ABSTRACT
Covid-19 remains the worldwide highlight because it is still growing rapidly and has greatly impacted human activities. Preventing its transmission by detecting to allow other actions to be taken continues to be carried out. Various research efforts have been performed to detect Covid-19. Along with developing its detection, technology can be conducted by image processing or machine learning. The detection in this study was carried out using X-ray images of Covid-19 positive people, totaling 101 images, propagated through pre-processing to 404 images. Then, these images were compared with the X-ray images of normal people amounting to 202 and the X-ray images of pneumonia-positive people totaling 390. The extraction process was performed using the Haar wavelet transformation by classifying the data using Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) methods. The Fine KNN model obtained the best accuracy with an average of 94.66%. © 2022 IEEE.
ABSTRACT
COVID-19 has significantly influenced living in recent years. Almost all countries have carried out all limitations to reduce its spread. Detection is highly required for further handling of COVID-19. In this study, the detection was performed using classification on 1,184 X-ray images, specifically 404 X-ray images of COVID-19 positive people, 390 X-ray images of normal people and 390 X-ray images of pneumonia positive people. The image data were extracted with the Haar wavelet algorithm and classified using the Support Vector Machine (SVM) and K-Nearest Neighbor (KNN);each had three classification models. The Quadratic SVM model obtained the best result with an accuracy of 79.8%. © 2022 IEEE.