<![CDATA[The development of computer vision technology and artificial intelligence has driven innovation in automation in various fields, including the agricultural sector and fruit trading. The process of identifying fruit quality, which is generally done manually, is still vulnerable to human error and inconsistencies. Based on these problems, this study aims to develop an automated system to detect the difference between fresh and rotten fruit using a deep learning-based You Only Look Once (YOLO) algorithm integrated with the OpenCV library. The system is designed in the form of a web application that is easy for fruit sellers to use. The dataset used consists of images of apples, mangoes, and bananas labeled through Roboflow into two categories, namely fresh and rotten. The model was trained using YOLOv11, then tested with new data that had never been used before. The test results showed high performance with an accuracy of 99.01%, mAP@50 of 0.925, precision of 0.93, recall of 0.90, and F1-score of 0.91. Based on these results, the system is able to detect the condition of the fruit automatically and in real-time with an excellent level of accuracy. This implementation proves that the integration between YOLO and OpenCV is effective in improving the efficiency, accuracy, and consistency of the fruit quality identification process.]]>
Copyrights © 2025
