Implementation of The K-Means Clustering Algorithm in Determining The Rate of Indramayu Mango Fruit

This research aims to classify the ripeness levels of Indramayu mangoes using the K-Means Clustering algorithm based on HSV (Hue, Saturation, Value) color features. The process begins with capturing mango images, followed by preprocessing steps such as normalization and resizing to enhance image quality. Next, color feature extraction is conducted, focusing on the Hue value as an indicator of color changes that characterize ripeness levels. The optimal number of clusters is determined using the Elbow method, resulting in two clusters: ripe mangoes and unripe mangoes. The clustering quality evaluation is performed using the Silhouette Score, which indicates an accuracy of 80%. The results demonstrate that the K-Means algorithm successfully classifies Indramayu mangoes, generating 495 image data divided into two main categories. This study contributes to improving the efficiency of automated mango ripeness classification, with potential applications in the agricultural industry.