KLASIFIKASI PARU-PARU PADA CITRA SINAR-X COVID-19 MENGGUNAKAN SEGEMENTASI OTSU, EKSTRAKSI FITUR GLCM, DAN SUPPORT VECTOR MACHINE

Irawan Dharma Sukowati (Faculty of Engineering UNWIKU)
Dody Wahjudi (Faculty of Engineering UNWIKU)
Isra' Nuur Darmawan (Faculty of Engineering UNWIKU)
Rahardian Luthfi Prasetyo (Faculty of Engineering UNWIKU)

Teodolita: Media Komunikasi Ilmiah di Bidang teknik
Vol 26, No 2 (2025): Teodolita: Media Komunkasi Ilmiah di Bidang Teknik

Lung damage due to Covid-19 can be analyzed through chest X-ray images, but manual interpretation is subjective. This study aims to classify lung damage in Covid-19 X-ray images using Gray Level Co-occurrence Matrix (GLCM) texture feature extraction and machine learning methods. The dataset consists of 100 labeled X-ray images stored in Excel and CSV formats. The research stages include image preprocessing, GLCM feature extraction, model training, and evaluation using precision, recall, f1-score, and accuracy. The results show that the model is capable of classifying with fairly good accuracy even though the test data used is only a portion of the dataset. The SVM model used produces an accuracy of 57% on the test data with fairly stable classification performance. This approach has the potential to be used as a support system for diagnosing lung damage due to Covid-19.