The Journal of the Convergence on Culture Technology (문화기술의 융합)

The International Promotion Agency of Culture Technology (국제문화기술진흥원)
권호 표지
DOI QR코드

DOI QR Code

A Study on the Improvement of Construction Site Worker Detection Performance Using YOLOv5 and OpenPose

YOLOv5 및 OpenPose를 이용한 건설현장 근로자 탐지성능 향상에 대한 연구

  • Received : 2022.08.30
  • Accepted : 2022.09.09
  • Published : 2022.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

The construction is the industry with the highest fatalities, and the fatalities has not decreased despite various institutional improvements. Accordingly, real-time safety management by applying artificial intelligence (AI) to CCTV images is emerging. Although some research on worker detection by applying AI to images of construction sites is being conducted, there are limitations in performance expression due to problems such as complex background due to the nature of the construction industry. In this study, the YOLO model and the OpenPose model were fused to improve the performance of worker detection and posture estimation to improve the detection performance of workers in various complex conditions. This is expected to be highly useful in terms of unsafe behavior and health management of workers in the future.

건설업은 사망자 수가 가장 많이 발생하는 산업이며, 다양한 제도 개선에도 사망자는 크게 줄어들지 않고 있다. 이에 따라, CCTV 영상에 인공지능(AI)을 적용한 실시간 안전관리가 부각되고 있다. 건설현장의 영상에 대한 AI를 적용한 근로자 탐지연구가 진행되고 있지만, 건설업의 특성상 복잡한 배경 등의 문제로 인해 성능 발현에 제한이 있다. 본 연구에서는 근로자의 탐지 및 자세 추정에 대한 성능 향상을 위해 YOLO 모델과 OpenPose 모델을 융합하여, 복잡 다양한 조건에서의 근로자에 대한 탐지 성능을 향상시켰다. 이는 향후 근로자의 불안전안 행동 및 건강관리 측면에서 활용도가 높을 것으로 예상된다.

Keywords

Acknowledgement

이 논문은 2021년 및 2022년도 정부(교육부)의 재원으로 한국연구재단의 지원을 받아 수행된 기초연구사업임(No. 2021R1A6A3A01086763, No. 2022R1I1A1A01061658).

References

  1. Ministry of Employment and Labor, "'2017-2021 Industrial Accident and Death Statistics", 2022.
  2. Ministry of Land, Infrastructure and Transport, "Smart construction technology development project", 2019.
  3. Y. S. Kim, T. K. Oh, C. S. Kim, N. E. Lee, C. S. Hong, S. Y. Lee, Y. G. Yoon, "A Study on the Actual Condition Analysis and Activation Plan of Smart Construction Safety Technology by the Survey", Journal of the Korean Society of Safety, Vol. 37, No. 1, pp. 30-40, 2022. https://doi.org/10.14346/JKOSOS.2022.37.1.30
  4. J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, "You only look once: Unified, real-time object detection," In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.779-788, 2016.
  5. Ultralytics. YOLOv5 [Internet] https://github.com/ultralytics/yolov5
  6. Cao, Zhe, Gines Hidalgo, Tomas Simon, Shih-En Wei, and Yaser Sheikh. "OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields." ArXiv:1812.08008 [Cs], May 30, 2019. https://arxiv.org/abs/1812.08008.
  7. Osokin, Daniil. "Real-Time 2D Multi-Person Pose Estimation on CPU: Lightweight OpenPose." ArXiv:1811.12004 [Cs], November 29, 2018. https://arxiv.org/abs/1811.12004.