<![CDATA[Fruit quality is an important factor that affects nutritional value, consumption safety, and market value of agricultural products. Apples, as one of the most widely consumed fruits, are prone to quality degradation due to spoilage, which is often difficult to accurately identify through human visual observation. Manual sorting of apples is subjective, time-consuming, and prone to errors. Therefore, this study aims to develop an automatic apple quality detection and classification system using the You Only Look Once version 5 (YOLOv5) deep learning method. Apple quality is classified into two categories, namely fresh apples and rotten apples, based on digital images. The dataset used in this study consists of 4,035 images obtained from the Roboflow platform, comprising 2,925 training images, 707 validation images, and 403 testing images. All images were resized to 640 × 640 pixels without data augmentation. The model was trained for 50 epochs using GPU acceleration on Google Colab. Model performance was evaluated using a confusion matrix on the testing dataset. The experimental results show that the YOLOv5 model successfully classified all testing images correctly without any misclassification, indicating excellent detection and classification performance. These results demonstrate that YOLOv5 is an effective and reliable method for automatic apple quality detection and has strong potential for application in agriculture and the food industry to improve efficiency and accuracy in fruit quality inspection.]]>
Copyrights © 2026
