Journal of the Institute of Convergence Signal Processing (융합신호처리학회논문지)

The Korea Institute of Convergence Signal Processing (한국융합신호처리학회)
권호 표지

The Modified Fall Detection Algorithm based on YOLO-KCF for Elderly Living Alone Care

독거노인 케어를 위한 개선된 YOLO-KCF 기반 낙상감지 알고리즘

  • Kang, Kyoung-Won (Dept. of Information, Communication & Software Engineering, Tongmyong University) ;
  • Park, Soo-Young (Dept. of Youth Science, Kyonggi University)
  • 강경원 (동명대학교 정보통신소프트웨어공학과) ;
  • 박수영 (경기대학교 대학원 청소년학과)
  • Received : 2020.03.23
  • Accepted : 2020.06.09
  • Published : 2020.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

As the number of elderly people living alone increases, the frequency of fall accidents is also increasing. Falls are a threat to the health of older adults and can reduce their ability to remain independent. To solve this problem, we need real-time technology to recognize and respond to the critical condition of the elderly living alone. Therefore, this paper proposes a modified fall detection algorithm based on YOLO-KCF that can check one of the emergency situations in real time for the elderly living alone. YOLO can detect not only the detection of objects, but also the behavior of objects, namely stand and fall. Therefore, this paper can detect fall using the ratio of change of boundary box between stand and falling situation, and this algorithm can improve the shortcomings of KCF.

독거노인이 증가하면서 낙상 사고 빈도도 높아지고 있다. 낙상은 노인들의 건강을 위협할 뿐만 아니라, 독립적인 생활을 위협할 수 있다. 이 문제를 해결하기 위해서는 독거노인의 위급한 상태를 인식하고 대응할 수 있는 실시간 기술이 필요하다. 따라서 본 논문은 독거노인을 위해 긴급 상황 중 하나인 낙상을 실시간으로 확인할 수 있는 YOLO-KCF를 기반 개선된 낙상 감지 알고리즘을 제안한다. YOLO는 물체의 검출뿐 아니라 서 있는 행동과 쓰러지는 행동 유행을 감지할 수 있다. 따라서 본 논문은 서 있는 행동 유형과 쓰러지는 행동 유형간의 경계 박스의 형태 변화를 이용하여 낙하를 검출할 수 있으며, KCF의 단점을 개선할 수 있다.

Keywords

References

  1. S. Ren, K. He, R. Girshick, J. Sun, "Faster R-CNN: Towards real-time object detection with region proposal networks," Proc. of the Int'l Conf. on Neural Information Processing Systems, pp. 91-99, 2015.
  2. J. Dai, Y. Li, K. He, J. Sun, "R-FCN: Object detection via region-based fully convolutional networks," Proc. of the Int'l Conf. on Neural Information Processing Systems, pp. 379-387, 2016.
  3. T. Y. Lin, P. Dollar, R. Girshick, K. He, B. Hariharan, S. Belongie, "Feature pyramid networks for object detection," Proc. of IEEE Conf. on Computer Vision and Pattern Recognition, pp. 2117-2125, 2017.
  4. J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, "You Only Look Once: Unified, Real-Time Object Detection," Proc. of IEEE Conf. on Computer Vision and Pattern Recognition, pp. 779-788, 2016.
  5. W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C. Y. Fu, A. C. Berg, "SSD: Signle shot multibox detector," European Conf. on Computer Vision, pp. 21-37, 2016.
  6. T. Y. Lin, P. Goyal, R. Girshick, K. He, P. Dollar, "Focal loss for dense object detection," Int'l Conf. on Computer Vision, pp. 2980-2988, 2017.
  7. J. Redmon, A. Farhadi, "YOLO9000: Better, faster, stronger," Proc. of IEEE Conf. on Computer Vision and Pattern Recognition, pp. 6517-6525, 2017.
  8. Redom, Darknet: Open source neural networks in C, http://pjreddie.com/darknet/
  9. J. H. Henriques, R. Caseiro, P. Martins, J. Batista, "High-speed tracking with kernelized correlation filters," TPAMI, vol. 37, no. 3, pp. 583-596, 2015. https://doi.org/10.1109/TPAMI.2014.2345390
  10. S. H. Hwang, S. B. Pan, "Fall detection system using the open source hardware and RGB camera," Journal of KIIT, vol. 16, no. 4, pp. 19-24, 2016.
  11. Fall detection Dataset, http://le2i.cnrs.fr/Fall-detectionDataset.