Performance Analysis of Gray Level-Based Texture Features Using Chest X-Ray Images

Abstract

A grown-up COVID-19 recognition approach for chest X-ray images is presented in this paper. The COVID-19 virus is still having a terrible impact on people's health and quality of life around the world. The first essential step in the fight against COVID-19 is effectively screening those who are infected, and imaging of the chest represents a few of the main diagnostic methods. This study used chest X-ray images, this study is intended to automatically identify COVID-19 patients while optimizing detection accuracy by employing a gray-level-based method for extracting features and some classifiers. The dataset consists of 196 COVID-19 and 196 ordinary X-rays of the chest images. The initial phase involves pre-processing a dataset of 392 chest X-ray pictures. The preprocessed images were used to extract grey-level-based characteristics in the second step.  The third phase presents the KNN classifier to attain 90.1% accuracy in patient classification.