Implementasi Pengolahan Citra Digital Untuk Mengidentifikasi Penyakit Tanaman Anggrek Menggunakan Algoritma K-Nearest Neighbor (K-NN)

Abstract

Orchids are one of the ornamental plants that are widely cultivated in Indonesia because of their beautiful flower strands and various flower patterns and have a high selling value, so many people cultivate these orchids. However, not many people know that orchids are one of the plants that tend to be prone to disease. Currently, the process of identifying orchid plant diseases is still done manually, causing other orchids to become infected with the disease. Therefore, it is important to create a system that can identify orchid plant diseases using digital image processing using Gray-level Co-occurrence Matrix feature extraction and the K-nearest Neighbor (K-NN) method. The first process in this research is to convert the RGB image to Grayscale before extracting features using the GLCM algorithm from several orchid images collected with different classes as training data. The training data is used by the KNearest Neighbor algorithm to determine the identity of new data with the closest distance value to produce a classification. The result of this research is that the system can classify orchid plant diseases which have been divided into 4 classes, namely anthracnose, flower spot, leaf spot, and fusarium wilt. Based on testing with 400 training data and 32 test data, it produces an accuracy rate of 84%.