Identification of Herbal Plants Using Morphology Method and K-Nearest Neighbour Algorithm (KNN)

Abstract The large number of herbal plants and the lack public knowledge about types medicines make it difficult for people to distinguish medicinal plants. So, provide information public, an identification system that could identify recognize a plant is needed. Leaves identificiation based on Morphological method with K-Nearest Neighbor algorithm. Leafs data will be used in this study Basil Leaves, Orthosiphon leaves, Painted Nettle Betel Teak leaves which are categorized as flowering from Lamiaceae or Labiatae family classification process. Image processing techniques obtain pattern shape characteristics each leaf known recognition morphological features. This using 200 separated into 2 types, 70% basic 30% test data. With Feature Extraction area parameters well compactness metric features leaves. accuracy results confusion matrix testing K3 values obtained by 80%, K5 K7 76.67% K9 showing 78, 67%.