Smartphone-Based Human Sitting Behaviors Recognition Using Inertial Sensor
Abstract
:1. Introduction
- The inertial sensor in a smartphone is helpful in monitoring the sitting behaviors of office workers continuously.
- Determining the time spent on different sitting behaviors, whether correct or incorrect.
- The user can detect the correct and incorrect sitting behaviors.
2. Framework of Smartphone-Based Sitting Detection
2.1. Data Collection and Preprocessing
- Left movement;
- Right movement;
- Front movement;
- Back movement;
- Straight movement.
Hardware Platform
2.2. Feature Extraction
2.2.1. Morphological Features
2.2.2. Entropy-Based Features
2.3. Feature Subset Selection
2.4. Sitting Behavior Recognition Techniques
3. Results and Discussion
3.1. Performance Analysis of Classifiers with Feature Selection of Accelerometer, Gyroscope, and Magnetometer
3.2. Performance Analysis of the Classifiers with Feature Selection of Accelerometer and Gyroscope
3.3. Performance Analysis of the Classifiers with Feature Selection of Accelerometer
3.4. Analysis of Results
4. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
Abbreviations
CT | Computerized tomography |
DT | Decision tree |
FDA | Flexible discriminant analysis |
FSR | Force-sensing resistor |
HM | Harmonic mean |
HSB | Human sitting behaviors |
IMU | Inertial measurement unit |
kNN | k-nearest neighbors |
MPU | Multi-core processing unit |
PCA | Principal component analysis |
RBF | Radial basis function |
RMS | Root mean square |
SD | Standard deviation |
SPMS | Sitting posture monitoring system |
SSI | Simple squared integral |
SVM | Support vector machine |
References
- Haynes, S.; Williams, K. Impact of seating posture on user comfort and typing performance for people with chronic low back pain. Int. J. Ind. Ergon. 2008, 38, 35–46. [Google Scholar] [CrossRef]
- Lis, A.M.; Black, K.M.; Korn, H.; Nordin, M. Association between sitting and occupational LBP. Eur. Spine J. 2007, 16, 283–298. [Google Scholar] [CrossRef] [Green Version]
- Magnæs, B. Body position and cerebrospinal fluid pressure. J. Neurosurg. 1976, 44, 687–697. [Google Scholar] [CrossRef]
- Xu, W.; Huang, M.C.; Amini, N.; He, L.; Sarrafzadeh, M. Ecushion: A textile pressure sensor array design and calibration for sitting posture analysis. IEEE Sens. J. 2013, 10, 3926–3934. [Google Scholar] [CrossRef]
- Ma, C.; Li, W.; Gravina, R.; Cao, J.; Li, Q.; Fortino, G. Activity level assessment using a smart cushion for people with a sedentary lifestyle. Sensors 2017, 17, 2269. [Google Scholar] [CrossRef] [Green Version]
- Roh, J.; Park, H.J.; Lee, K.J.; Hyeong, J.; Kim, S.; Lee, B. Sitting posture monitoring system based on a low-cost load cell using machine learning. Sensors 2018, 18, 208. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Xipei, R.; Yu, B.; Lu, Y.; Chen, Y.; Pu, P. Health—Sit: Designing posture based interaction to promote exercise during fitness breaks. Int. J. Hum. Com Puter. Interact. 2019, 35, 870–885. [Google Scholar] [CrossRef] [Green Version]
- Robertson, M.; Amick, B.C., III; DeRango, K.; Rooney, T.; Bazzani, L.; Harrist, R.; Moore, A. The effects of an office ergonomics training and chair intervention on worker knowledge, behavior and musculoskeletal risk. Appl. Ergon. 2009, 40, 124–135. [Google Scholar] [CrossRef] [PubMed]
- Chanchai, W.; Songkham, W.; Ketsomporn, P.; Sappakitchanchai, P.; Siriwong, W.; Robson, M.G. The impact of ergonomics intervention on psychosocial factors and musculoskeletal symptoms among office workers. Int. J. Ind. Ergon. 2011, 41, 671–676. [Google Scholar]
- Goossens, R.H.M.; Netten, M.P.; Van der Doelen, B. An offiice chair to influence the sitting behavior of office workers. Work 2012, 41, 2086–2088. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Menndez, C.C.; Amick, B.C., III; Robertson, M.; Bazzani, L.; DeRango, K.; Rooney, T.; Moore, A. A replicated field intervention study evaluating the impact of a highly adjustable chair and office ergonomics training on visual symptoms. Appl. Ergon. 2012, 43, 639–644. [Google Scholar] [CrossRef] [Green Version]
