COVID-19, attacking the lung and respiratory system, can infect people of all ages since it is easily transmitted through contact with sufferers. However, despite having similar symptoms to pneumonia, COVID-19 is more dangerous as late diagnosis can lead to death. COVID-19 and pneumonia diagnosis can be performed using chest X-ray images. This study aims to detect COVID-19 in chest X-ray images using Gradient Vector Flow Snake (GVFS) segmentation and Hu Moment extraction, with Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) as the classification methods. Chest X-ray image data were obtained from Joseph Paul Cohen's Github open source from various hospitals in various parts of the country by taking three image classes: COVID-19, normal, and pneumonia. Hu Moment Cubic KNN outperformed other methods by producing the greatest training accuracy of 56.20% and the highest testing accuracy of 66.30%.
