This research focuses on developing a micro-controller-based banana ripeness sorting tool with the implementation of the K-Nearest Neighbors (KNN) algorithm for the classification of ripeness levels based on RGB color image processing using the OpenCV library. Banana is an important fruit in society because of their high nutritional content, but manual sorting of banana fruit is a challenge for farmers and officers. The tool built uses Arduino UNO as a controller, a conveyor belt with a dynamo motor, and a servo motor for sorting. The KNN method is used for classification based on banana skin color. The results showed that the success rate of sorting reached 100% at the neighboring value of K = 3, 93.33% at K = 5, and 86.66% at K = 1. This tool can be an efficient solution for automatically sorting bananas based on ripeness level with high accuracy.