<![CDATA[As a regional flagship commodity with a promising selling value, the process of grouping the maturity level of Medan Orange is still dominated by manual visual techniques. This often triggers data inconsistency and requires a long duration of processing due to personnel subjectivity factors. This research aims to compare the performance of two machine learning algorithms, namely KNN and SVM, in classifying the maturity level of Medan Orange fruit based on digital images. The dataset used is a primary dataset collected directly from Medan Orange farmers in field conditions. The research stages include image acquisition, pre-processing, extraction of HSV-based color features and GLCM-based textures, as well as classification of maturity levels into three classes, namely raw, semi-cooked, and mature. The performance of both algorithms is evaluated using accuracy, precision, and recall metrics. The research results show that the KNN algorithm has a superior performance compared to SVM, with an accuracy rate of 96,25%, while SVM produces an accuracy of 91,25%. This result shows that KNN is effective and more suitable to be applied to the automation system of classification of the maturity of Medan Orange fruit based on digital images.]]>
Copyrights © 2026
