Open AccessYOLO NETWORK TRAINING FOR FACE RECOGNITION IN MEDICAL MASKS
Author(s) -
E. K. Temyrkanova,
A. Saurambekova
Publication year - 2021
Publication title -
izvestiâ nacionalʹnoj akademii nauk respubliki kazahstan. seriâ fiziko-matematičeskaâ/izvestiâ nacionalʹnoj akademii nauk respubliki kazahstan. seriâ fiziko-matematičeskaâ
Language(s) - English
Resource type - Journals
eISSN - 2518-1726
pISSN - 1991-346X
DOI - 10.32014/2021.2518-1726.31
Subject(s) - computer science , convolutional neural network , artificial intelligence , facial recognition system , face (sociological concept) , task (project management) , face masks , field (mathematics) , computer vision , face detection , pattern recognition (psychology) , training set , covid-19 , medicine , engineering , social science , mathematics , disease , systems engineering , pathology , sociology , infectious disease (medical specialty) , pure mathematics
The detection of face masks is a very important issue for the safety and prevention of Covid-19. In the medical field, the mask reduces the potential risk of infection from an infected person, regardless of whether they have symptoms or not. Thus, the detection of masks on the face becomes a very important and complex task. The efficiency of facial recognition systems can significantly deteriorate due to occlusions, such as medical masks, hats, facial hair, and sunglasses. Currently, there are a number of different methods for recognizing objects in an image. One of the most popular methods is convolutional neural networks and their modifications. This article provides a brief description of the YOLO network, an example of training that can detect faces with a mask and without a mask, and the results of the work. The recognition model has been trained on different object recognition pre-trained models with the same data and evaluated on multiple environments to achieve good accuracy for limited identities.