Detection of personal protective equipment

In this study, a special detection algorithm was developed to determine whether personal protective equipment, which is required to be worn by occupational safety personel on construction sites, is used by employees. In this study, we focused on the moving areas for personal protective equipment used by people. The originality of this work is the creation of a new data set that is not included in the literature, designing a classification of the data obtained from the moving areas within the scope of the study, and the development of the Gaussian Auto Labeling algorithm. The helmet points marked on the images by the operator are used for training of interest areas automatically extracted from the moving areas. As a result of the study, it was possible to classify the helmets which are worn by the construction workers.