Open AccessMask Wearing Detection Method Based on the Skin Color and Eyes Detection
Author(s) -
Tian He
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1906/1/012012
Subject(s) - computer vision , computer science , artificial intelligence , task (project management) , interpretability , face (sociological concept) , skin color , judgement , face detection , feature extraction , facial recognition system , engineering , social science , systems engineering , sociology , law , political science
During the period of fighting against COVID-19, wearing masks is the main measure of protection in public places. In order to obey relevant regulations, railway stations, airports, shopping malls and other public places are equipped with staff to supervise citizens to wear masks. This method not only wastes human resources, but also has the disadvantage of significantly reducing the efficiency when the flow increases. In order to improve work efficiency, save human resources, and ensure the health and safety of staff, computer vision technology can be used to develop embedded device to deal with the task of automatic mask wearing detection. This paper proposes a mask wearing detection method based on the skin color and eyes detection. This method uses the ellipse skin model, the grayscale gradient features of eyes and the geometric relationship between eyes and other parts of face to locate the face region. And by calculating the coverage of skin color in nose and mouth area, this method gives a judgement of whether a man is wearing a mask properly. This method can basically deal with the task of mask wearing detection under the specific application scenario. It also has the advantages of small volume and strong interpretability.