- Taieb-Maimon, M.; Cwikel, J.; Shapira, B.; Orenstein, I. The effectiveness of a training method using selfmoelling webcam photos for reducing musculoskeletal risk among office workers using computers. Appl. Ergon. 2012, 43, 376–385. [Google Scholar] [CrossRef] [PubMed]
- Garcia, S.; Derrac, J.; Cano, J.-R.; Herrera, F. Prototype Selection for Nearest Neighbor Classification: Taxonomy and Empirical Study. IEEE Trans. Pattern Anal. Mach. Intell. 2012, 34, 417–435. [Google Scholar] [CrossRef] [PubMed]
- Theodoridis, S.; Pikrakis, A.; Koutroumbas, K.; Cavouras, D. Introduction to Pattern Recognition: A Matlab Approach; Academic Press: Waltham, NA, USA, 2010. [Google Scholar]
- Duda, R.O.; Hart, P.E.; Stork, D.G. Pattern classification. Int. J. Comput. Intell. Appl. 2001, 1, 335–339. [Google Scholar]
- Zhang, S. KNN-CF Approach: Incorporating Certainty Factor to kNN Classification. IEEE Intell. Inform. Bull. 2010, 1, 24–33. [Google Scholar]
- Zhang, S.; Li, X.; Zong, M.; Zhu, X.; Wang, R. Efficient kNN Classification with Different Numbers of Nearest Neighbors. IEEE Trans. Neural. Netw. Learn. Syst. 2017, 29, 1774–1785. [Google Scholar] [CrossRef] [PubMed]
- Rosero-Montalvo, P.D.; Peluffo-Ordonez, D.H.; Batista, V.F.L.; Serrano, J.; Rosero, E.A. Intelligent system for identification of wheel chair users posture using machine learning techniques. IEEE Sens. J. 2018, 19, 1936–1942. [Google Scholar] [CrossRef]
- Chang, C.; Lin, C. Libsvm: A library for support vector machines. ACM Trans. Intell. Syst. Technol. 2011, 2, 1–27. [Google Scholar] [CrossRef]
- Hall, M.A. Correlation-Based Feature Selection for Machine Learning. The University of Waikato. Available online: https://www.cs.waikato.ac.nz/~ml/publications/1999/99MH-Thesis.pdf (accessed on 25 September 2021).
- Burges, C.J. A Tutorial on Support Vector Machines for Pattern Recognition. Data Min. Knowl. Discov. 1998, 2, 121–167. [Google Scholar] [CrossRef]
- Cortes, C.; Vapnik, V. Support vector networks. Mach. Learn. 1995, 20, 273–297. [Google Scholar] [CrossRef]
- Gaglio, S.; Re, G.L.; Morana, M. Human Activity Recognition Process Using 3-D Posture Data. IEEE Trans. Hum.-Mach. Syst. 2014, 45, 586–597. [Google Scholar] [CrossRef]
- Miron, C.; Pasarica, A.; Costin, H.; Manta, V.; Timofte, R.; Ciucu, R. Hand Gesture Recognition based on SVM Classification. In Proceedings of the 2019 E-Health and Bioengineering Conference (EHB), Iasi, Romania, 21–23 November 2019. [Google Scholar]
- Attal, F.; Mohammed, S.; Dedabrishvili, M.; Chamroukhi, F.; Oukhellou, L.; Amirat, Y. Physical Human Activity Recognition Using Wearable Sensors. Sensors 2015, 15, 31314–31338. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Vergara, M.; Page, Á. System to measure the use of the backrest in sitting-posture office tasks. Appl. Ergon. 2000, 31, 247–254. [Google Scholar] [CrossRef]
- Zemp, R.; Fliesser, M.; Wippert, P.-M.; Taylor, W.R.; Lorenzetti, S. Occupational sitting behaviour and its relationship with back pain—A pilot study. Appl. Ergon. 2016, 56, 84–91. [Google Scholar] [CrossRef] [Green Version]
- Grandjean, E.; Hünting, W. Ergonomics of posture—Review of various problems of standing and sitting posture. Appl. Ergon. 1977, 8, 135–140. [Google Scholar] [CrossRef]
- Hyeong, J.-H.; Roh, J.-R.; Park, S.-B.; Kim, S.; Chung, K.-R. A Trend Analysis of Dynamic Chair and Applied Technology. J. Ergon. Soc. Korea 2014, 33, 267–279. [Google Scholar] [CrossRef] [Green Version]
- Wang, S.J.; Sommer, B.; Cheng, W.; Schreiber, F. The Virtual-Spine Platform—Acquiring, visualizing, and analyzing individual sitting behavior. PLoS ONE 2018, 13, e0195670. [Google Scholar] [CrossRef] [PubMed]
