<![CDATA[Tomatoes are plants whose fruits can be used to flavor dishes, fried rice, and can also be made into tomato juice. Due to their great benefits, the demand for tomatoes is increasing, even to the point of being exported overseas. For this reason, many farmers in Indonesia cultivate them. However, in order to be exported, the quality of the tomatoes to be shipped must meet the established standard, which is grade 1 quality. Currently, determining the quality of tomatoes is done manually by humans as quality control, which requires a long time, high operational costs, and is prone to errors. Therefore, this study will attempt to use YOLOv4 to identify tomato quality in three levels, namely good, fair, and poor. YOLOv4 will first be trained with 80% of the dataset totaling 300, and 20% will be used for testing. The test results show that YOLOv4 can correctly distinguish tomatoes with good, mid, and bad quality with an accuracy rate of to 66.2% and a detection time of less than 1 second for each object.]]>
Copyrights © 2025
