Mobile-Centric Supervised Machine Learning Approach for Elderly Fall Detection Using YOLOv8
Article Sidebar
Main Article Content
Abstract
The global surge in the elderly population has underscored the need for advanced safety systems tailored to seniors, particularly those living independently. Central to these systems is the pivotal role of fall detection in ensuring their welfare. This paper presents a cutting-edge fall detection system designed specifically for the elderly, leveraging supervised machine learning techniques with a mobile-centric approach. Departing from traditional hospital-centric setups, our system offers cost-effectiveness and improved mobility, facilitating deployment across diverse environments. The methodology comprises three core stages: data collection and annotation, model training, and inference. We curated a dataset of 1500 images categorized into three classes: standing, falling, and fallen, meticulously annotated using RoboFlow. Subsequent model training utilized YOLOv8, culminating in the inference stage, which underwent quantitative evaluation employing 10-fold cross-validation, yielding an average accuracy of 97.88%. Qualitative assessment across four distinct scenarios further validated our system, achieving an average accuracy of 95.92%. These results underscore the efficacy of our approach and lay the foundation for practical implementation and widespread adoption. Subsequent to the successful development of the core algorithm, we operationalized it for real-world applications by seamlessly integrating it with smartphones via TensorFlowLite. This integration underscores the synergy between algorithm design and software development, further facilitating the practical deployment and widespread acceptance of our system in diverse settings
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
References
Kumar, V., Badal, N., & Mishra, R., (2021). Elderly fall detection using IoT and image processing. Journal of Discrete Mathematical Sciences and Cryptography, 24(3), 681-695. https://doi.org/10.1080/09720529.2019.1692451
Alvarez, J., Li, C., & Philips, A. (2021). Enhancing fall detection accuracy using wearable sensors and machine learning algorithms. Journal of Assistive Technologies, 12(3), 217-230. https://doi.org/10.1108/JAT-12-2020-0078
Chen, Y., Yu, H., & Wang, Y. (2020). A wearable sensor-based fall detection system using machine learning algorithms. IEEE Sensors Journal, 20(15), 8765-8773. https://doi.org/10.1109/JSEN.2020.2993583
Lopes, I. M., Rodrigues, J. J., & Pławiak, P. (2020). A review of wearable solutions for elderly fall detection. Sensors, 20(3), Article 829.
https://doi.org/10.3390/s20030829
Kim, T., & Kim, M. S. (2021). Real-time fall detection using machine learning algorithms and wearable sensors. Sensors, 21(6), Article 2086. https://doi.org/10.3390/s21062086
Ma, X., Zhang, T., & Li, M. (2020). RGB-D camera-based fall detection system using deep learning. Journal of Visual Communication and Image Representation, 69, 103010. https://doi.org/10.1016/j.jvcir.2020.103010
Palacios-Navarro, G., Carrasco-Jiménez, J. C., & Pérez-Cisneros, M. (2021). Fall detection system using thermal and depth information fusion. Sensors, 21(2), Article 429. https://doi.org/10.3390/s21020429
Bouloudi, F., Charfi, I., & Soudani, A. (2019). Real-time fall detection using Kinect depth images and a support vector machine. Sensors, 19(11), Article 2457. https://doi.org/10.3390/s19112457
Yu, Z., Wang, Q., & Hao, K. (2020). Human fall detection system based on RGB-D cameras and convolutional neural networks. Journal of Ambient Intelligence and Humanized Computing, 11(5), 1897-1906. https://doi.org/10.1007/s12652-019-01322-3
Khan, M. A., & Porikli, F. (2019). Acoustic fall detection using deep learning-based sound analysis. Pattern Recognition, 93, 195-205.
https://doi.org/10.1016/j.patcog.2019.05.017
Grzeszick, R., & Jager, F. (2020). Acoustic-based fall detection using convolutional neural networks. Applied Acoustics, 164, 107315.
