摘要
文中针对印刷电路板(Printed Circuit Board,PCB)缺陷检测问题,提出了一种结合传统图像处理方法与深度学习技术的缺陷检测方法。首先,采用传统的图像预处理技术对PCB图像进行去噪、对比度增强和边缘检测等操作,为后续的缺陷检测奠定基础。随后,基于Faster R-CNN模型进行缺陷检测与分类。实验结果表明,该方法在不同缺陷类型的检测中表现出较高的准确性、精确率和召回率,不仅提高了PCB缺陷检测的效率,还为相关领域的应用提供了新的思路。
This paper proposes a defect detection method for printed circuit boards(PCBs)that combines traditional image processing methods with deep learning techniques.Firstly,traditional image preprocessing techniques are used to denoise,enhance contrast,and perform edge detection on PCB images,laying the foundation for subsequent defect detection.Subsequently,defect detection and classification were performed based on the Faster R-CNN model.The experimental results show that this method exhibits high accuracy,precision,and recall in detecting different types of defects.This method not only improves the efficiency of PCB defect detection,but also provides new ideas for applications in related fields.
作者
杭秋丽
HANG Qiuli(Inner Mongolia Electronic Information Vocational and Technical College,Hohhot 010070,China)
出处
《移动信息》
2025年第8期282-284,共3页
Mobile Information
