Development of an automated fruit classification system by using computer vision and deep learning

With the rapid development of technology in the Industry 4.0 era, computer vision and deep learning have emerged as key technologies supporting industrial systems such as classification and quality inspection with high accuracy, optimizing processes in production lines. In recent years, computer vision and data-driven intelligent systems have played an increasingly important role in industrial automation. Moreover, smart warehouse systems minimize unnecessary steps in product storage, optimizing the time required for import and export in industrial production lines. This paper proposes an automated import/export system integrated with the YOLOv8 network to classify various fruits (kumquats, longans, cherry tomatoes) and store them in an automated warehouse system. The import/export data is recorded on a personal web platform to track the system's input and output volumes. A multi-threading mechanism is also applied to ensure real-time data processing. Experimental results indicate that the overall system achieves a high accuracy rate of approximately 98%. A demonstration video illustrating the system operation and setup procedure is available in the project repository.