Abstract
Tomato fruit yields have a variety, not only ripe fruit but also the condition of tomatoes that are still raw or half-ripe. By naked eye, tomato ripeness can be seen by the difference in three colors, namely red, yellow and green. Each color difference describes the level of maturity. However, human decisions in distinguishing the color of ripe tomatoes can vary and are more subjective. With the help of an image processing system, the selection of tomato ripeness can be faster and more objective. This study uses the YOLOv4 algorithm which is implemented into a mobile computing device, namely the Raspberry Pi 3, to select the tomato image captured by a camera. The results of the YOLOv4 image process will be output to the actuator in the form of a conveyor equipped with a wiper that will sort tomatoes automatically. The result of this research is a prototype with the YOLOv4 algorithm that has been trained with the tomato dataset.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call