<![CDATA[Classifying fruit ripeness is a crucial stage in agriculture and food processing, ensuring optimal product quality standards. Specifically for pineapples, assessing ripeness manually requires considerable time and experience. According to data from the Bolaang Mongondow Department of Trade in 2019, around 8.75% of pineapples in Lobong Village—the largest pineapple-producing village in Bolaang Mongondow Raya—were spoiled or wasted, indicating that the harvest timing for pineapples was often inaccurate. In response to this challenge, research was conducted to introduce a deep learning methodology utilizing the Convolutional Neural Network (CNN) with YOLOv8 (You Only Look Once) version 8 to autonomously classify the ripeness stages of pineapples. By developing a pineapple ripeness classification system using the YOLO algorithm, it is expected to address these issues and assist farmers in determining the ripeness level of pineapples for sales and processing purposes. The use of CNN with YOLOv8 is chosen due to its ability to quickly and accurately detect objects based on images in real time. Additionally, the trained YOLOv8 model can classify the ripeness levels of pineapples, thereby helping farmers sort the fruit according to its ripeness stage. Tests conducted to measure the performance of the YOLOv8 algorithm in detecting and classifying pineapple ripeness showed promising results with an mAP value of 81%, Precision of 70.5%, and Recall of 75.9%, producing a satisfactory level of accuracy across various levels. This research can optimize the process of sorting pineapple ripeness stages, thereby improving product quality and enhancing farmers' competitiveness in both domestic and export markets]]>
