基于OpenCV的Micro-PCB点胶表面缺陷检测系统设计

Design of OpenCV-based Micro-PCB dispensing surface defect detection system

针对Micro-PCB体型小、板片薄不易做表面缺陷检测的弊端,构建了一套基于OpenCV视觉库,可对点胶表面进行缺陷检测的机器视觉系统。运用图像处理中的形态学算法,对金面表面实现0.2mm2以上的多点状、膜破渗金和抗镀无法描述、黑色圆形点状等缺陷的检测。测试结果表明,该系统检测精度较高且运行稳定。

Aiming at the drawbacks of Micro-PCB′s small size and thin boards that are not easy to do surface defect detection,a machine vision system based on OpenCV vision library that can detect defects on dispensing surfaces is constructed.Using morphological algorithms in image processing,the surface of the gold surface to achieve more than 0.2 mm2 multi-dot,film broken gold seepage and anti-plating can not be described,black round dots and other defects in the detection.Test results show that the system has high detection accuracy and stable operation.