Corona virus illness (COVID-19) is also an illness caused by the severe acute metabolic process syndrome (Severe Acute metabolism Syndrome) virus. Those that are infected with the Covid-19 virus experienced moderate respiratory illness and recovered with non-special treatments. However, some of us became seriously unwell and required medical attention. As a primary step in combating COVID-19 is effective screening of infected patients, with one of the key screening approaches being radiology examination using chest radiography. It was found in early studies that patients present abnormalities in chest radiography photos that are characteristic of those infected with COVID-19. Impelled by this and the affected by the open-source efforts of the analysis community, in this study we propose a CNN convolutional neural network for the detection of COVID-19 cases from chest X-ray (CXR) photos. The dataset used is COVID-19 RADIOGRAPHY database that is publicly available. All the pictures are in Portable Network Graphics (PNG) file format. We achieved 94% of training accuracy.