Corona Virus Detection using Digital Image Processing

Abstract

The Corona Virus epidemic has thrown the entire world into chaos, bringing life to a screeching halt and taking thousands of lives. It continues to pose a serious hazard to public health. The COVID-19 pandemic's global expansion has resulted in considerable losses. The fact that the corona virus was detected quickly is one of the most crucial challenges that medical and healthcare departments are dealing with. As a result, it's critical to confirm the suspected case's diagnosis, not only to make the following step easier for the patients, but also to limit the number of infected persons. In image identification, the deep convolutional neural network has made tremendous progress, particularly in the field of auxiliary medical diagnosis technologies. Neural networks have been effectively utilized to identify pneumonia from CT scans, with results that outperform radiologists. As a result, deep learning has played a critical part in the response to the COVID-19 outbreak, making it possible to appropriately judge and respond to the outbreak. This project presents a model that employs the Convolutional Neural Network (CNN) deep learning algorithm with some basic layers. The proposed method applies deep learning's analytical and diagnostic capabilities to CT scan images, presenting an image classifier based on the CNN and VGG16 models to classify chest CT scan images. The goal of this model is to transfer learning, model integration, and classification of these images into two categories: normal and Covid19-positive.