基于机器视觉的高密度电路板缺陷检测系统  

Defects Inspection System of HID PCB Based on Machine Vision

作　　者：熊光洁[1,2] 马树元[2] 聂学俊[1,2] 武思远 汤晓华[1] 

机构地区：[1]北京工商大学机械工程学院,北京100048 [2]北京理工大学机械与车辆工程学院,北京100081 [3]北京星河泰视特科技有限公司,北京100094

出　　处：《计算机测量与控制》2011年第8期1824-1826,共3页Computer Measurement &Control

基　　金：北京市科委创新项目(K20090217L)

摘　　要：为减少高密度电路板的缺陷误报率,研究一种新型自动光学检测系统(AOI);系统采用自行研制的多色LED照明系统,利用机器视觉获取被测PCB的图像,通过图像处理软件系统快速准确地识别出各种缺陷;系统利用获取的彩色图像信息,根据各种缺陷的特征信息不同,采用OPENCV对各种缺陷的检测算法进行改进,使得系统性能有很大改进;对30块同类HDI型PCB的36300个检测点进行测试,测试结果证明,系统PCB缺陷的检出率高达99.87%,误报率只有0.32%。An improved automated optical inspection system （AOI） is researched to decrease defects false alarm rate of HDI PCB. With a new multicolor LED illuminator, the system can capture the tested PCB image by using machine vision, and identify the various defects quickly and accurately through image processing software systems in this paper. The performance system of this AOI has been greatly improved by using improved hardware system and algorithms which was programmed on OPENCV platform by using the colorful information of captured images. The 36300 testing points of 30 HDI PCB are detected, and the results of this experiment prove that the PCB defect detection rate of the AOI inspection system is improved to 99.87% and the false alarm rate of defects down to 0. 32%.

关 键 词：机器视觉 高密度电路板 多色LED照明 缺陷检测 OPENCV 

分 类 号：TP216[自动化与计算机技术—检测技术与自动化装置]