Journal of the Korea Institute of Information and Communication Engineering (한국정보통신학회논문지)

The Korea Institute of Information and Commucation Engineering (한국정보통신학회)
권호 표지
DOI QR코드

DOI QR Code

Study of regularization of long short-term memory(LSTM) for fall detection system of the elderly

장단기 메모리를 이용한 노인 낙상감지시스템의 정규화에 대한 연구

  • Jeong, Seung Su (Department of Electrical, Electronic, and Control Eng., Hankyong National University) ;
  • Kim, Namg Ho (Convergence Technology Campus of Korea Polytechnic) ;
  • Yu, Yun Seop (ICT&Robotics Eng. and IITC, Hankyong National University)
  • Received : 2021.08.20
  • Accepted : 2021.09.12
  • Published : 2021.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we introduce a regularization of long short-term memory (LSTM) based fall detection system using TensorFlow that can detect falls that can occur in the elderly. Fall detection uses data from a 3-axis acceleration sensor attached to the body of an elderly person and learns about a total of 7 behavior patterns, each of which is a pattern that occurs in daily life, and the remaining 3 are patterns for falls. During training, a normalization process is performed to effectively reduce the loss function, and the normalization performs a maximum-minimum normalization for data and a L2 regularization for the loss function. The optimal regularization conditions of LSTM using several falling parameters obtained from the 3-axis accelerometer is explained. When normalization and regularization rate λ for sum vector magnitude (SVM) are 127 and 0.00015, respectively, the best sensitivity, specificity, and accuracy are 98.4, 94.8, and 96.9%, respectively.

본 논문에서는 고령자의 낙상상황을 감지할 수 있는 텐서플로우 장단기 메모리 기반 낙상감지 시스템의 정규화에 대하여 소개한다. 낙상감지는 고령자의 몸에 부착한 3축 가속도 센서 데이터를 사용하며, 총 7가지의 행동 패턴들에 대하여 학습하며, 각각 4가지는 일상생활에서 일어나는 패턴이고, 나머지 3가지는 낙상에 대한 패턴이다. 학습시에는 손실함수(loss function)를 효과적으로 줄이기 위하여 정규화 과정을 진행하며, 정규화 과정은 데이터에 대하여 최대최소 정규화, 손실함수에 대하여 L2 정규화 과정을 진행한다. 3축 가속도 센서를 이용하여 구한 다양한 파라미터에 대하여 정규화 과정의 최적의 조건을 제시한다. 낙상 검출율면에서 SVM을 이용하고 정규화 127과 정규화율 λ 0.00015일 때 Sensitivity 98.4%, Specificity 94.8%, Accuracy 96.9%로 가장 좋은 모습을 보였다.

Keywords

Acknowledgement

This research was supported by the Basic Science Research Program through NRF of Korea funded by the Ministry of Education (NRF-2019R1F1A1060383)

References

  1. Top Ten Reviews for Fall Detection of Seniors. Purch.com [Internet]. Available: http://www.toptenreviews.com/health/senior-care/best-fall-detection-sensors/
  2. S. S. Khan and J. Hoey, "Review of fall detection techniques: A data availability perspective," Medical Engineering & Physics, vol. 39, pp. 12-22, Jan. 2017. DOI: 10.1016/j.medengphy.2016.10.014.
  3. S. Khojasteh, J. R. Villar, C. Chira, V. M. Gonzalez, and E. de la Cal, "Improving fall detection using an on-wrist wearable accelerometer," Sensors, vol. 18, no. 5, pp. 1350, Apr. 2018. DOI: 10.3390/s18051350.
  4. T. Xu, Y. Zhou, and J. Zhu, "New advances and challenges of fall detection systems: A survey" Applied Sciences, vol. 8, no. 3, pp. 418, 2018. DOI: 10.3390/app8030418.
  5. H. Li, A. Shrestha, H. Heidari, J. L. Kernec, and F. Fioranelli, "Bi-LSTM network for multimodal continuous human activity recognition and fall detection," IEEE Sensors Journal, vol. 20, no. 3, pp. 1191-1201, 2019. https://doi.org/10.1109/jsen.2019.2946095
  6. S. Hochreiter and J. Schmidhuber, "Long Short-Term Memory," Neural Computation, vol. 9, no. 8, pp. 1735-1780, Nov. 1997. DOI: 10.1162/neco.1997.9.8.1735.
  7. M. Roondiwala, H. Patel, and S. Varma, "Predicting Stock Prices Using LSTM," International Journal of Science and Research, vol. 6, no. 4, pp. 1754-1756, Apr. 2017.
  8. T. Hayashi, S. Watanabe, T. Toda, T. Hori, J. L. Roux, and K. Takeda, "Duration-Controlled LSTM for Polyphonic Sound Event Detection," IEEE/ACM Trans. Audio Speech Lang. Process, vol. 25, no. 11, pp. 2059-2070, 2017. DOI: 10.1109/taslp.2017.2740002.
  9. D. Freedman, R. Pisani, and R. Purves, Statistics : Fourth International Student Edition, New York, NY: W. W. Norton & Company, 2007.
  10. Regularization for Simplicity: Lambda, Google Developers, 2020 [Internet]. Available: https://developers.google.com/machine-learning/crash-course/regularization-for-simplicity/lambda?hl=ko.
  11. Vanishing gradient problem, wikipedia [Internet]. Available: https://en.wikipedia.org/wiki/Vanishing_gradient_problem.
  12. N. H. Kim and Y. S. Yu, "Fall Recognition Algorithm Using Gravity-Weighted 3-Axis Accelerometer Data," Journal of the Institute of Electronics and Information Engineers, vol. 50, no. 6, pp. 254-259, Jun. 2013. https://doi.org/10.5573/IEEK.2013.50.6.254
  13. M. K. Dahouda and I. Joe, "A Deep-Learned Embedding Technique for Categorical Features Encoding," IEEE Access, vol. 9, pp. 1, 2021. DOI: 10.1109/access.2021.3104357.
  14. K. Li, X. Zhao, J. Bian, and M. Tan, "Sequential learning for multimodal 3D human activity recognition with Long-Short Term Memory," in Proceeding of 2017 IEEE International Conference on Mechatronics and Automation (ICMA), pp. 1556-1561, 2017. DOI: 10.1109/icma.2017.8016048.
  15. LSTMs for Human Activity Recognition, github 2020 [Internet]. Available: https://github.com/guillaume-chevalier/LSTM-Human-Activity-Recognition.
  16. T. V. Laarhoven, "L2 regularization versus batch and weight normalization," arXiv:1706.05350v1, 2017.
  17. D. Lim, C. Park, N. H. Kim, S. H. Kim, and Y. S. Yu, "Fall-Detection Algorithm Using 3-Axis Acceleration: Combination with Simple Threshold and Hidden Markov Model," Journal of Applied Mathematics, vol. 2014, pp. 8, 2014. DOI: 10.1155/2014/896030.