RANCANG BANGUN SISTEM PENGECEKAN SAFETY HELMET DAN MASKER PADA PROYEK PEMBANGUNAN DENGAN METODELOGI YOLOv5 DENGAN MIKROKONTROLER SEBAGAI INDIKATOR

Abstract

The use of Personal Protective Equipment (PPE) is a very important role in occupational health and safety, especially safety helmets that can minimize the impact of falling small and large objects, besides safety helmets, masks are no less important to prevent dust and small particles from entering the respiratory tract. To ensure that field workers wear PPE, especially safety helmets and masks, a system is needed that can automatically check workers before entering the work area. This research aims to find out how to create an AI model to detect safety helmets and masks using the You Only Look Once version 5 (YOLOv5) method and find out how much accuracy, besides that the purpose of this research is to find out how to make an indicator system using a microcontroller with result from YOLOv5. After experimenting with the tools made, YOLOv5 successfully detected safety helmets and masks with an mAP50 value of 0.995 and F1-score of 0.997 and ESP32 successfully provided indicators with a success value of 100% from 20 trials.