AUTOMATED DETECTION OF PERSONAL PROTECTIVE EQUIPMENT (PPE) USING YOLOV8 FOR CONSTRUCTION SAFETY

• Main Author: Raditya, Yosafat
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/85005
• Institution: Institut Teknologi Bandung

Workplace safety in construction environments is a top priority to reduce the risk of accidents and ensure compliance with safety standards. Proper use of Personal Protective Equipment (PPE) is crucial in achieving this goal. CCTV is often employed to monitor PPE usage; however, human limitations in continuously observing CCTV footage can lead to gaps in supervision. Therefore, an automated system is needed to monitor CCTV recordings and detect PPE usage. This final project aims to design a system, develop and implement a model, and evaluate the system based on the conditions in the construction environment. The PPE detection system will utilize the YOLOv8 model to recognize and detect PPE. The system's development follows the CRISP-DM methodology, which includes the stages of Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, and Deployment. The outcome of this project is a reliable PPE detection system that can be operated in construction environments, thereby supporting the implementation of workplace safety. After evaluating the system in a construction environment, the developed PPE detection system achieved a precision of 0.848, a recall of 0.560, and an F1-score of 0.676.