Deep Learning-Based Sitting Posture Recognition from Pressure Distribution Across Hard and Soft Seat Environment
Source abstract
Abstract
This study investigates the classification of nine sitting postures using pressure distribution data from hard and soft seat surfaces. Three neural network architectures (FNN, CNN, ResNet) were evaluated under single-surface and mixed-domain training regimes. While all models achieved high accuracy (>96%) when trained on a mixed-domain dataset, significant performance degradation occurred in cross-domain testing. CNN demonstrated superior capability in leveraging spatial pressure features under mixed training conditions, while FNN exhibited relatively better cross-domain robustness. Results indicate that model performance highly depends on architectural inductive biases and training data diversity. These findings underscore the importance of employing representative multi-surface datasets for ensuring generalization in practical sitting posture recognition systems.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Ergonomics and Musculoskeletal Disorders · Pressure Ulcer Prevention and Management · Non-Invasive Vital Sign Monitoring
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Laboratory of Human Kinesiology & Performance, School of Physical Education, Shenzhen University, Shenzhen 518060, China
School of Fashion and Textiles, The Hong Kong Polytechnic University, Hong Kong SAR 999077, China
Funding
National Key Research and Development Program of China
References (37)
Show source-provided references
1. Sitting posture during occupational driving causes low back pain; evidence-based position or dogma? A systematic review10.1177/0018720819871730 ↗
2. Occupational sitting behaviour and its relationship with back pain—A pilot study10.1016/j.apergo.2016.03.007 ↗
3. Guerra, B.M.V., Ramat, S., Beltrami, G., and Schmid, M. (2023). Recurrent network solutions for human posture recognition based on kinect skeletal data. Sensors, 23.10.3390/s23115260 ↗
4. Abnormal sitting posture recognition based on multi-scale spatiotemporal features of skeleton graph10.1016/j.engappai.2023.106374 ↗
5. Tang, H.-Y., Tan, S.-H., Su, T.-Y., Chiang, C.-J., and Chen, H.-H. (2021). Upper body posture recognition using inertial sensors and recurrent neural networks. Appl. Sci., 11.10.3390/app112412101 ↗
6. Machine learning algorithms application for the proposed sitting posture monitoring system10.1016/j.procs.2022.07.031 ↗
7. Sitting posture recognition using a spiking neural network10.1109/JSEN.2020.3016611 ↗
8. A comparative study on real-time sitting posture monitoring systems using pressure sensors10.2478/jee-2023-0055 ↗
9. A privacy-preserving and unobtrusive sitting posture recognition system via pressure array sensor and infrared array sensor for office workers10.1016/j.aei.2022.101690 ↗
10. Effects of different sitting postures on back shape and hip pressure
11. Kamiya, K., Kudo, M., Nonaka, H., and Toyama, J. (2008, January 8–11). Sitting posture analysis by pressure sensors. Proceedings of the 2008 19th International Conference on Pattern Recognition, Tampa, FL, USA.
