Research Concerning Accuracy Verification of Pattern Identification Assumed Helmet Wearing Using Deep Learning

Abstract

In order to eliminate labor accidents at the construction sites, it is important to implement effective safety management measures. For example, there is a measure that identify and warn a person who is at risk of intrusion into a dangerous place or contact with construction machines using cameras. In existing research, a technology has been developed for identifying people using deep learning for surveillance cameras in towns. However, at construction sites it is necessary to identify workers with similar clothes and construction machines. Thus, there is a possibility that the existing technology could not be applied. Therefore, in this research, we tried to identify the person wearing the helmet with patterns using deep learning for purpose of safety management in the construction fields.