DOI QR코드

DOI QR Code

Configuration and Application of a deep learning-based fall detection system

딥러닝 기반 낙상 감지 시스템의 구성과 적용

  • Jong-Seok Woo (Dept. of Artificial Intelligence Convergence, Pukyong National University) ;
  • Lionel Kyenyeneye (Dept. of Artificial Intelligence Convergence, Pukyong National University) ;
  • Sang-Joong Jung (Dept. of Applied Artificial Intelligence, Dongseo University) ;
  • Wan-Young Chung (Dept. of Artificial Intelligence Convergence, Pukyong National University)
  • 우종석 (부경대학교 인공지능융합학과) ;
  • 리오넬 (부경대학교 인공지능융합학과) ;
  • 정상중 (동서대학교 인공지능응용학과) ;
  • 정완영 (부경대학교 인공지능융합학과)
  • Received : 2023.11.08
  • Accepted : 2023.12.28
  • Published : 2023.12.31

Abstract

Falling occurs unexpectedly during daily activities, causing many difficulties in life. The purpose of this study was to establish a system for fall detection of high-risk occupations and to verify their effectiveness by collecting data and applying it to predictive models. To this end, a wearable device was configured to detect fall by calculating acceleration signals and azimuths through acceleration sensors and gyro sensors. In addition, the study participants wore the device on their abdomen and measured necessary data from falls-related movements in the process of performing predetermined activities and transmitted it to the computer through a Bluetooth device present in the device. The collected data was processed through filtering, applied to fall detection prediction models based on deep learning algorithms which are 1D CNN, LSTM and CNN-LSTM, and evaluate the results.

낙상은 일상의 활동 중에 예기치 않게 발생하여 생활에 많은 어려움을 초래한다. 본 연구는 고위험 직종 종사자들의 낙상 감지를 위한 시스템을 구성하고 자료를 수집하여 예측 모델에 적용함으로써 그 유효성을 검증하는 것을 목적으로 하였다. 이를 위해 가속도센서와 자이로센서를 통해 가속도 신호와 방위각을 산출하여 낙상 여부를 감지하는 웨어러블 기기를 구성하였다. 그리고 연구 참여자들이 이 기기를 복부에 착용하고 정해진 활동을 수행하는 과정에서 낙상과 관련한 동작으로부터 필요한 데이터를 측정하고 기기 내에 존재하는 블루투스 장치를 통해 컴퓨터로 전송하였다. 이렇게 수집된 데이터를 필터링 등을 통해 처리하여 딥러닝 알고리즘들인 1D CNN, LSTM, CNN-LSTM에 근거한 낙상 감지 예측 모델들에 적용하고 그 결과를 평가하였다.

Keywords

References

  1. 질병관리청 국가건강정보포털  https://health.kdca.go.kr/healthinfo/biz/health/ gnrlzHealthInfo/gnrlzHealthInfo/gnrlzHealthInfoView.do
  2. 고용노동부 정책자료실, '2022년 산업재해현황분석', https://www.moel.go.kr/info/publict/publictDataView.do?bbs_seq=20231202281
  3. 소방청, '2023 소방청 통계연보'  https://www.nfa.go.kr/nfa/releaseinformation/statistical-information/main/?boardId=bbs_0000000000000019&mode=view&cntId=55
  4. Jae-Jung Kim, Min-Ho Park, Sin-Ho Kim, Ji-Yun Seo, Do-Un Jeong, "Implementation of the Fall Detection System based on IMU Sensor Applicable to Industrial Sites," 한국정보통신학회 여성 ICT 학술대회 논문집, Vol. 2021 No. 8, pp.33-36, 2021.
  5. Pil-Seong Jeong, Yang-Hyun Cho, "Fall Detection System based Internet of Things," Journal of the Korea Institute Of Information and Communication Engineering, Vol. 19 No. 11, pp.2546-2553, 2015. https://doi.org/10.6109/jkiice.2015.19.11.2546
  6. Lin, C. W., Z. H. Ling, Y. C. Chang, C. J. Kuo, "Compressed -domain Fall Incident Detection for Intelligent Homecare", Tje Journal of VLSI Signal Processing, Vol. 49, No. 3, pp. 393-408, 2007 https://doi.org/10.1007/s11265-007-0092-3
  7. In-Kyung Kim, Dae-Hee Kim, Song Noh, Jae-Koo Lee, "1D CNN and Machine Learning Methods for Fall Detection," KIPS Transactions on Software and Data Engineering, Vol. 10 No. 3, pp.85-90, 2021.
  8. Q. Li, J. A. Stankovic, M. Hanson, A. Barth, "Accurate, Fast Fall Detection Using Gyroscopes and Accelerometer-Derived Posture Information." Sixth International Workshop Wearable and Implantable Body Sensor Networks(BSN) 2009, Berkeley, CA, 3-5 June 2009,
  9. Bourke, A. K., G. M. Lyons, "A threshold-based fall-detection algorithm using a bi-axial gyroscope sensor", Medical Engineering and Physics, Vol. 30, pp.84-90, 2008. https://doi.org/10.1016/j.medengphy.2006.12.001
  10. Jeong-Hyun Roh, Jin-Heon Kim, "Fall detection based on acceleration sensor attached to wrist using feature data in frequency space", Smart media journal Vol. 10 No. 3, 31-38, 2021. https://doi.org/10.30693/SMJ.2021.10.3.31
  11. Luna-Perejon, F. Dominguez-Morales, M. J. Civit-Balcells, A. "Wearable Fall Detector using Recurrent Neural Networks.", Sensors 19, 4885, 2019.
  12. Khojasteh, Samad Barri, Villar, Jose R. Chira, Camelia, Gonzalez, Victor M. Enrique de la Cal, "Improving fall detection using an on-wrist wearable accelerometer.", Sensors 18, 1350, 2018.
  13. Hyun Park, Jun-Mo Park, Yeon-Chul Ha, "Implementation of Acceleration Sensor-based Human activity and Fall Classification Algorithm," The Journal of Korea Institute of Convergence Signal Processing, Vol. 23 No. 2, pp.76-83, 2022.
  14. Do-Un Jeong, Wan-Young Chung, "Posture and activity monitoring using a 3-axis accelerometer," Journal of Sensor Science and Technology, Vol. 16 No. 6, pp.467-474, 2007. https://doi.org/10.5369/JSST.2007.16.6.467
  15. R. M. Gibson, A. Amira, N. Ramzan, P. Casaseca-de -la-Higuera, and Z. Pervez, "Multiplecomparator classifier framework for accelerometer-based fall detection and diagnostic," Applied Soft Computing, Vol. 39, pp.94-103, 2016. https://doi.org/10.1016/j.asoc.2015.10.062