Klasifikasi Penyakit Glaukoma dengan Menggunakan Metode Support Vector Machine dengan Ekstraksi Local Binary Pattern (LBP) dan Gray Level Co-Occurrence Matrix (GLCM)

Abstract

Glaucoma is a disease caused by increased intraocular pressure in the eye, which can lead to vision loss. It causes a characteristic optic nerve head to appear on funduscopic examination. Glaucoma is the second leading cause of blindness in the world after cataract. It was responsible for 8% of blindness, including refractive errors in 2010. The application of information technology in detecting glaucoma disease plays an important role in providing information to detect glaucoma disease early. The SVM method is one of the machine learning methods used for classification and regression processes. SVM utilizes optimal linear or non-linear separators to separate two classes of data. Some of the feature extractions used are LBP which is defined as the size of the grayscale texture produced by the surrounding texture. Grayscale texture is the grayscale converted into binary samples that use thresholds to describe texture properties. There is an extension of the GLCM feature that is processed by first calculating the angle and distance, then analyzing how often the summation of contrast differences in the image at each pixel appears. Comparison of data schemes used ranging from 70:30, 80:20, and 90:10 resulted in the best accuracy value in the 90:10 data comparison scheme in LBP feature extraction with a value of 72.5%