The Application of Image Processing for Mask Detection

In recent years, highly contagious viruses are on the rise, making it easy for people to contract the virus from one another when exposed to it. Face masks are used to prevent cross contamination as part of an infection control strategy. Without having the mask on, this increases the chances of contracting the virus as those viruses are being transmitted through droplets and sprays in the air. Due to any inconsiderate individual that refuses to put the mask on in high-risk places like health facilities, this action can endanger the lives of their loved ones as well as people around them. Therefore, an automatic system that is able to determine if a person is wearing a mask or not is proposed. This system is completed in three different stages which are face detection, facial features detection and mask detection. The face detection is first implemented into the system to determine the location of faces in the image. Then, the facial features such as eyes, nose and lips are determined. In order to implement both approaches, the Viola-Jones algorithm along with the coding system called cascade object detector method is used to detect all individuals' faces as well as all features. Lastly, the system is able to detect if an individual is masked or unmasked by allowing the system to distinguish those facial features. If no facial features are found by the system, the individual will be categorized as masked and vice versa. Upon successful implementation, the system will be able to help reduce the spread of diseases by performing automatic mask detection in enclosed areas, especially in healthcare institutes.