A low-complexity activity classification algorithm with optimized selection of accelerometric features

Abstract

Activity classification consists in detecting and classifying a sequence of activities, choosing from a limited set of known activities, by observing the outputs generated by (typically) inertial sensor devices placed over the body of a user. To this end, machine learning techniques can be effectively used to detect meaningful patterns from data without explicitly defining classification rules. In this paper, we present a novel Body Sensor Network (BSN)-based low complexity activity classification algorithm, which can effectively detect activities performed by the user just analyzing the accelerometric signals generated by the BSN. A preliminary (computationally intensive) training phase, performed once, is run to automatically optimize the key parameters of the algorithm used in the following (computationally light) online phase for activity classification. In particular, during the training phase, optimized subsets of nodes are selected in order to minimize the number of relevant features and keep a good compromise between performance and time complexity. Our results show that the proposed algorithm outperforms other known activity classification algorithms, especially when using a limited number of nodes, and lends itself to real-time implementation.

Keywords

Activity classification machine learning Body Sensor Networks accelerometers automatic feature selection

Get full access to this article

View all access options for this article.

References

Aylward and

Paradiso, A compact, high-speed, wearable sensor network for biomotion capture and interactive media, in: Proc. of the 6th International Conference on Information Processing in Sensor Networks (IPSN), Cambridge, MA, USA, April 2007, pp. 380–389. doi:10.1145/1236360.1236408.

Bao and

Intille, Activity recognition from user annotated acceleration data, in: Pervasive Computing, April 2004, pp. 1–17. doi:10.1007/978-3-540-24646-6_1.

Chavarriaga,

Bayati and

J.d.R.

Millán, Unsupervised adaptation for acceleration-based activity recognition: Robustness to sensor displacement and rotation, Personal and Ubiquitous Computing 17(3) (March 2013), 479–490. doi:10.1007/s00779-011-0493-y.

Chavarriaga,

Sagha,

Calatroni,

S.T.

Digumarti,

Troster,

J.d.R.

Millan and

Roggen, The opportunity challenge: A benchmark database for on-body sensor-based activity recognition, Pattern Recognition Letters 34(15) (November 2013), 2033–2042. doi:10.1016/j.patrec.2012.12.014.

R.O.

Duda,

P.E.

Hart and

D.G.

Stork, Pattern Classification and Scene Analysis, 2nd edn, Wiley-Interscience, New York, NY, USA, 2000.

Fleury,

Vacher and

Noury, SVM-based multimodal classification of activities of daily living in health smart homes: Sensors, algorithms, and first experimental results, IEEE Trans. on Information Technology in Biomedicine 14(2) (March 2010), 274–283. doi:10.1109/TITB.2009.2037317.

Giuberti and

Ferrari, Simple and robust BSN-based activity classification: Winning the first BSN contest, in: Proc. 4th International Symposium on Applied Sciences in Biomedical and Communication Technologies (ISABEL), Barcelona, Spain, October 2011.

Giuberti and

Ferrari, BSN-based activity classification: A low complexity windowing-&-classification approach, Advances in Science and Technology 85 (2013), 53–58. doi:10.4028/www.scientific.net/AST.85.53.

Guenterberg,

Ostadabbas,

Ghasemzadeh and

Jafari, An automatic segmentation technique in body sensor networks based on signal energy, in: Fourth International Conference on Body Area Networks (BodyNets), Los Angeles, CA, USA, April 2009, pp. 1–7.

10.

Huynh and

Schiele, Analyzing features for activity recognition, in: Proc. of the 2005 Joint Conference on Smart Objects and Ambient Intelligence: Innovative Context-Aware Services: Usages and Technologies, Grenoble, France, 2005, pp. 159–163.

11.

Jafari,

Li,

Bajcsy,

Glaser and

Sastry, Physical activity monitoring for assisted living at home, in: Proc. of International Workshop on Wearable and Implantable Body Sensor Networks (BSN), Aachen, Germany, March 2007, pp. 213–219. doi:10.1007/978-3-540-70994-7_37.

12.

A.K.

Jain,

R.P.W.

Duin and

Mao, Statistical pattern recognition: A review, IEEE Trans. on Pattern Analysis and Machine Intelligence 22(1) (January 2000), 4–37. doi:10.1109/34.824819.

13.

Mannini and

A.M.

Sabatini, Machine learning methods for classifying human physical activity from on-body accelerometers, Sensors 10(2) (February 2010), 1154–1175. doi:10.3390/s100201154.

14.

Opportunity Dataset, http://archive.ics.uci.edu/ml/datasets/OPPORTUNITY+Activity+Recognition.

15.

Patel,

Park,

Bonato,

Chan and

Rodgers, A review of wearable sensors and systems with application in rehabilitation, Journal of NeuroEngineering and Rehabilitation 9(21) (April 2012), 1–17.

16.

Pirttikangas,

Fujinami and

Nakajima, Feature selection and activity recognition from wearable sensors, in: International Symposium on Ubiquitous Computing Systems, Seoul, Korea, October 2006, pp. 516–527. doi:10.1007/11890348_39.

17.

S.J.

Preece,

J.Y.

Goulermas,

L.P.J.

Kenney,

Howard,

Meijer and

Crompton, Activity identification using body-mounted sensors – A review of classification techniques, Physiological Measurement 30(4) (April 2009), R1. doi:10.1088/0967-3334/30/4/R01.

18.

Roggen,

Calatroni,

Rossi,

Holleczek,

Forster,

Troster,

Lukowicz,

Bannach,

Pirkl,

Ferscha,

Doppler,

Holzmann,

Kurz,

Holl,

Chavarriaga,

Sagha,

Bayati,

Creatura and

J.d.R.

Millan, Collecting complex activity datasets in highly rich networked sensor environments, in: 2010 7th International Conference on Networked Sensing Systems (INSS), Kassel, Germany, June 2010, pp. 233–240. doi:10.1109/INSS.2010.5573462.

19.

Sagha,

J.d.R.

Millán and

Chavarriaga, Detecting anomalies to improve classification performance in opportunistic sensor networks, in: Proc. of 2011 IEEE International Conference on Pervasive Computing and Communications Workshops (PERCOM Workshops), Seattle, WA, USA, March 2011, pp. 154–159. doi:10.1109/PERCOMW.2011.5766860.

20.

D.M.

Sherrill,

M.L.

Moy,

J.J.

Reilly and

Bonato, Using hierarchical clustering methods to classify motor activities of copd patients from wearable sensor data, Journal of NeuroEngineering and Rehabilitation 2(1) (June 2005), 1–14. doi:10.1186/1743-0003-2-16.

21.

J.A.

Ward,

Lukowicz,

Tröster and

T.E.

Starner, Activity recognition of assembly tasks using body-worn microphones and accelerometers, IEEE Transactions on Pattern Analysis and Machine Intelligence 28(10) (October 2006), 1553–1567. doi:10.1109/TPAMI.2006.197.

22.

Yang,

Iyengar,

S.S.

Sastry,

Bajcsy,

Kuryloski and

Jafari, Distributed segmentation and classification of human actions using a wearable motion sensor network, in: Proc. of the Computer Vision and Pattern Recognition Workshops on Human Communicative Behavior Analysis (CVPRW), Anchorage, AK, USA, June 2008, pp. 1–8.

23.

Zappi,

Stiefmeier,

Farella,

Roggen,

Benini and

Tröster, Activity recognition from on-body sensors by classifier fusion: Sensor scalability and robustness, in: Proc. of 3rd International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP), Melbourne, Australia, December 2007, pp. 281–286.