Tugas Akhir
Quail egg classification and sorting system based on yolo for production efficiency / Muhammad Ali Fikri Mubarok
Abstrak
Quail is one of the poultry species that produces eggs with high economic value and is widely farmed in Indonesia. The manual process of sorting quail eggs by quality still faces several challenges including being time-consuming relying heavily on human labor and prone to classification errors. Therefore an automated system is needed that can classify and sort quail eggs in real-time to improve production efficiency. This research aims to design and develop a quail egg classification and sorting system based on the You Only Look Once version 5 (YOLOv5) algorithm. The system is designed to detect and classify eggs into two categories namely good and defective based on the visual condition of the eggshell and automatically sort them using a servo motor. The test results show that the system is capable of classifying quail eggs with an average accuracy of 92%. The precision for the good category reached 93.75% and 90.38% for the defective category with recall values of 90% and 94% respectively. The sorting success rate reached 92% with a False Detection Rate (FDR) of 8%. In addition the system successfully reduced the sorting time from 300 seconds to 1 second significantly improving the efficiency of farmers operational time. The implementation of the YOLOv5-based quail egg classification and sorting system has proven effective in increasing production efficiency reducing manual intervention and enhancing the classification accuracy of quail eggs. This research is expected to serve as a foundation for the further development of automatic classification technology in the poultry farming sector while supporting innovation towards a modern AI-based livestock industry.