An Approach to Automatic Classification of Construction Workers by Degree of Risk

Most of construction accidents are caused by the unsafe behavior of construction workers. To reduce the risk of accidents, safety manages should be assigned to manage safety. However, it is difficult for safety manages to manage risks of all worksites. To address this issue, it is necessary to continuously monitor worksites in a timely manner. However, since it is difficult that a safety manager identify hazard areas in many worksites where various tasks are carried out simultaneously, it is necessary to be able to monitor safety of high-risk worksites and identify worksites with high risk of accidents. Therefore, we propose a method to detect and classify workers by degree of risk for the worksites where hazard area has been identified.