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

Worker Collision Safety Management System using Object Detection

객체 탐지를 활용한 근로자 충돌 안전관리 시스템

  • Lee, Taejun (Department of Computer Engineering, Paichai University) ;
  • Kim, Seongjae (Department of Computer Engineering, Paichai University) ;
  • Hwang, Chul-Hyun (Department of Big Data, Hanyang Women's University) ;
  • Jung, Hoekyung (Department of Computer Engineering, Paichai University)
  • Received : 2022.07.25
  • Accepted : 2022.07.28
  • Published : 2022.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, AI, big data, and IoT technologies are being used in various solutions such as fire detection and gas or dangerous substance detection for safety accident prevention. According to the status of occupational accidents published by the Ministry of Employment and Labor in 2021, the accident rate, the number of injured, and the number of deaths have increased compared to 2020. In this paper, referring to the dataset construction guidelines provided by the National Intelligence Service Agency(NIA), the dataset is directly collected from the field and learned with YOLOv4 to propose a collision risk object detection system through object detection. The accuracy of the dangerous situation rule violation was 88% indoors and 92% outdoors. Through this system, it is thought that it will be possible to analyze safety accidents that occur in industrial sites in advance and use them to intelligent platforms research.

최근 인공지능, 빅데이터, 사물인터넷 기술이 안전사고 예방을 위한 화재 감지, 가스나 유해 물질 감지 등 다양한 솔루션에서 활용되고 있다. 2021년 고용노동부에서 발간한 산업 재해 발생 현황에 따르면, 2020년과 비교해 재해율, 재해자 수, 사망자 수가 증가하였으며 최근에는 중대재해 처벌 등에 관한 법률과 같은 안전조치를 강화하는 등 제도적, 사회적 관심이 높아지고 있다. 본 논문에서는 한국지능정보사회진흥원(NIA)에서 제공한 데이터셋 구축 가이드라인을 참고하여 현장에서 직접 수집해 데이터셋을 직접 구축하고 YOLOv4로 학습하여 객체 탐지를 통해 충돌위험 객체 탐지 시스템을 제안하고자 한다. 위험 상황 규칙 위반에 대한 정확도는 실내 88%, 실외 92%의 탐지 성능을 보였다. 이러한 시스템을 통해 산업 현장에서 발생하는 안전사고를 사전에 분석해 지능형 플랫폼 연구에 활용이 가능할 것으로 사료된다.

Keywords

Acknowledgement

This work was supported by the Korea Medical Device Development Fund grant funded by the Korea government(the Ministry of Science and ICT, the Ministry of Trade, Industry and Energy, the Ministry of Health & Welfare, the Ministry of Food and Drug Safety) (Project Number: 1415181564)

References

  1. S. E. Yang, S. O. Lee, and H. K. Jung, "IoT Platform System for Electric Fire Prediction and Prevention," Journal of the Korea Institute of Information and Communication Engineering, vol. 26, no. 2, pp. 223-229, Feb. 2022. https://doi.org/10.6109/JKIICE.2022.26.2.223
  2. C. H. Hwang, H. S. Kim, and H. K. Jung, "Detection and Correction Method of Erroneous Data Using Quantile Pattern and LSTM," Journal of Information and Communication Convergence Engineering, vol. 16, no. 4, pp. 242-247, Dec. 2018. https://doi.org/10.6109/JICCE.2018.16.4.242
  3. Korea Ministry of Employment and Labor, Policy archives - occupational accident status in 2021 [Internet]. Available: https://www.moel.go.kr/policy/policydata/view.do?bbs_seq=20220300882.
  4. Severe Accident Penalty Act. Pub. L. No. 17907. § 2. 2022. [Online]. Available: https://www.law.go.kr/%EB%B2%95%EB%A0%B9/%EC%A4%91%EB%8C%80%EC%9E%AC%ED%95%B4%EC%B2%98%EB%B2%8C%EB%93%B1%EC%97%90%EA%B4%80%ED%95%9C%EB%B2%95%EB%A5%A0.
  5. Korea National Information Society. AI dataset building guide and quality guide [Internet]. Available: https://aihub.or.kr/intro/guide.
  6. A. Bochkovskiy, C.-Y. Wang and H.-Y. M. Liao, "YOLOv4: optimal speed and accuracy of object detection," arXiv preprint arXiv:2004.10934, 2020.
  7. P. Khandelwal, A. Khandelwal, S. Agarwal, D. Thomas, N. Xavier, and A. Raghurama, "Using Computer Vision to enhance Safety of Workforce in Manufacturing in Post COVID World," arXiv preprint arXiv:2005.05287, 2020.
  8. R. Padilla, S. L. Netto, and E. A. B. da Silva, "A Survey on Performance Metrics for Object-Detection Algorithms," in 2020 International Conference on Systems, Signals and Image Processing (IWSSIP), Niteroi, Brazil, pp. 237-242, 2020.