摘要
为了解决球栅阵列(BGA)焊点气孔缺陷的在线检测问题,开发了一种特殊的结构光检测技术。同时用环形红光和十字形绿光发射二极管(LED)照明BGA芯片,通过环形和十字形的中心,布置一个带有远心透镜的电荷耦合器(CCD)摄像机记录小球上红色圆环和绿色十字的图像。根据图像半径方向上的纹理频谱分布特征,引入人工神经网络(ANN)算法对BGA芯片的气孔缺陷进行检测。采用真实的BGA焊点进行实验,结果验证了方法的精度和可行性。
A novel structural light technique was developed for ball grid array (BGA) void on-line inspection.A light emitting diode (LED) ring light and a LED cross light were used to illuminate the BGA package with a charge-coupled device (CCD) to capture the image through the center hole of the ring light and the cross light.By extracting the spectral energy distribution features as a function of radius from the center of the spectrum using artificial neural network (ANN),we evaluated these techniques using actual BGA bumps.As a result,the correct rate of judgment reached 96.9%.which clearly showed that our method could be useful in the practice.
出处
《微型电脑应用》
2008年第10期32-33,26,共3页
Microcomputer Applications