Automatic Detection of Covid-19 based on Xception Network with Optimized CNN

Abstract

An exact and quick finding of Covid-19 patients plays a major part in the initial period of medicinal treatment and prevention. Automatic recognition of COVID-19 cases utilizing lung X-ray images may help lessen the effect of this infection on human civilization. In recent years, computer vision is the major solution for diagnosing the covid-19 disease by utilizing X-ray images. Besides, many researchers had presented efficient Artificial Intelligence (AI)methods for identifying Covid-19 disease. Nevertheless, the accuracy and time consumption of the model is further to be improved. Thus, in this work, we proposed a novel Xception network with an optimized convolution neural network (OCNN). The CNN architecture is enhanced by using the adaptive seagull optimization (ASO) algorithm. There are three stages of the approach: pre-processing, feature extraction, and classification. At first, a median filter is applied to each image to reduce the noise present in the input image. Then, the noise-free images are fed to the Xception network to extract the features of the images. A convolution network is used to classify an image as positive or negative after feature extraction. The performance of the proposed approach is analyzed based on various metrics and performance compared with other techniques.