COVID-19 is spread by the pandemic coronavirus and is spreading very fast in the world. Every single development impact has taken place in every direction due to the COVID-19 which has also affected the health care system. We can prevent the virus from turning the COVID-19 by wearing a mask. In this review, we will see that individuals need to distinguish regardless of whether individuals are wearing veils where there are CCTV cameras. MobileNetV1 is a machine learning tool used to create a face mask application. We will collect photos of people who will be wearing masks or not wearing masks or ordinary photographs will be used in custom CV text which will be the first step in how our model will be produced, followed by pre-data analysis and data sorting, testing, and finally modeling. The calculation is 98.73–99.80% exact in deciding if an individual is wearing a veil or not. MobileNetV1 might be picked over models like MobileNetV2 or Efficient Net for facial covering discovery because of its straightforwardness and computational effectiveness. It gives a decent harmony between speed and precision, making it reasonable for ongoing applications, for example, live video handling, where fast direction is urgent. MobileNetV1 additionally requires less computational assets, which is worthwhile in conditions with equipment constraints, like cell phones or implanted frameworks. In addition, its far reaching reception and broad documentation can work with simpler execution and investigation. While more up-to-date models like MobileNetV2 and Efficient Net offer enhancements in precision and proficiency, the decision of MobileNetV1 frequently mirrors a useful thought of asset limitations, adequate exactness for the errand, and the requirement for effective power utilization, especially in battery-worked gadgets.