DOI QR코드

DOI QR Code

Statistical Analysis of Major Accident Reports and Development of a Real-time Detection Model for Portable Ladder and Safety Helmet

이동식사다리 중대재해 통계 분석 및 이동식사다리와 안전모 실시간 탐지 기계학습 모델 개발

  • Choi, Seung-Ju (Department of Safety Engineering, University of Ulsan) ;
  • Jung, Kihyo (School of Industrial Engineering, University of Ulsan)
  • 최승주 (울산대학교 안전보건전문학과) ;
  • 정기효 (울산대학교 산업경영공학부)
  • Received : 2020.12.03
  • Accepted : 2021.03.04
  • Published : 2021.03.31

Abstract

The leading source of occupational fatalities is a portable ladder in Korea because it is widely used in industry as work platform. In order to reduce victims, it is necessary to establish preventive measures for the accidents caused by portable ladder. Therefore, this study statistically analyzed injury death by portable ladder for recent 10 years to investigate the accident characteristics. Next, to monitor wearing of safety helmet in real-time while working on a portable ladder, this study developed an object detection model based on the You Only Look Once(YOLO) architecture, which can accurately detect objects within a reasonable time. The model was trained on 6,023 images with/without ladders and safety helmets. The performance of the proposed detection model was 0.795 for F1 score and 0.843 for mean average precision. In addition, the proposed model processed at least 25 frames per second which make the model suitable for real-time application.

References

  1. A. Kuznetsova, H. Rom, N. Alldrin, et al.(2020), The open images dataset V4: Unified image classification, object detection, and visual relationship detection at scale. IJCV.
  2. C. Song, Y. Kwon, D. Kim, K. Kang(2013), "A study on the improvement measures for the prevention of fall accidents on the ladder at construction sites." 2013 Spring Conference of Korea Saf. Manag. Sci., 225-234.
  3. D. M. W. Powers(2011), "Evaluation: From precision, recall and F-Measure to ROC, informedness, markedness & correlation." J. Machine Learning Technologies, 2(1):37-63.
  4. G. Jocher, Y. Kwon, Y., Guigarfr, J. Veitch-Michaelis, et al.(2020), "Ultralytics/volov3: 43.1mAP@0.5:0. 95 on COCO2014(version v7)." Zenodo. http://doi.org/10.5281/zenodo.3785397 https://doi.org/10.5281/zenodo.3785397
  5. H. Kim, S. Lee, W. Jung, B. Ryu(2009), "A study on the preventive measures against fall injuries in manufacturing industry focusing on the portable ladders." J. Korean Soc. Saf., 24(6):136-143.
  6. H. Lee(2018), "An efficient deep learning platform for object detection." Master's thesis, Soongsil University.
  7. H. Sim, K. Kang(2017), "A study on the death accident analysis of ladder and prevention measures for fall accidents." J. Korea Saf. Manag. Sci., 19(4):95-104. https://doi.org/10.12812/KSMS.2017.19.4.95
  8. I. Krasin, T. Duerig, N. Alldrin, et al.(2017), Open images: A public dataset for large-scale multi-label and multi-class image classification, 2017. https://stor-age.googleapis.com/openimages/web/index.html
  9. International Labour Organization(ILO)(2020), Statistics on safety and health at work. https://ilostat.ilo.org/topics/safety-and-health-at-work/
  10. J. Redmon, A. Farhadi(2017a), "YOLO9000: Better, faster, stronger." The IEEE Conference on Computer Vision and Pattern Recognition(CVPR), IEEE.
  11. J. Redmon, A. Farhadi(2017b), YOLOv3: An incremental improvement. arXiv preprint arXiv:1804.02767.
  12. J. Redmon, S. Divvala, R. Girshick, A. Farhadi (2016), "You only look once: Unified, real-time object detection." The IEEE Conference on Computer Vision and Pattern Recognition(CVPR), IEEE.
  13. M. Everingham, L. V. Gool, C. K. I. Williams, et al.(2010), "The pascal Visual Object Classes(VOC) challenge." Int. J. Comput. Vis., 88:303-338. https://doi.org/10.1007/s11263-009-0275-4
  14. Ministry of Employment and Labor(MOEL)(2018), Analysis of industrial accidents in 2017.
  15. Ministry of Employment and Labor(MOEL)(2020), Status of industrial accidents in 2019. http://www.kosha.or.kr/kosha/data/industrialAccidentStatus.do
  16. N. D. Nath, A. H. Behzadan, S. G. Paal(2020), "Deep learning for site safety: Real-time detection of personal protective equipment." Automation on Construction, 112:103085. https://doi.org/10.1016/j.autcon.2020.103085
  17. Occupational Safety and Health Research Institute (OSHRI)(2015), Cause of industrial accidents in 2014. OSHRI Research Report.
  18. S. Park, S. Yoon, J. Heo(2019), "Image-based automatic detection of construction helmets using R-FCN and transfer learning." J. Korea Soc. Civil Eng., 39(3):399-407. https://doi.org/10.12652/KSCE.2019.39.3.0399
  19. The Korea Occupational Safety and Health Agency (KOSHA)(2010-2019), Industrial accident statistics.
  20. W. Jang, D. Shin(2009), "WSN safety monitoring using RSSI-based ranging technique in a construction site." J. of Korean Soc. of Societal Security, 2(2):49-54.
  21. Xie, Liangbin(2019), "Hardhat." Havard Dataverse, V1. http://doi.org/10.7910/DVN/7-CBGOS