- Bortone, I.; Argentiero, A.; Agnello, N.; Sabato, S.S.; Bucciero, A. A two-stage approach to bring the postural assessment to masses: The KISS-Health Project. In Proceedings of the IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI), Valencia, Spain, 1–4 June 2014; pp. 371–374. [Google Scholar]
- Bortone, I.; Argentiero, A.; Agnello, N.; Denetto, V.; Neglia, C.; Benvenuto, M. The PoSE Project: An Innovative Approach to PromoteHealthy Postures in Schoolchildren. In Proceedings of the International Conference on E-Learning, E-Education, and Online Training, Bethesda, MD, USA, 18−20 September 2014. [Google Scholar]
- Arif, M.; Ahmed, K. Physical activities monitoring using wearable acceleration sensors attached to the body. PLoS ONE 2015, 10, e0130851. [Google Scholar] [CrossRef] [Green Version]
- Zdemir, A.; Billur, B. Detecting falls with wearable sensors using machine learning techniques. Sensors 2014, 14, 10691–10708. [Google Scholar] [CrossRef]
- Benocci, M.; Farella, E.; Benini, L.A. Context-Aware Smart Seat. In Proceedings of the 2011 4th IEEE International Workshop on Advances in Sensors and Interfaces (IWASI), Savelletri diFasano, Italy, 28–29 June 2011; pp. 104–109. [Google Scholar]
- Zhang, S.; Mccullagh, P.; Nugent, C.; Zheng, H.; Baumgarten, M. Optimal model selection for posture recognition in home-based healthcare. Int. J. Mach. Learn. Cybern. 2010, 2, 1–14. [Google Scholar] [CrossRef]
- OnePlus 6 specs, Oneplus.in. 2021. Available online: https://www.oneplus.in/6/specs (accessed on 21 September 2021).
- Arif, M.; Mohsin, B.; Ahmed, K.; Ahamed, S.I. Better physical activity classification using smart phone acceleration sensor. J. Med Syst. 2014, 38, 95. [Google Scholar] [CrossRef]
- Phinyomark, A.; Hirunviriya, S.; Limsakul, C.; Phukpattaranont, P. Evaluation of EMG feature extraction for hand movement recognition based on Euclidean distance and standard deviation. In Proceedings of the ECTI-CON2010: The 2010 ECTI International Confernce on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, Chiang Mai, Thailand, 19–21 May 2010; pp. 856–860. [Google Scholar]
- Mannini, A.; Sabatini, A.M. Machine learning methods for classifying human physical activity from on body accelerometers. Sensors 2010, 10, 1154–1175. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Hall, M.; Eibe, F.; Holmes, G.; Pfahringer, B.; Reutemann, P.; Witten, I.H. The WEKA data mining software: An update. ACM SIGKDD Explor. Newsl. 2009, 11, 10–18. [Google Scholar] [CrossRef]
- Hall, M.A.; Smith, L.A. Feature Selection for Machine Learning: Comparing a Correlation-Based Filter Approach to the Wrapper. In Proceedings of the FLAIRS Conference, Orlando, FL, USA, 1–5 May 1999; pp. 235–239. [Google Scholar]
- Aha, D.W.; Kibler, D.; Albert, M.K. Instance based learning algorithms. Mach. Learn. 1991, 6, 37–66. [Google Scholar] [CrossRef] [Green Version]
- Russell, E.; Kennedy, J. Particle swarm optimization. In Proceedings of the IEEE International Conference On Neural Networks, Citeseer, Perth, WA, Australia, 27 November–1 December 1995; pp. 1942–1948. [Google Scholar]
- Sinha, V.K.; Maurya, A.K. Calibration of Inertial Sensor by Using Particle Swarm Optimization and Human Opinion Dynamics Algorithm. Int. J. Instrum. Control. Syst. 2017, 47, 777–778. [Google Scholar] [CrossRef]
- Reddy, S.; Mun, M.; Burke, J.; Estrin, D.; Hansen, M.; Srivastava, M. Using mobile phones to determine transportation modes. ACM Trans. Sens. Netw. 2010, 6, 1–27. [Google Scholar] [CrossRef]
- VenkataPhanikrishna, B.; Pławiak, P.; Jaya Prakash, A. A Brief Review on EEG Signal Pre-processing Techniques for Real-Time Brain-Computer Interface Applications. TechRxiv. Preprint. 2021. Available online: https://doi.org/10.36227/techrxiv.16691605.v1 (accessed on 20 September 2021).
