2024, 8, 9 | Christine Dewi, Danny Manongga, Hendry, Evangs Mailoa and Kristoko Dwi Hartomo
This study focuses on developing an accurate face mask detection system using deep learning and the YOLOv8 object detection algorithm. The primary goal is to annotate and classify face mask entities in images, particularly in the context of the COVID-19 pandemic, where mask-wearing guidelines are crucial for public health. The researchers combined the Face Mask Dataset (FMD) and the Medical Mask Dataset (MMD) to create a comprehensive dataset for training and testing the model. The YOLOv8 model, known for its advanced architecture and performance, was used to detect and identify face masks. The study achieved a detection accuracy of 99.1% for the "good" class (properly worn masks), surpassing previous models. The results demonstrate that the proposed model is a reliable method for detecting faces obscured by medical masks and outperforms other models in terms of both accuracy and precision. The study also discusses the limitations of YOLOv8, such as challenges with small objects and partial occlusions, and suggests future research directions, including the application of explainable artificial intelligence (XAI) for medical mask identification.This study focuses on developing an accurate face mask detection system using deep learning and the YOLOv8 object detection algorithm. The primary goal is to annotate and classify face mask entities in images, particularly in the context of the COVID-19 pandemic, where mask-wearing guidelines are crucial for public health. The researchers combined the Face Mask Dataset (FMD) and the Medical Mask Dataset (MMD) to create a comprehensive dataset for training and testing the model. The YOLOv8 model, known for its advanced architecture and performance, was used to detect and identify face masks. The study achieved a detection accuracy of 99.1% for the "good" class (properly worn masks), surpassing previous models. The results demonstrate that the proposed model is a reliable method for detecting faces obscured by medical masks and outperforms other models in terms of both accuracy and precision. The study also discusses the limitations of YOLOv8, such as challenges with small objects and partial occlusions, and suggests future research directions, including the application of explainable artificial intelligence (XAI) for medical mask identification.