https://doi.org/10.1016/j.apacoust.2020.107315
Wang, Z., & Liu, Y. (2020). Fall detection using acoustic signals and machine learning techniques. IEEE Sensors Journal, 20(7), 3661-3670. https://doi.org/10.1109/JSEN.2019.2951383
Pannurat, N., Nantajeewarawat, E., & Haddawy, P. (2020). Fall detection using environmental sound signals and machine learning. Sensors, 20(6), Article 1746. https://doi.org/10.3390/s20061746
Miao, F., Zhang, X., & Wang, W. (2019). Radar-based fall detection using machine learning algorithms. IEEE Transactions on Biomedical Engineering, 66(2), 419-428. https://doi.org/10.1109/TBME.2018.2839078
Zhou, K., He, Y., & Wu, F. (2020). Ultra-wideband radar for fall detection: Signal processing and system design. Digital Signal Processing, 98, 102647. https://doi.org/10.1016/j.dsp.2019.102647
Wang, Z., & Guo, L. (2020). Fall detection using ultra-wideband radar and deep learning. IEEE Access, 8, 53657-53666. https://doi.org/10.1109/ACCESS.2020.2986087
Mahmood, A., & Hassan, M. (2020). Frequency-modulated continuous-wave radar-based fall detection using deep learning. IEEE Transactions on Industrial Informatics, 16(7), 4361-4370. https://doi.org/10.1109/TII.2019.2953813
Alsheikh, M. A., & Selim, M. M. (2020). Hybrid fall detection system using wearable sensors and RGB-D cameras. Journal of Ambient Intelligence and Humanized Computing, 11(10), 4427-4439. https://doi.org/10.1007/s12652-019-01344-x
Mahmud, M. S., & Wang, X. (2019). A review of sensor fusion techniques for fall detection. IEEE Access, 7, 48819-48833. https://doi.org/10.1109/ACCESS.2019.2907910
Eskofier, B. M., Lee, S. I., & Kupnik, M. (2020). Sensor fusion and machine learning for robust fall detection in real-world environments. Sensors, 20(7), Article 1901. https://doi.org/10.3390/s20071901
Hossain, M. S., Muhammad, G., & Alhamid, M. F. (2020). Hybrid fall detection system using wearable sensors and ambient sensors. IEEE Access, 8, 132836-132847. https://doi.org/10.1109/ACCESS.2020.3015645
He, T., & Wang, H. (2020). Multi-sensor fusion for fall detection using deep learning techniques. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 28(4), 922-931. https://doi.org/10.1109/TNSRE.2020.2978803
Luo, B. (2023). Human fall detection for smart home caring using YOLO networks. International Journal of Advanced Computer Science and Applications, 14(4), 59-68. https://doi.org/10.14569/IJACSA.2023.0140409
Gao, P. (2023). Development of YOLO-based model for fall detection in IoT smart home applications. International Journal of Advanced Computer Science and Applications, 14(10), 1118-1125. https://dx.doi.org/10.14569/IJACSA.2023.01410117
Kan, X., Zhu, S., Zhang, Y., & Qian, C. (2023). A lightweight human fall detection network. Sensors, 23(22), 9069, 1-20. https://doi.org/10.3390/s23229069
Wang, Y., Chi, Z., Liu, M., Li, G., & Ding, S. (2023). High-performance lightweight fall detection with an improved YOLOv5s algorithm. Machines, 11(8), 818, 1-17. https://doi.org/10.3390/machines11080818
Gomes, M. E. N., Macêdo, D., Zanchettin, C., de-Mattos-Neto, P. S. G., & Oliveira, A. (2022). Multi-human fall detection and localization in videos. Computer Vision and Image Understanding, 220, 103442, 1-13. https://doi.org/10.1016/j.cviu.2022.103442
Raza, A., Yousaf, M. H., & Velastin, S. A. (2022). Human fall detection using YOLO: A real-time and AI-on-the-edge perspective. In 2022 12th International Conference on Pattern Recognition Systems (ICPRS) (pp. 1-6). https://doi.org/10.1109/ICPRS54038.2022.9854070
Zhao, D., Song, T., Gao, J., Li, D., & Niu, Y. (2021). YOLO-Fall: A novel convolutional neural network model for fall detection in open spaces. In 2021 IEEE International Conference on Emergency Science and Information Technology (ICESIT) (Vol. 12, pp.26137-26149).
https://doi.org/10.1109/ACCESS.2024.3362958
Yin, Y., Lei, L., Liang, M., Li, X., He, Y., & Qin, L. (2021). Research on fall detection algorithm for the elderly living alone based on YOLO. In 2021 IEEE International Conference on Emergency Science and Information Technology (ICESIT) (pp. 403-408). ttps://doi.org/10.1109/ICESIT53460.2021.9696459
Wang, X., & Jia, K. (2020). Human fall detection algorithm based on YOLOv3. In 2020 IEEE 5th International Conference on Image, Vision and Computing (ICIVC) (pp. 50-54). IEEE. https://doi.org/10.1109/ICIVC50857.2020.9177447