OvOptix: Optical Based Egg Sorting Technology

This paper introduces a machine learning and AI-based system for detecting internal and external irregularities in table eggs, addressing the limitations of current egg sorting machines. By leveraging Internet of Things (IoT) technology, the system determines if an egg is defective. YOLOv8 and machine vision were integrated for enhanced detection, using a web camera to capture images and reveal defects not visible through traditional methods. The system's lightweight design is ideal for real-time detection with limited computational resources. A custom egg tray facilitated dataset collection, achieving 96.8% effectiveness, 95.1% reliability, and 91.2% usability. The system demonstrated live detection with an average speed of 450 milliseconds, proving that the OvOptix machine can efficiently meet the high production demands of the egg market.