발간년도 : [2022]
| 논문정보 |
|
| 논문명(한글) |
[Vol.17, No.3] Deep Learning-based Worker Personal Protective Equipment and Face Identification System |
|
| 논문투고자 |
Keunwoo Lee, Seokwoo Lee, Hakseong Kim, Hoekyung Jung |
|
| 논문내용 |
One of the most common safety accidents in industrial sites is accidents caused by not wearing personal protective equipment such as helmets, safety rings, and safety boots. According to domestic industrial accident analysis, 37.3% of all deaths occurred without a helmet on. Accident prevention systems in industrial sites require automated technology application. Recently, many efforts have been made to introduce an automated safety management method using image recognition technology based on computer vision and deep learning. Therefore, in this paper, a study was conducted on worker wearing personal protective equipment and face identification for safety management in industrial sites. An image data set of industrial site workers and personal protective equipment wearing was prepared, and a YOLOv5s model for personal protective equipment detection and a FaceNet model for face recognition were applied. Through the experiment, performance such as accuracy, precision, reproducibility, harmonic mean, and face pair comparison was confirmed for the detection of worker's personal protective equipment and face recognition. This paper is differentiated in that it is possible to simultaneously check the identity of the worker and compliance with the wearing of personal protective equipment through face recognition and personal protective equipment detection by the worker image, and deep learning-based worker wearing and identity of personal protective equipment was confirmed. |
|
| 첨부논문 |
|
|
|
|
|
자료실 > 논문지 
