Issues‎ > ‎vol23n1‎ > ‎

jzs-10847

Detection and classification of falling in elderly people using customized deep learning algorithm

Bnar Abdulsalam Abdulrahman1* and Aree Ali Mohammed2

1Student affairs office, Komar University of Science and Technology, Kurdistan Region, Iraq

2Computer Science Department, College of Science, University of Sulaimani, Kurdistan Region, Iraq

*Corresponding author’s e-mail: bnar.abdoulsalam@komar.edu.iq

22 November 2020   Accepted: 29 November 2020 Published online 20 June 2021 Original: 15 October  Revise   

DOI link:  https://doi.org/10.17656/jzs.10847


Abstract
This research work proposes a fall detection system in elderly people based on customized human body detection using You Only Look Once (YOLO) version-3 algorithm. This scheme provides a high accuracy rate of classification for different cases (stand, sit, and fall). To design such system, IoT based-fall detection is implemented. The Raspberry pi is used to process the tested images captured from the live camera. Then, the YOLO detects the human body and classifies them into three categories. The proposed system has trained for 100 images for each class type based on positive samples. Moreover, in the test phase, static image, and live camera have been used to show the performance of the system in term of the accuracy detection. Finally, the accuracy rate is determined for different distances from the camera in order to improve the validity of the classification. Test results indicate that the proposed system is invariant for the light and environmental conditions and has a good accuracy when the live camera is used. The accuracy rates average of a live camera for different distances are 100%, 95%, and 90% for (stand, sit, and fall) classes respectively.

Key Words: Biometric, Fall detection, Fall classification Customized deep learning , Accuracy detection.


References

[1] M. Pareraki et al. "Biosensors and Internet of Things in smart healthcare applications: challenges and opportunities". Biosensors and Internet of Things in smart healthcare applications: challenges and opportunities. Vol. 7, pp. 25-53. (2020).

[2] P. Gope and T. Hwang. "BSN-Care: A Secure IoT- Based Modern Healthcare System Using Body Sensor Network". IEEE Sensors Journal. Vol. 16, no. 5, pp. 1368–1376. (2016).

[3]R. K. Kodali, G. Swamy, and B. Lakshmi. "An implementation of IoT for healthcare". IEEE Recent Adv. Intell. Comput.Syst. RAICS Vol. 2015, pp. 411–416. (2016).

[4] Koppar, Anant R, and Venugopalachar Sridhar. "A workflow solution for electronic health records to improve healthcare delivery efficiency in rural India," In eHealth, Telemedicine, and Social Medicine, 2009. eTELEMED’09. International Conference on, pp. 227–232, IEEE, 2009 

[5] S. Sivagami, D. Revathy, and L. Nithyabharathi. "Smart Health Care System Implemented usingIoT". Int. J. Contemp. Res. Comput. Sci. Technol. Vol. 2, No. 3, pp. 641–646. (2016).

[6] Mubashir M., Shao L., Seed L. "A survey on fall detection: Principles and approaches". Neurocomputing, pp. 144–152. (2013).

[7] F. Hussain, M. Umair, M. Ehatisham, I. Pires, T. Valente, N. Garcia, and N. Pombo. "An Efficient Machine Learning-based Elderly Fall Detection Algorithm". The Ninth International Conference on Sensor Device Technologies and Applications, pp. 88-93. (2018).

[8] T. Shany, S. J. Redmond, M. R. Narayanan, and N. H. Lovell. "Sensors-based wearable systems for monitoring of human movement and falls". IEEE Sensors Journal. Vol. 12, No. 3, pp. 658–670. (2012).

[9] N. Joshi and S. Nalbalwar. "A Fall Detection and Alert System for an Elderly using Computer Vision and Internet of Things". 2ndIEEE International Conference On Recent Trends in Electronics Information & Communication Technology, pp. 1276-1281. (2017).

[10] D. Anastasios and D. Nikolaos. "Adaptive Deep Learning for a Vision-based Fall Detection". Association for Computing Machinery Conference. pp. 558-565, (2018).

[11] B. Stephanie, W. Xiang, and I. Atkinson. "Internet of Things for Smart Healthcare: Technologies, Challenges, and Opportunities". IEEE Access, Vol. 5, pp. 26521-26544. (2017).

[12] M. Padmanabhan, P. Yuan , G. Chada and H. V. Nguyen. "Physician-Friendly Machine Learning: A Case Study with Cardiovascular Disease Risk Prediction". Journal of Clinical Medicine, Vol. 8, No. 10, pp. 1-13. (2019).

[13] A. Yazar, A. Cetin, and F. Erden. "Multi-sensor ambient assisted living system for fall detection". IEEE International Conference on Acoustics, Speech, and Signal Processing, (2014).

[14] F. Wu, H. Zhao, Y. Zhao, and H. Zhong. "Development of a Wearable-Sensor-Based Fall Detection System"International Journal of Telemedicine and Applications. Vol. 2015, pp. 1-11. (2015).

[15] K. Miguel, A. Brunete, M. Hernando, and E. Gambao. "Home Camera-Based Fall Detection System for the Elderly", Journal of Sensors, Vol. 17, No. 28, 2017.

[16] K. Lu and E. Chu. "An Image-Based Fall Detection System for the Elderly". Journal of Applied Science,. Vol. 8. (2018).

[17]M. Yu, L. Gong, and S. Kollias. "Computer Vision Based Fall Detection by a Convolutional Neural Network". Proceedings of ACM International Conference on Multimodal Interaction. pp. 416-420. (2017).

[18]Z. Zhang, C. Conly, and V. Athitsos. "A Survey on Vision-based Fall Detection". Proceedings of the ACM International Conference on Pervasive Technologies related to Assistive Environments. Article No. 46. (2015).

[19] O. Seredin A. Kopylov, S. Huang , and D. Rodionov. "A Skeleton Features-Based Fall Detection using Microsoft Kinect V2 with One Class-Classifier Outlier Removal". Int. Workshop on Photogrammetric & Computer Vision Techniques for Video Surveillance, Biometrics and Biomedicine. Vol. XLII-2/W12, pp. 189-195. (2019).

[20] X. Kong, L. Chen, Z. Wang, Y. Chen, L. Meng, and H.Tomiyama. "Robust Self-Adaptation Fall-Detection System Based on Camera Height". Journal of Sensors. Vol. 19, No. 3768. (2019).

[21] Qi-Chao Mao , Hong-Mei Sun, Yan-Bo Liu and Rui-Sheng Jia. "Mini-YOLOv3: Real-Time Object Detector for Embedded Applications". 2941547, IEEE Access, DOI 10.1109/ACCESS.2019