Your browser doesn't support javascript.
Detection of Proper Orientation of Face Mask Using Deep Neural Network
12th International Conference on Biomedical Engineering and Technology, ICBET 2022 ; : 44-50, 2022.
Article in English | Scopus | ID: covidwho-1962430
ABSTRACT
The pandemic caused by the coronavirus resulted in health protocols regulations to prevent the disease's spread. The World Health Organization recommended using facemasks as a preliminary measure that governments around the world could use help prevent the outbreak. The study's goal is to implement computer vision for facial detection with faces wearing a face mask. The study aims to (1) capture facial features of masked faces using a Web Camera;and (2) detect a face with proper face mask orientation using deep neural network and mobile net model. (3) For the system to detect whether a person is appropriately wearing a face mask following IATF and WHO standards. The study only allowed one person to face the camera while wearing Standard Surgical Face Masks and being within eye level of the camera if ambient lighting was available. Face Mask Detection, with the Raspberry Pi, has checked the proper face mask orientation for 50 different individuals. The Confusion Matrix Testing determined the following values 80 percent for the FMD's Accuracy, 62.5 percent for the FMD's Precision, 71.4 percent for the FMD's Recall, Sensitivity, and True Positive Rate, 66.6 percent for the F1 Score, and 16.6 percent for the False Positive Rate. © 2022 ACM.
Keywords

Full text: Available Collection: Databases of international organizations Database: Scopus Language: English Journal: 12th International Conference on Biomedical Engineering and Technology, ICBET 2022 Year: 2022 Document Type: Article

Similar

MEDLINE

...
LILACS

LIS


Full text: Available Collection: Databases of international organizations Database: Scopus Language: English Journal: 12th International Conference on Biomedical Engineering and Technology, ICBET 2022 Year: 2022 Document Type: Article