Real-Time Welding Defect Detection with YOLOv8 on Raspberry Pi

Welding plays a vital role in industries such as manufacturing, automotive, and construction, where ensuring the strength and reliability of welded joints is essential for structural stability and safety. Defects like cracks, porosity, and incomplete fusion can weaken joints, increasing the risk of failures, raising costs, and posing safety concerns. Traditional inspection techniques, such as manual visual checks and nondestructive testing, often require significant time, financial investment, and skilled personnel, making them inefficient and prone to human error. This research introduces a real-time, cost-effective approach to welding defect detection by leveraging YOLOv8 and Raspberry Pi. The system is designed to capture high-resolution images of welds, analyze them using a deep learning model trained on a labeled dataset, and accurately classify defects Compared to traditional methods, this solution enhances defect detection efficiency while remaining affordable, particularly for small and medium-sized enterprises (SMEs) that may not have access to high-end quality control equipment. By integrating this automated system into industrial processes, companies can improve welding quality, reduce rework costs, and enhance workplace safety, ultimately contributing to more reliable and cost-efficient manufacturing practices.