Research Article
Comparative Analysis of Masked and Unmasked for Face Recognition Using VGG Face and MTCNN
@INPROCEEDINGS{10.4108/eai.5-10-2022.2327473, author={Hanif Naufal Arif Sunarko and Risanuri Hidayat and Rudy Hartanto}, title={Comparative Analysis of Masked and Unmasked for Face Recognition Using VGG Face and MTCNN}, proceedings={Proceedings of the 5th International Conference on Applied Engineering, ICAE 2022, 5 October 2022, Batam, Indonesia}, publisher={EAI}, proceedings_a={ICAE}, year={2023}, month={6}, keywords={face recognition covid-19 mask dataset vgg-face mtcnn}, doi={10.4108/eai.5-10-2022.2327473} }- Hanif Naufal Arif Sunarko
Risanuri Hidayat
Rudy Hartanto
Year: 2023
Comparative Analysis of Masked and Unmasked for Face Recognition Using VGG Face and MTCNN
ICAE
EAI
DOI: 10.4108/eai.5-10-2022.2327473
Abstract
Face recognition is a system that is widely used in various fields such as security, attendance system, and other fields. Currently Covid-19 is still a major problem around the world and almost everyone is protecting themselves with masks. This is a problem for the face recognition system. This happen because most of the faces are covered by masks so that face recognition system will be difficult to recognize the face. This paper will do a comparison between a dataset without a mask and a mixed dataset. This study was conducted to find out how the effect of the dataset used on the accuracy of face recognition system either with masks or without masks and to find out how well the performance of face recognition with different dataset. VGG Face and MTCNN are used to detect and recognize faces based on landmarks. This study compares the level of accuracy, level of precision and level of sensitivity. The result shows that using a mixed dataset containing masked and unmasked faces will increase the accuracy rate from 86.7% to 93.3%. For the level of precision increased from 87.7% to 93.5%. And the Sensitivity level increased from 86.7% to 93.3%.
