening for physical activity in family practice: evaluation of two brief assessment tools. , 2005, American journal of preventive medicine.

[11]  Angelo M. Sabatini,et al.  Machine Learning Methods for Classifying Human Physical Activity from On-Body Accelerometers , 2010, Sensors.

[12]  Mi Zhang,et al.  Motion primitive-based human activity recognition using a bag-of-features approach , 2012, IHI '12.

[13]  Changseok Bae,et al.  Unsupervised learning for human activity recognition using smartphone sensors , 2014, Expert Syst. Appl..

[14]  Tullio Vernazza,et al.  Analysis of human behavior recognition algorithms based on acceleration data , 2013, 2013 IEEE International Conference on Robotics and Automation.

[15]  Stefan Woltran,et al.  Generalizations of Dung Frameworks and Their Role in Formal Argumentation , 2014, IEEE Intelligent Systems.

[16]  Tara O’Brien,et al.  Acceptability of wristband activity trackers among community dwelling older adults. , 2015, Geriatric nursing.

[17]  Gerhard Tröster,et al.  Probabilistic parsing of dietary activity events , 2007, BSN.

[18]  Gerhard Tröster,et al.  Recognition of dietary activity events using on-body sensors , 2008, Artif. Intell. Medicine.

[19]  Daniela Micucci,et al.  Falls as anomalies? An experimental evaluation using smartphone accelerometer data , 2015, J. Ambient Intell. Humaniz. Comput..

[20]  Paul Lukowicz,et al.  Towards wearable sensing-based assessment of fluid intake , 2010, 2010 8th IEEE International Conference on Pervasive Computing and Communications Workshops (PERCOM Workshops).

[21]  Oliver Amftand.  Recognition of dietary activity events using on-body sensors , 2008 .

[22]  Edward D. Lemaire,et al.  Feature Selection for Wearable Smartphone-Based Human Activity Recognition with Able bodied, Elderly, and Stroke Patients , 2015, PloS one.

[23]  Davide Anguita,et al.  A Public Domain Dataset for Human Activity Recognition using Smartphones , 2013, ESANN.

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

[25]  Petia Radeva,et al.  Human Activity Recognition from Accelerometer Data Using a Wearable Device , 2011, IbPRIA.

[26]  Tibert Verhagen,et al.  Health empowerment through activity trackers: An empirical smart wristband study , 2016, Comput. Hum. Behav..

[27]  Lei Wang,et al.  Analysis of filtering methods for 3D acceleration signals in body sensor network , 2011, International Symposium on Bioelectronics and Bioinformations 2011.

[28]  G M Lyons,et al.  Comparison of the performance of the activPAL Professional physical activity logger to a discrete accelerometer-based activity monitor. , 2007, Medical engineering & physics.

[29]  Ramón F. Brena,et al.  Long-Term Activity Recognition from Accelerometer Data , 2013 .

[30]  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.

[31]  Yujie Dong,et al.  Detecting Periods of Eating During Free-Living by Tracking Wrist Motion , 2014, IEEE Journal of Biomedical and Health Informatics.

[32]  Ritu Gupta,et al.  Classification of team sport activities using a single wearable tracking device. , 2015, Journal of biomechanics.

[33]  Jozsef Suto,et al.  Efficiency investigation of artificial neural networks in human activity recognition , 2017, Journal of Ambient Intelligence and Humanized Computing.

[34]  Stefano Cagnoni,et al.  Multi-sensor system for detection and classification of human activities , 2011, Journal of Ambient Intelligence and Humanized Computing.