To Design and Implementation of SARS-COVID-19 Detection Using CNN and VGG19 under Deep Learning Techniques

Abstract

On behalf of the World Health Organization (WHO) reports the COVID-19 (coronavirus) pandemic makes daily threats to the health system—resources for health in most countries must be more and more reasonable for the community. There are different problems, such as the number of health persons, the number of intensive care units, or the number of beds. According to previous research, the main problem of SARS-COVID-19 is detecting the human body in COVID-19 infection. This research focuses on SARS-COVID-19 detection when a man's body is infected due to the coronavirus. Research indicates comparing three human body types: standard, pneumonia, and SARS-CoV-19 detection. The study uses deep learning techniques to detect the human body precisely. The two methods of Deep Learning (DL), Convolutional Neural Network (CNN), and Visual Geometric Group (VGG19) are used for COVID-19 detection and comparison to normal body, pneumonia, and SARS COVID-19 detection under four performance parameters as accuracy, recall, F-score and precision. The outcome of this research is that VGG19 is better than CNN under SARS-COVID-19 detection.