Study on the Implementation of Image Processing Systems for Detection of Mask Usage

Abstract

There are still many people who do not comply with government regulations regarding the importance of wearing masks during the Covid-19 pandemic. Image processing system can used to detect objects in the form of a human face that uses a mask and or does not use a mask in real time. It is built using the Python programming language and several libraries including OpenCV, Tensorflow, Keras, Imutils, NumPy, Matplotlib, and SciPy. Objects can be detected and defined from a distance of 50-200 cm. The intensity of the light and the angle at which the photo is taken greatly affect the detection results. The percentage of accuracy of people who wear masks and who do not wear masks in this project has exceeded 80%.