摘要
印刷电路板自动视觉检测方法一般只应用了PCB图像的灰度信息,且关注于导路的几何缺陷。本文提出了用于PCB上金手指表面检测的两个熵测度,以检测颜色异常与结构变化。实验结果表明该方法的检测效果较好,可实现金手指在线检测。
Generally, auto-visual inspection of printed circuit board (PCB) utilizes exclusively information on gray-scale of PCB images, and focuses on geometric defects of the transmitting circuit. Two entropy measures were presented and applied to the surface of gold finger on PCB to test color abnormity and structural transformation. The experimental result proves the feasibility of the on-line detecting of gold finger.
出处
《机床与液压》
北大核心
2008年第1期117-119,123,共4页
Machine Tool & Hydraulics
基金
广东省自然科学基金项目(7007362)
国家自然科学基金项目(50475044)
广州市黄埔区科技计划项目(0713)
关键词
熵测度
印刷电路板
缺陷检测
机器视觉
Entropy measure
Printed circuit board
Defect-detecting
Machinery vision