Non-hard-hat use (NHU) is related to many construction accidents, so NHU inspection is crucial to safety management, in which automatic NHU monitoring plays an essential role. Existing computer vision–based NHU inspection methods lack capabilities in identifying workers and helping take real-time action. Previous sensor-based NHU inspection methods require direct skin contact, which would be uncomfortable for workers. In addition, previous sensor-based methods could be deceived by objects other than human heads and could not achieve real-time alarms. This study aims to address these problems by implementing real-time alarming, monitoring, and locating for NHU in construction based on sensor, mobile, web, and cloud techniques. A smart hard-hat system is developed using an Internet of Things (IoT)-based architecture including (1) a hard hat with an infrared beam detector and thermal infrared sensor for nonintrusive NHU detection; (2) radio-frequency identification (RFID) triggers for locating NHU with an average detection error of less than 10 cm; (3) a smartphone application for personalized warnings; (4) a web application for data visualization and alarms for managers; and (5) a cloud sever for data storage and retrieval. The proposed system enables both workers and managers to take timely actions against NHU. The system performance is evaluated in a laboratory test and validated in a field application. It is indicated that the proposed system is accurate and reliable, showing potential to promote safety inspection and supervision in construction.
1 October 2017 in Advanced Engineering Informatics
Struck‐by accidents often cause serious injuries in construction. Monitoring of the struck‐by hazards in terms of spatial relationship between a worker and a heavy vehicle is crucial to prevent such accidents. The computer vision‐based technique has been put forward for monitoring the struck‐by hazards but there exists shortages such as spatial relationship distortion due to two‐dimensional (2D) image pixels‐based estimation and self‐occlusion of heavy vehicles. This study is aimed to address these problems, including the detection of workers and heavy vehicles, three‐dimensional (3D) bounding box reconstruction for the detected objects, depth and range estimation in the monocular 2D vision, and 3D spatial relationship recognition. A series of experiments were conducted to evaluate the proposed method. The proposed method is expected to estimate 3D spatial relationship between construction worker and heavy vehicle in a real‐time and view‐invariant manner, thus enhancing struck‐by hazards monitoring at the construction site.