Covid-19 Image Segmentation based on Masi Entropy and HGS Optimization Algorithm

The covid-19 epidemic has killed a lot of people over the past three years. A lot of solutions have been proposed to prevent the outbreak of the disease, including the fast diagnosis which helps to control the number of people infected across the world by providing the best treatment to the patient in a timely manner. A chest X-ray is a quick way to detect the infected area in the lungs. However, it is difficult to determine whether the lungs are infected or not? Hence, this paper proposes a powerful computer vision technology that is an efficient approach of multi-level thresholding based on the hunger game search (HGS) optimization algorithm. This algorithm used Masi's entropy as an objective function until the best solution was arrived at and, in particular, to determine the infection areas of lungs. Thus there are two kinds of images sets used to demonstrate the efficiency of the proposed approach. The first one is Non-COVID-19 chest X-ray images used as a reference and the second one COVID-19 chest X-ray images to test and detect affected areas in the lungs.