Mask detection using automated machine learning techniques: A response strategy to the COVID-19 pandemic

Abstract

Amidst the global backdrop of the COVID-19 pandemic, the imperative of judicious mask usage has emerged as a pivotal facet of public health. Conventional monitoring methods, such as manual checks, prove inadequate in addressing the demands of a pandemic of this scale. Although traditional machine learning techniques offer a potential solution for mask detection, their time-consuming nature poses challenges for real-time applications. In response, this study delves into the realm of automated machine learning techniques, focusing on the EasyDL platform due to its user-friendly interface and robust algorithms. This study explores automated machine learning for efficient mask detection during the COVID-19 pandemic. Using the EasyDL platform, we achieved a 94.3% precision rate in mask detection with only 213 minutes of training on a 6006-image dataset. This approach proves more time-effective than traditional methods, making it suitable for real-time applications and large-scale monitoring. The combination of high accuracy and efficiency showcases the potential of automated machine learning in public health, enabling swift responses to health threats. To sum up, this research symbolizes a significant progress in terms of applying artificial intelligence to address the chanlleges of the public health. It could be estimated that these discoveries will stimulate further researches, and pave the way for developing more efficient and more effective mask detecting tools, which would be likely to contribute to the progress of public health management in the future.