Implementasi Metode KNN Untuk Klasifikasi Penyakit Daun Teh Berdasarkan Fitur Tekstur GLCM

Abstract

This study evaluates the tea leaf disease classification method using Gray Level Co-occurrence Matrix (GLCM) texture feature extraction and the K-Nearest Neighbor (K-NN) algorithm. The dataset used consists of 1,800 tea leaf images divided into six categories, namely Red Leaf Spot, Algal Leaf Spot, Anthracnose, Bird Eye Spot, Brown Blight, and Healthy. Each image undergoes a preprocessing process in the form of conversion to grayscale, resizing, and normalization. The extracted GLCM global texture features include contrast, correlation, energy, and homogeneity. The classification process uses the K-NN algorithm with variations in K values ​​from K = 1 to k = 10, and is carried out on three training and test data division scenarios, namely 90:10, 80:20, 70:30. The evaluation was carried out using a confusion matrix with accuracy, precision, recall, and F1-score metrics. The results show that the model with a combination of GLCM and K-NN produces the highest accuracy of 95.83% at a value of K = 1 with a data division scenario of 80:20. These findings indicate that this approach is effective in classifying tea leaf diseases.