Research on Portable Surface Defect Detection System Based on Channel Pruning Technology

This study addresses the prevalent issue of surface defect detection in industrial manufacturing, particularly in high-precision domains such as semiconductor and automotive parts production, where traditional manual inspection methods fall short in terms of efficiency and accuracy. In response to these challenges, this paper introduces a portable surface defect detection system based on channel pruning technology. The research incorporates an enhanced YOLOv8 object detection algorithm and global channel pruning techniques to optimize algorithm performance, thereby enhancing the system's operational efficiency and detection accuracy under resource-constrained conditions. Key findings include the system's ability to effectively identify and classify various surface defects, significantly accelerating detection speed and accuracy. Additionally, the system's design facilitates seamless integration with existing manufacturing systems and offers considerable scalability.