S. No | Authors | Type of Sensors | Classifiers | Accuracy (%) | Limitations |
---|---|---|---|---|---|
1 | Xu, Wenyao et al. [4] | Textile sensor array in a smart cushion chair | Naïve Bayes network | 85.90 | The recognition rate is less. |
2 | Roh, Jongryun et al. [6] | Low-cost load cells (P0236-I42) | SVM using RBF kernel | 97.20 | No. of subjects used is less, and power consumption is more. |
3 | Taieb-Maimon, Meirav et al. [12] | Webcam, Rapid Upper Limb Assessment (RULA) tool. | Sliced inverse regression | 86.0 | Analyzed only three symptom scales as back symptoms, arm symptoms, and neck pain severity. |
4 | Arif, Muhammad et al. [33] | Colibri wireless IMU | kNN | 97.90 | Dataset tested is small, and the optimal set of sensors need to be placed at the appropriate locations on the body. |
5 | Zdemir et al. [34] | The MTw sensor unit, MTw software development kit | Random forest | 90.90 | Cost is high, and the convergence time is more. |
6 | Rosero-Montalvo et al. [18] | Ultrasonic sensor, pressure sensor, Arduinonano, LiPobattery | kNN | 75.0 | Accuracy reported is much less. |
7 | Benocci et al. [35] | FSR, digital magnetometer, accelerometer | kNN | 92.70 | The number of subjects used in the experiment is less. |
8 | Shumei Zhang et al. [36] | HTC smartphone (HD8282) | kNN | 92.70 | A posture-aware reminder system can be attached. |
S. No | Physical Activities | No. of Instances | Time (in Seconds) |
---|---|---|---|
1 | A1: Left movement | 35,565 | 712 |
2 | A2: Right movement | 37,757 | 756 |
3 | A3: Front movement | 33,268 | 665 |
4 | A4: Back movement | 29,460 | 590 |
5 | A5: Straight movement | 27,451 | 549 |
Total | 163,501 | 3272 |
S. No | Selected Features | S. No | Selected Features |
---|---|---|---|
1 | Total-acceleration | 15 | Z-magnetometer-SD |
2 | Total-magnetometer | 16 | Z-accelerometer-skewness |
3 | Y-accelerometer-MAV | 17 | X-gyroscope-skewness |
4 | X-gyroscope-MAV | 18 | Y-gyroscope-skewness |
5 | Y-gyroscope-MAV | 19 | Z-gyroscope-skewness |
6 | Y-magnetometer-MAV | 20 | Y-magnetometer-skewness |
7 | X-accelerometer-HM | 21 | Y-accelerometer-LEE |
8 | X-gyroscope-HM | 22 | Y-magnetometer-LEE |
9 | Y-accelerometer-Var | 23 | X-gyroscope-SSI |
10 | Z-accelerometer-Var | 24 | X-accelerometer-WE |
11 | Z-magnetometer-Var | 25 | X-gyroscope-WE |
12 | X-gyroscope-SD | 26 | X-magnetometer-WE |
13 | Y-gyroscope-SD | 27 | Y-magnetometer-WE |
14 | X-magnetometer-SD |
S. No | Activities | KNN (K = 3) | KNN (K = 5) | KNN (K = 7) | KNN (K = 11) | SVM | Naive Bayes |
---|---|---|---|---|---|---|---|
1 | A1: Left movement | 99.20 | 99.91 | 99.91 | 99.92 | 99.99 | 98.51 |
2 | A2: Right movement | 99.97 | 99.97 | 99.97 | 99.95 | 99.98 | 99.06 |
3 | A3: Front movement | 99.96 | 99.96 | 99.96 | 99.94 | 99.98 | 99.15 |
4 | A4: Back movement | 99.67 | 99.72 | 99.70 | 99.68 | 99.76 | 91.89 |
