ABSTRACT:

The "Steel Surface Defect Detection" project leverages YOLOv8 for real-time, high-accuracy detection of six steel defect types: crazing, inclusion, patches, pitted surface, rolled-in scale, and scratches. The system is developed using Python for backend operations, with Flask as the framework, while HTML, CSS, and JavaScript are used for a responsive frontend interface. By harnessing YOLOv8's object detection capabilities, the system identifies and classifies surface defects on steel, facilitating quality control and defect management in industrial processes. The integration of Flask enables a seamless user experience, allowing users to interact with the system to upload images and visualize defect detection results. This solution aims to optimize production quality and reduce waste by ensuring defect-free products reach the market.

Keywords: Steel defect detection, YOLOv8, Flask framework, real-time classification, surface quality control.