Machine Learning Approach for Rhizomes Classification Based on Color
Bayu Agustian, Maimunah

Universitas Muhammadiyah Magelang

Abstract

Rhizome plants that are often used as traditional ingredients or herbs such as temu hitam, temulawak, and temumangga. The types of rhizome are similar in color, shape, and odor feature. In this study, the types of rhizomes based on their color were classified into three classes, namely temu hitam, temulawak, and temumangga. The color of the rhizomes is obtained from image which is taken using camera. Image feature extraction process is done to get the color characteristics of the image by calculating the image RGB value. The values that have been obtained are classified using several methods in machine learning, namely Decision Tree, Naive Bayes, and KNN. The amount of data used in this study was 120 for all classes with the ratio of the number of training data and testing data is 70% and 30%. At the beginning of the classification stage, data cleaning is carried out then make training data and testing data to create a classification model. The classification results obtained the accuracy value of each method and then the method with the best accuracy was selected.

Keywords: machine learning, classification, herbal plant

Topic: Engineering