Hairnet Detection and Alerts for Hygiene Compliance in Food Processing using Deep Transfer Learning Technique

The study focuses on creating and assessing a strong system for detecting and alerting the presence of hairnets to ensure hygiene compliance in food processing. This is achieved by utilizing the YOLOv8 architecture with a transfer learning method. This study aimed to develop a model that can effectively distinguish between those who are wearing hairnets and those who are not, specifically in food processing facilities, to ensure strict compliance with hygiene regulations. The YOLOv8 architecture, utilized via transfer learning, exhibited outstanding performance with a Mean Average Precision (mAP) of 98.2%, Precision of 96.3%, and Recall of 96.5%. The evaluation of the model was performed on data that had not been previously seen, demonstrating high rates of detection for those who adhere to cleanliness requirements, despite limits in nearby settings. The results validate the system's competence in precise object detection, while also emphasizing areas that require future improvement. The results of this study show potential for practical use, highlighting the effectiveness of the system in enforcing cleanliness standards in food processing facilities. Subsequent efforts will prioritize enhancing the model's functionalities to overcome its limits, so guaranteeing thorough and accurate identification in all situations, leading to enhanced practicality in real-world settings.