Performance Comparison Study of Cherry Tomato Ripeness Detection Using Multiple YOLO Models

Millions of tons of cherry tomatoes are produced annually, with the harvesting process being crucial. This paper presents a deep learning-based approach to distinguish the ripeness of cherry tomatoes during the harvesting process. It specifically analyzes the performance of the YOLO (You Only Look Once)v5 and YOLOv8 models, which are designed to rapidly detect the ripeness of cherry tomatoes in real time. The proposed system is expected to resolve rural issues caused by population decline and enhance productivity in cherry tomato harvesting environments. Future research will aim to apply this technology to harvesting robots for the automatic recognition and harvesting of cherry tomatoes.