Deep Neural Architecture for Face mask Detection on Simulated Masked Face Dataset against Covid-19 Pandemic

Abstract

The dangerous COVID-19 (SARS-CoV-2) is rising steadily and globally, with more than 72,851,747 confirmed cases observed to WHO including 1,643,339 deaths till 17 December 2020. The country's economy is now almost fully halted, people are stuck up and investment becomes deteriorating. So, this is turning to worry of the government for a development and health. Health organizations are often desperate for evolving decision-making innovations to overcome this viral virus and encourage people to receive rapid and effective responses in realtime. Thus, it is important to create auto-mechanisms as a preventive shield to ensure healthy humanity against SARS-CoV-2. Advanced analytics methods and other strategies could also empower researchers, learners and the pharmaceutical industry to acknowledge the hazardous COVID-19 and speed it up care procedures by efficiently testing vast volumes of research data. The prevention method consequence is being used to effectively manage, calculate, forecast and monitor current infected people and future potential cases. Therefore, we proposed CNN and VGG16 based deep learning models to incorporate and enforce AI-based precautionary measures to detect the face mask on Simulated Masked Face Dataset (SMFD). This technique is capable of recognizing masked and unmasked faces to help monitor safety breaches, facilitate the use of face masks, and maintain a secure working atmosphere.

WebFace260M: A Benchmark for Million-Scale Deep Face Recognition

“…Due to the sudden outbreak of the coronavirus epidemic, there is yet no comprehensive real-world masked face recognition benchmark. Evaluations on simulated masked images [59], [61] may result in questioned conclusions, and small-scale masked sets [10], [12], [17] can not comprehensively reflect the performance of algorithms. Real-world masked test set RMFRD [95] consists of 525 identities and 5K masked faces, but there exist annotation noises.…”

“…In contrast with simulated [59], [61] or relatively small [10], [12], [17], [95] masked test sets, a real-world comprehensive benchmark for evaluating MFR is developed in this work. Based on the SFR identities, we further collect masked faces for these celebrities.…”

M2FRED: Mobile Masked Face REcognition Through Periocular Dynamics Analysis

“…In [35], the authors propose an advanced deep learning model for face mask detection in real-time video streams. In this regard, Negi et al [36] employ two well-known deep neural network architectures with transfer learning for face mask detection using the Simulated Masked Face Recognition Dataset. Recently, the focus has shifted to subject recognition in the presence of masks, which is nowadays known as Masked Face Recognition (MFR).…”

Identifying corn leaves diseases by extensive use of transfer learning: a comparative study

“…It can take images and pull out features layer after layer. New studies have shown that intelligent computer processor vision is getting better at things like finding brain and breast cancer [15], finding cars without helmets [16], and finding face masks [17]. During the review of extant literature, different deep neural networks [18]-[21] are used to find the correct results.…”