Identification of Leaf Diseases of Medicinal Plants Using K-Nearest Neighbor Based on Color, Texture, and Shape Features

Medicinal plants are plants with significant potential for development in Indonesia, as the Indonesian population is one of the largest users of medicinal plants in Asia. However, with the passage of time, the production of medicinal plants has not been matched by proper cultivation practices, making the leaves of these plants susceptible to diseases. To obtain high-quality plants, early efforts need to be made by farmers to recognize the types of diseases and provide appropriate treatment. Currently, the process of plant disease identification is often carried out by farmers through visual observation of the patterns of damage on various plant organs, including leaves, making the determination of the type of disease subjective [1]. This method leads to farmers using incorrect remedies or pesticide applications to treat medicinal plant leaf diseases (Mungki Astiningrum). To assist farmers in identifying disease types more accurately, early efforts should be made to equip them with adequate knowledge about plant diseases [2]. Through this research, it is hoped that the identified issues can assist farmers, especially in Bali, in the process of disease type recognition and provide appropriate treatment. The main objective of this study is to identify diseases in medicinal plants through leaf image analysis based on color and texture feature extraction, by comparing the K-Nearest Neighbor (K-NN) and Naïve Bayes methods. This comparison aims to determine which method is most suitable for identifying leaf diseases in medicinal plants.