Freshness Condition Detection, Size Classification with Weight Prediction and Sortation of Philippine Egg Produce Using YOLOv8 and OpenCV

Manual egg grading is a time-consuming and laborintensive operation in the poultry industry. Multiple personnel are needed in the operation since multiple processes are involved in the production chain. Nevertheless, manual egg grading is still being practiced and carried out, particularly by small and medium-sized poultry farms. These farms often resort to this method due to the high cost of egg-grading sorter machines available on the market. This study, therefore, introduces a costeffective system that leverages computer vision for grading eggs sequentially. The system initially detects egg freshness with YOLOv8 and then predicts weight and size using OpenCV. Eggs classified as good undergo complete grading and are sorted by size per the Philippine National Standard for Table Eggs (PNS/BAFS 35:2018). Eggs classified as rotten, however, are immediately sorted into the tray designated for rotten eggs, skipping the weight and size analysis. With an overall accuracy of 93.75%, the system and method established in this study present a great alternative to traditional egg grading methods. The results support the system's potential for usage in small-scale production settings.