AI-Based Part Identification and Classification Using YOLOv8 and Flask

The rapid advancement of artificial intelligence (AI) and computer vision has enabled the development of systems capable of identifying and classifying parts of complex items with high accuracy. This paper focuses on building an AI-based system to detect and classify parts of an item, such as components of a machine, car parts, or sections of a product. The system leverages state-of-the-art object detection models, specifically YOLOv8 (You Only Look Once), to achieve real-time and accurate part identification. The paper involves several key steps: data collection and annotation, model selection and training, and deployment of the trained model for inference. A custom dataset is created, containing images of the item with annotated parts. The YOLOv8 model is fine-tuned on this dataset to detect and classify the parts. The trained model is then integrated into a Flask-based web application, allowing users to upload images and receive predictions in real-time. The system demonstrates high accuracy and efficiency, making it suitable for industrial applications, quality control, and educational purposes. By automating the process of part identification, this paper reduces human error, saves time, and enhances productivity. The results highlight the potential of AI in transforming traditional workflows and enabling smarter decision-making in various domains.