Open AccessEkstraksi Fitur Citra Radiografi Thorax Menggunakan DWT dan Moment Invariant
Author(s) -
I Gede Pasek Suta Wijaya,
Ditha Nurcahya Avianty,
Fitri Bimantoro,
Rina Lestari
Publication year - 2021
Publication title -
j-cosine (journal of computer science and informatics engineering)
Language(s) - English
Resource type - Journals
ISSN - 2541-0806
DOI - 10.29303/jcosine.v5i2.423
Subject(s) - pattern recognition (psychology) , artificial intelligence , invariant (physics) , discrete wavelet transform , haar wavelet , medicine , computer science , mathematics , wavelet , wavelet transform , mathematical physics
COVID-19 is an infectious disease caused by thecoronavirus family, namely severe acute respiratorysyndrome coronavirus 2 (SARS-CoV-2). The fastest methodto identify the presence of this virus is a rapid antibody or
antigen test, but confirming the positive status of a COVID-19 patient requires further examination. Lung examination
using chest X-ray images taken through X-rays of COVID-19patients can be one way to confirm the patient's conditionbefore/after the rapid test. This paper proposes a featureextraction model to detect COVID-19 through chestradiography using a combination of Discrete WaveletTransform (DWT) and Moment Invariant features. In thiscase, haar wavelet transform and seven Hu moments wereused to extract image features in order to find unique featuresthat represent chest radiographic images as suspectedCOVID-19, pneumonia, or normal. To find out theuniqueness of the proposed features, it is coupled with thekNN and generic ANN classification techniques. Based on theperformance parameters assessed, it turns out that thewavelet-based and moment invariant thorax radiographicimage feature model can be used as a unique featureassociated with three categories: Normal, Pneumonia, andCovid-19. This is indicated by the accuracy value of 82.7% inthe kNN classification technique and the accuracy, precision,and recall of 86%, 87%, and 86% respectively with the ANNclassification technique.