Study on Automatic Detection of PCB Defects Based on Machine Vision
基于机器视觉PCB裸板缺陷自动检测研究

摘要

随着科技的不断进步，PCB生产厂商越来越多，产品更新换代也非常快，但在某些情况下会产生断料、塑形等缺陷。PCB裸板的加工是基于人工检测和机器视觉技术来进行缺陷识别的一种方法，但是人工眼中存在许多不确定因素，而且易受到光源影响，导致无法获取完整图像。针对PCB裸板的特点和检验方法,本文以机器视觉技术为基础,提取PCB裸板的缺陷图片，对采集到的样本图像进行处理，基于机器视觉PCB裸板缺陷自动检测进行研究。

Abstract

With the continuous progress of science and technology, more and more PCB manufacturers, product upgrading is also very fast, but in some cases will produce broken material cut, shaping and other defects. The processing of PCB bare plate is a method of defect identification based on manual detection and machine vision technology, but there are many uncertainties in the artificial eye, and they are susceptible to the light source, leading to the inability to obtain a complete image. According to the characteristics and inspection methods of PCB bare plate, this paper extracts the defect pictures of PCB bare plate, processes the collected sample images, and studies them based on the automatic detection of machine vision PCB bare plate defects.