Public Health Face Mask Detection of Covid-19 Utilizing Convolutional Neural Network (CNN)

Coronavirus Disease (COVID-19) has gained special attention worldwide. The transmission of the COVID- 19 virus has globally spread, including Indonesia. Indonesia has been experiencing a crisis in the area of health and economic sectors due to the COVID-19 pandemic, encouraging the government to implement a program such as Large- Scale Social Restrictions (PSBB) in each region because of the rapid spread of the virus, urging public services and places to wear masks. During this time, the detection of masks is manually conducted with observations from security personnel, in which human error often occurs at the time of mask detection. This study proposes a mask detection system (Face Mask Detection) by utilizing image processing. Face mask detection is commonly acknowledged as the detection of whether a person wears a mask. The study applied the Convolutional Neural Networks (CNN) method. More specifically, Neural Networks' Convolutional learning algorithm uses the extraction of features from imagery that would be further studied by several hidden layers. The system thus far employs a combination of object detection classification, image, and object tracking to develop a system that detects the usage of a mask in images or videos. The datasets were variously obtained with images of faces using hijabs, hats, and not using attributes. In addition, the images were gathered from various countries such as Asia, Europe, and the Americas. The face mask detection system obtained an accuracy value of 0.9893% and training loss of 0.0284% during the test