Construction sites are inherently dynamic due to continuous resource changes. As construction projects become bigger and more complex, this dynamism can lead to more frequent incidents on jobsites. Safety training, personal protective equipment, and regulations have been applied to improve the safety of construction sites. However, because of the dynamic and congested nature of construction sites, these practices still cannot guarantee safety. This paper proposes a vision-based hazard avoidance system that proactively informs workers of potentially dangerous situations. The system enables workers to recognize and consequently avoid dangers before accidents occur by displaying augmented hazard information on a wearable device. The system comprises of three modules; a vision-based site monitoring module that utilizes image capture device and wearable devices to identify site hazards, a safety assessment module that uses captured image data and fuzzy-based reasoning to evaluate the safety level of each object, and a visualization module that provides actionable information such as hazard orientation, distance, and safety level. The safety information provided by the proposed system can mitigate hazards and improve construction site safety.
Bibliographical noteFunding Information:
This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP; Ministry of Science, ICT and Future Planning) (NRF-2014R1A2A1A11052499 and No. 2011-0030040). The authors would like to thank Doosan Engineering & Construction Co. Ltd. for granting access to their construction site. The authors also would like to thank Inhae Ha, Seongdeok Bang, and Somin Park for helping acquisition of data used in this study.
All Science Journal Classification (ASJC) codes
- Control and Systems Engineering
- Civil and Structural Engineering
- Building and Construction