The COVID-19 epidemic is causing a major outbreak in more than 150 countries worldwide, which has a profound impact on the health and well-being of many people worldwide. One of the most important steps in combating COVID-19 is the ability to detect infected patients early, and place them under special care. Diagnosis of radiography and radiology is probably one of the fastest ways to diagnose patients. To accelerate the detection of COVID-19 infections by X-ray images, this study developed a new diagnostic platform using a convolution neural network (CNN) that can assist radiologists with diagnostic detection of COVID-19 pneumonia. Such a tool can save time in translating chest X-rays and increase accuracy and thus improve our medical capacity to detect and diagnose COVID-19. The idea of the study is that a collection of X-ray medical imaging images are used to train CNN which can distinguish between sound and useful information and then use this training to interpret new images by detecting patterns that show certain diseases such as corona infection in individual images. Continue to use transfer learning strategies such as VGG 16, RESNET50, Inception V3 and find out which process provides the best accuracy in obtaining COVID-19.