<![CDATA[So far, rice companies determine the quality of rice through 2 stages, namely visual tests and laboratory tests. Laboratory tests are said to take quite a long time, while visual tests are carried out manually, by estimation or by human eye vision, so errors often occur in determining the quality of rice due to fatigue. and doubts in determining the quality of rice. Based on this problem, this research developed an application for identifying rice quality using Hue, Saturation, Value (HSV) color extraction with an identification method using K-Nearest Neighbor (KNN) and applying an evaluation results method using Euclidean Distance, in order to determine the level of accuracy. higher with digital processing. Therefore, this research carried out the process of identifying the quality of rice into 3 classes, namely medium 2, medium 1 and premium. With the KNN identification method, and the dataset used is 240 training data and 60 test data. The highest value is k=3 with an accuracy of 93.33%, precision of 93.33% and recall of 93.33%. So identifying rice quality based on HSV color image features using the K-Nearest Neighbors (KNN) method is suitable for use as intended.]]>
Copyrights © 2024