5 | A5: Straight movement | 99.89 | 99.58 | 98.31 | 97.68 | 99.77 | 98.71 |
S. No | Activity | A1 | A2 | A3 | A4 | A5 | |
---|---|---|---|---|---|---|---|
True Class | 1 | A1 | 10,669 | 0 | 0 | 0 | 1 |
2 | A2 | 1 | 11,476 | 0 | 0 | 0 | |
3 | A3 | 1 | 0 | 10,055 | 1 | 0 | |
4 | A4 | 0 | 0 | 0 | 8762 | 21 | |
5 | A5 | 0 | 0 | 2 | 16 | 8044 | |
Predicted Class |
S. No | Activities | KNN (K = 3) | KNN (K = 5) | KNN (K = 7) | KNN (K = 11) | SVM | Naive Bayes |
---|---|---|---|---|---|---|---|
1 | A1: Left Movement | 98.24 | 97.88 | 97.56 | 96.95 | 97.10 | 84.55 |
2 | A2: Right Movement | 99.78 | 99.76 | 99.73 | 99.69 | 99.88 | 99.37 |
3 | A3: Front Movement | 99.55 | 99.53 | 99.52 | 99.51 | 99.66 | 92.43 |
4 | A4: Back Movement | 98.95 | 99.08 | 99.17 | 99.26 | 98.78 | 85.66 |
5 | A5: Straight Movement | 95.87 | 95.12 | 94.57 | 93.51 | 98.31 | 95.31 |
S. No | Activity | A1 | A2 | A3 | A4 | A5 | |
---|---|---|---|---|---|---|---|
True Class | 1 | A1 | 10,341 | 7 | 108 | 8 | 185 |
2 | A2 | 3 | 11,460 | 3 | 2 | 5 | |
3 | A3 | 130 | 2 | 10,038 | 1 | 2 | |
4 | A4 | 1 | 1 | 0 | 8746 | 106 | |
5 | A5 | 69 | 4 | 1 | 59 | 7868 | |
Predicted Class |
S. No | Activities | KNN (K = 3) | KNN (K = 5) | KNN (K = 7) | KNN (K = 11) | SVM | Naive Bayes |
---|---|---|---|---|---|---|---|
1 | A1: Left movement | 99.63 | 99.57 | 99.50 | 99.46 | 98.94 | 82.18 |
2 | A2: Right movement | 99.95 | 99.93 | 99.92 | 99.91 | 99.89 | 99.80 |
3 | A3: Front movement | 99.90 | 99.88 | 99.87 | 99.87 | 99.84 | 90.62 |
4 | A4: Back movement | 99.68 | 99.72 | 99.70 | 99.75 | 99.60 | 91.06 |
5 | A5: Straight movement | 99.31 | 99.20 | 99.15 | 98.88 | 99.30 | 96.28 |
S. No | Activity | A1 | A2 | A3 | A4 | A5 | |
---|---|---|---|---|---|---|---|
True Class | 1 | A1 | 10,634 | 6 | 18 | 1 | 14 |
2 | A2 | 3 | 11,467 | 1 | 0 | 1 | |
3 | A3 | 9 | 0 | 10,049 | 0 | 1 | |
4 | A4 | 1 | 0 | 0 | 8769 | 127 | |
5 | A5 | 15 | 1 | 0 | 39 | 7994 | |
Predicted Class |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Sinha, V.K.; Patro, K.K.; Pławiak, P.; Prakash, A.J. Smartphone-Based Human Sitting Behaviors Recognition Using Inertial Sensor. Sensors 2021, 21, 6652. https://doi.org/10.3390/s21196652
Sinha VK, Patro KK, Pławiak P, Prakash AJ. Smartphone-Based Human Sitting Behaviors Recognition Using Inertial Sensor. Sensors. 2021; 21(19):6652. https://doi.org/10.3390/s21196652
Chicago/Turabian StyleSinha, Vikas Kumar, Kiran Kumar Patro, Paweł Pławiak, and Allam Jaya Prakash. 2021. "Smartphone-Based Human Sitting Behaviors Recognition Using Inertial Sensor" Sensors 21, no. 19: 6652. https://doi.org/10.3390/s21196652
APA StyleSinha, V. K., Patro, K. K., Pławiak, P., & Prakash, A. J. (2021). Smartphone-Based Human Sitting Behaviors Recognition Using Inertial Sensor. Sensors, 21(19), 6652. https://doi.org/10.3390/s21196652