An Efficient Innovative Masked Face Recognition Using FaceNet Algorithm over Visual Geometry Group (VGG-16) Algorithm for Better Efficiency

Aim: The COVID-19 virus has made wearing masks a habit for living beings, the objective of this project is to provide a system with better accuracy for masked facial identification using the FaceNet Algorithm in comparison with the Visual Geometry Group Algorithm. Materials and Methods: In a total of $\mathbf{1 1 2}$ samples, Each group has $\mathbf{5 6}$ samples and the number of iterations is $\mathbf{1 0}$ for each group. The $\mathbf{G}$ power for calculating statistical tests is set at 80%. A total of 2800 images which make 2240 training images and 560 tested images-make up the research dataset, which is obtained from Kaggle.com combined with self-obtained images. Results: The accuracy results for the Innovative FaceNet algorithm is (87.7%), and the Visual Geometry Group (VGG-16) technique is ($\mathbf{9 0. 2 \%}$). The significance value is 0.001 ($\mathbf{p}\lt0.05$) which is statistically significant for the two algorithms considered for identity recognition of masked faces. Conclusion: The accuracy has significantly improved in the Visual Geometry Group (VGG-16) algorithm compared to the FaceNet approach.