For enhancing accuracy on real-time face mask detection, a novel convolutional neural network algorithm is used instead of a deep neural network algorithm.

Abstract

Aim: This paper is a comparative study of novel convolutional neural networks (N-CNN) and deep neural network algorithms to enhance the reliability of real time face mask detection. Materials and Methods: Sample size of Novel convolutional neural network algorithm (N=20) and deep neural network algorithm (N=20) methods are simulated by varying the NCNN parameter and deep neural parameter to optimise the pH sample size is calculated using G power 80% for just two groups and can find 40 samples utilised in this work. Result: According to obtained results, a novel convolutional neural network has significantly better accuracy (94.98%) when compared with deep neural network accuracy (85.41%). The statistical significance difference between novel  convolutional neural network and deep neural network was found to be p=0.000 (p<0.05). Conclusion: Novel  convolutional neural network algorithm produces greater outcomes in predicting face masks to improve accuracy percentage than deep neural network algorithm.