DOI QR코드

DOI QR Code

Implementation of an alarm system with AI image processing to detect whether a helmet is worn or not and a fall accident

헬멧 착용 여부 및 쓰러짐 사고 감지를 위한 AI 영상처리와 알람 시스템의 구현

  • Yong-Hwa Jo (Department of Information & Communication AI Engineering, Kyungnam University) ;
  • Hyuek-Jae Lee (Department of Information & Communication AI Engineering, Kyungnam University)
  • 조용화 (경남대학교 정보통신AI공학과) ;
  • 이혁재 (경남대학교 정보통신AI공학과)
  • Received : 2022.08.29
  • Accepted : 2022.09.23
  • Published : 2022.09.30

Abstract

This paper presents an implementation of detecting whether a helmet is worn and there is a fall accident through individual image analysis in real-time from extracting the image objects of several workers active in the industrial field. In order to detect image objects of workers, YOLO, a deep learning-based computer vision model, was used, and for whether a helmet is worn or not, the extracted images with 5,000 different helmet learning data images were applied. For whether a fall accident occurred, the position of the head was checked using the Pose real-time body tracking algorithm of Mediapipe, and the movement speed was calculated to determine whether the person fell. In addition, to give reliability to the result of a falling accident, a method to infer the posture of an object by obtaining the size of YOLO's bounding box was proposed and implemented. Finally, Telegram API Bot and Firebase DB server were implemented for notification service to administrators.

본 논문은 실시간 영상 분석을 통해서 산업현장에서 활동하는 여러 근로자의 영상 객체를 추출해 내고, 추출된 이미지로 부터 개별 영상 분석을 통해 헬멧의 착용 여부와 낙상 사고 여부를 확인하는 방법을 구현한다. 근로자의 영상 객체를 탐지하기 위해서 딥러닝 기반 컴퓨터 비전 모델인 YOLO를 사용하였으며, 추출된 이미지를 이용하여 헬멧의 착용여부를 판단하기 위해 따로 5,000장의 다양한 헬멧 학습 데이터 이미지를 만들어서 사용하였다. 또한, 낙상사고 여부를 판단하기 위해서 Mediapipe의 Pose 실시간 신체추적 알고리즘을 사용하여 머리의 위치를 확인하고 움직이는 속도를 계산하여 쓰러짐 여부를 판단하였다. 결과에 신뢰성을 주기위한 방법으로 YOLO의 바운딩 박스의 크기를 구하여 객체의 자세를 유추하는 방법을 추가하고 구현하였다. 최종적으로 관리자에게 알림 서비스를 위하여 텔레그램 API Bot과 Firebase DB 서버를 구현하였다.

Keywords

References

  1. Analysis of Industrial Accidents in 2020, Available: https://url.kr/7xfh6u
  2. Jeung, Sueng Hyo, Lee, Yong-Soo, Kim, ChangEun, "Improved System for Establishing a Culture to Wear Personal Protective Gear," Journal of the Korea Institute of Construction Safety, 2(1), pp. 16-20, 2019. https://doi.org/10.20931/JKICS.2019.2.1.016
  3. S. Ren, K.He, R. Girshick, and J. Sun (2016). "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks" https://arxiv.org/pdf/1506.01497.pdf (Faster-CNN)
  4. J. Redmon, S. Divvala, R. Girshick, and Ali Farhadi (2016). "You Only Look Once: Unified, Real-Time Object Detection"https://arxiv.org/pdf/1506.02640.pdf (YOLO)
  5. A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao (2020). "YOLOv4: Optimal Speed and Accuracy of Object Detection" https://arxiv.org/pdf/2004.10934.pdf (YOLOv4)
  6. https://zenodo.org/record/6222936 (YOLOv5)
  7. YOLOv5 in PyTorch > ONNX > CoreML > TFLite, Available: https://url.kr/gpsmfj
  8. Mediapipe - Google, Available: https://url.kr/hay576
  9. S.Garg, A.Saxena, and R. Gupta, "Yoga pose classification: a CNN and MediaPipe inspired deep learning approach for real-world application," Journal of Ambient Intelligence and Humanized Computing, 03 June 2022.
  10. Receiving push notifications with Telegram Bot API, Available: https://shared.co.kr/196
  11. FireBase Realtime Database, Available: https://firebase.google.com/docs/database?hl=ko
  12. YOLO_Label github, Available: https://url.kr/ulnvh4
  13. Yoon-Kyu Kang, Hee-Yong Kang, and Dalsoo Weon, "Human Skeleton Keypoints based Fall Detection using GRU," Journal of the Korea AcademiaIndustrial cooperation Society Vol. 22, No. 2 pp. 127-133, 2021.