SISTEM DETEKSI COVID-19 BERDASARKAN METODE GRADIENT VECTOR FLOW SNAKE, SUPPORT VECTOR MACHINE, DAN K-NEAREST NEIGHBOR

Covid-19 is a virus that has the ability to spread very quickly, causing the World Health Organization (WHO) to designate Covid-19 as a pandemic on January 30, 2020. The weakness of the system on chest x-ray images for covid-19 detection is constrained by the problem of doctors' accuracy in assess chest x-ray images. In previous studies, image detection was carried out on all parts of the chest x-ray image, while in this study it was focused on the lungs using Gradient Vector Flow Snake segmentation. This study aims to detect the covid-19 virus using the Gradient Vector Flow Snake (GVFS) segmentation method, the Gray Level Cooccurrence Matrix (GLCM) and Hu Moment extraction methods, and the classification method used is the Support Vector Machine (SVM) and K-Nearest Neighbor (KNN). This study uses chest x-ray image data which consists of 3 image classes, namely covid-19, normal and pneumonia. This study produces the best accuracy value of 83.20% on the Gray Level Co-occurrence Matrix Weighted KNearest Neighbor for the training process and 93.65% on the Gray Level Cooccurrence Matrix Fine K-Nearest Neighbor for the testing process