Printed circuit boards(PCB)are essential components for connecting and fixing electronic components.They are widely integrated in modern electronic devices,including computers,mobile phones,televisions,digital cameras,and various devices.With the rapid development of integrated circuit and semiconductor technology,the manufacturing of PCBs has become increasingly sophisticated and miniaturized,which has exacerbated the challenges of defect detection.Traditional algorithms are inefficient and have limited accuracy,which cannot meet the needs of actual production applications.In contrast,artificial intelligence-based PCB defect detection algorithms are expected to achieve higher accuracy and efficiency,thanks to their ability to quickly and accurately identify new defect types.This paper comprehensively analyzes PCB defect detection algorithms based on traditional machine learning and deep learning,introduces 8 publicly available PCB defect datasets,discusses traditional machine learning-based and deep learning-based defect detection methods in detail,compares their algorithm performance,advantages,and limitations,summarizes the challenges currently faced in the field of PCB defect detection,and looks forward to the research trends of PCB defect detection in the future.
维普
万方数据