12. Nadeem, M., Elbasi, E., Zreikat, A.I., and Sharsheer, M. (2024). Sitting posture recognition systems: Comprehensive literature review and analysis. Appl. Sci., 14.10.3390/app14188557 ↗
13. Ahmad, J., Andersson, H., and Sidén, J. (November, January 29). Sitting posture recognition using screen printed large area pressure sensors. Proceedings of the 2017 IEEE Sensors, Glasgow, UK.10.1109/ICSENS.2017.8233944 ↗
14. Wan, Q., Zhao, H., Li, J., and Xu, P. (2021). Hip positioning and sitting posture recognition based on human sitting pressure image. Sensors, 21.10.3390/s21020426 ↗
15. Fan, Z., Hu, X., Chen, W.-M., Zhang, D.-W., and Ma, X. (2022). A deep learning based 2-dimensional hip pressure signals analysis method for sitting posture recognition. Biomed. Signal Process Control, 73.10.1016/j.bspc.2021.103432 ↗
16. Benocci, M., Farella, E., and Benini, L. (2011, January 28–29). A context-aware smart seat. Proceedings of the 2011 4th IEEE International Workshop on Advances in Sensors and Interfaces (IWASI), Savelletri di Fasano, Italy.10.1109/IWASI.2011.6004697 ↗
17. Roh, J., Park, H., Lee, K.J., Hyeong, J., Kim, S., and Lee, B. (2018). Sitting posture monitoring systemm based on a low-cost load cell using machine learning. Sensors, 18.10.3390/s18010208 ↗
18. Katayama, H., Mizomoto, T., Rizk, H., and Yamaguchi, H. (2022, January 21–25). You work we care: Sitting posture assessment based on point cloud data. Proceedings of the 2022 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops), Pisa, Italy.10.1109/PerComWorkshops53856.2022.9767292 ↗
19. Ishac, K., and Suzuki, K. (2018). Lifechair: A conductive fabric sensor-based smart cushion for actively shaping sitting posture. Sensors, 18.10.3390/s18072261 ↗
20. Muppavram, S., Patel, N., and Nadeem, M. (2018, January 1–6). Posture alert. Proceedings of the 2018 IEEE Region Ten Symposium (Tensymp), Sydney, Australia.10.1109/TENCONSpring.2018.8692017 ↗
21. Bourahmoune, K., Ishac, K., and Amagasa, T. (2022). Intelligent posture training: Machine-learning-powered human sitting posture recognition based on a pressure-sensing IoT cushion. Sensors, 22.10.3390/s22145337 ↗
22. Ergonomics of posture-Review of various problems of standing and sitting posture10.1016/0003-6870(77)90002-3 ↗
23. The effects of the seat cushion contour and the sitting posture on surface pressure distribution and comfort during seated work10.13075/ijomeh.1896.01582 ↗
24. System to measure the use of backrest in sitting-posture office tasks10.1016/S0003-6870(99)00056-3 ↗
25. Ahn, S., Jeong, Y., Kim, D., and Kim, H. (2015, January 21–23). Development of the non-wearable system with FSR sensors for correction of sitting position. Proceedings of the 2015 Second International Conference on Computing Technology and Information Management (ICCTIM), Johor, Malaysia.10.1109/ICCTIM.2015.7224608 ↗
26. Real time visualization of asymmetrical sitting posture10.1016/j.procs.2019.08.024 ↗
27. Krauter, C., Angerbauer, K., Calepso, A.S., Achberger, A., Mayer, S., and Sedlmair, M. (2024, January 11–16). Sitting posture recognition and feedback: A literature review. Proceedings of the CHI Conference on Human Factors in Computing Systems, Honolulu, HI, USA.10.1145/3613904.3642657 ↗
28. Methods for sitting posture evaluation: Static posture and applications10.1111/j.1475-1305.2007.00334.x ↗
29. Multimodal data-based deep learning model for sitting posture recognition toward office workers’ health promotion10.1016/j.sna.2022.114150 ↗
30. Stern, L., Fernie, G., and Roshan Fekr, A. (2024). A novel in-bed body posture monitoring for decubitus ulcer prevention using body pressure distribution mapping. BioMed Eng. OnLine, 23.10.1186/s12938-024-01227-x ↗
31. Liu, S., Wang, X., and Sridhar, G. (2023, January 5–6). A case study of the effects of foam properties and support surface on seat pressure distribution. Proceedings of the Fourth International Comfort Congress, Amberg, Germany.
32. Discussion on comfort of vehicle seat cushion based on best-worst scaling method
33. Visualizing data using t-SNE
34. How to use t-SNE effectively10.23915/distill.00002 ↗
35. Kumar, R.N., Surendran, R., and Madhusundar, N. (2024, January 6–8). An intelligent human pose detection comparing with random forest and recurrent neural network. Proceedings of the 2024 8th International Conference on Electronics, Communication and Aerospace Technology (ICECA), Coimbatore, India.
36. Classification algorithm for sitting postures using weighted random forest10.1049/ipr2.70126 ↗
37. Biorthogonal wavelet trees in the classification of embedded signal classes for intelligent sensors using machine learning applications10.1016/j.jfranklin.2006.10.003 ↗