Exploring Pre-Trained Model Potential for Reflective Vest Real Time Detection With YOLOv8 Models

Workplace safety is crucial across all industries, particularly in dynamic environments like construction and road infrastructure. The application of computer vision holds promise in enhancing safety measures by automatically detecting hazards and identifying risks in real-time. Despite its potential, computer vision’s role in workplace safety remains under-explored. This study investigates the effectiveness of three pre-trained versions of the YOLOv8 model in detecting reflective vests, a critical safety element, using real-world data. The YOLOv8 model offers fast and accurate recognition, enabling efficient monitoring of work activities and timely alerts for safety breaches. Results demonstrate the model’s capability in improving workplace safety while maintaining manageable computational demands. This research underscores the significance of leveraging advanced technological solutions like CV to mitigate risks and ensure the well-being of employees in hazardous work environments.