Optimizing Coffee Ripeness Classification Using Yolov5 for Automated Detection and Sorting

The quality of agricultural products, particularly coffee beans, is crucial in today's global market, which demands precise ripeness classification due to its high commercial value. Traditional manual methods in coffee plantations, heavily reliant on human labor to determine quality, often result in inefficiencies and inaccuracies. To address this issue, this study developed an automated coffee ripeness detection system using the YOLOv5 machine learning algorithm, combined with Raspberry Pi, webcam, and servo motor. By integrating YOLOv5, the system enables real-time classification of coffee beans into three categories: ripe, unripe, and rotten, with an average accuracy of 90% during real-time testing. This system not only reduces dependence on manual labor but also improves process efficiency across various environmental conditions. The findings suggest that the application of this technology can significantly enhance productivity in the coffee industry, while providing a foundation for further advancements in automation and classification methodologies in the agricultural sector.