期刊文献+

多引脚类贴装器件的炉前检测算法 被引量:3

Inspection algorithm of before-furnace for multi-leads surface mounted devices
下载PDF
导出
摘要 为了适应多引脚类贴装器件炉前检测的实时性要求,提出了一种针对引脚特征的检测算法。首先分析了多引脚类器件在结构光下的特征;然后设计了一种差分算子,提取引脚的水平和垂直差分,接着对差分值作积分投影,从而避免了设置边缘阈值的困难,在此基础上,利用最大邻域梯度法和滑动滤波窗算法定位引脚位置;最后提取引脚的面积特征并引入模式匹配方法。实验结果表明:本文算法可有效的检测多引脚类器件缺件、错件、偏移、歪斜和翘脚等缺陷,整体准确率为97.7%,检测单个器件的时间约为28ms,能满足在线检测系统高准确率和实时性的要求。 According to the real-time inspection requirements of before-furnace for the multi-leads surface mounted devices, an inspection algorithm based on the lead features was presented. Firstly, the features of the devices under the structure light sources were analyzed. Secondly, the horizontal and vertical differences were extracted with the difference arithmetic operator. Then the differences were projected in order to avoid the difficulty of right thresholding. Based on the integral projections of the differences, the horizontal and vertical borders of the leads were obtained by the maximum neighbor gradient and the sliding filter window algorithm, respectively. Finally, the area features in the leads were ex- tracted and the pattem matching method was applied. The experiment results show that the proposed method can identify the defects such as missing devices, wrong devices, shifts, skews and leads lift effectively, the total success rate was 97.7%, and the time consuming for single device was about 28 ms. Therefore it can satisfy the requests of high accuracy and real-time character for on-line inspection system.
出处 《电子测量与仪器学报》 CSCD 2011年第11期998-1005,共8页 Journal of Electronic Measurement and Instrumentation
基金 国家杰出青年科学基金(50825504)
关键词 自动光学检测 贴装器件 多引脚 积分投影 automatic optical inspection (AOI) mounted devices multi-leads integral projection
  • 相关文献

参考文献4

二级参考文献50

  • 1胡登鹏,李宏伟,郭英,樊昌周.基于决策树的便携式心音诊断仪[J].仪器仪表学报,2006,27(z2):1511-1512. 被引量:2
  • 2邵晓芳,孙即祥,王亮亮,田素芬.改进的Canny算法[J].电光与控制,2006,13(6):53-55. 被引量:6
  • 3李志敏,林越伟,黄俊,张凤阳,万睿,张晶,黄凡.PCB走线检测的预处理算法[J].光学精密工程,2007,15(2):272-276. 被引量:13
  • 4俞龙江,杨英,孙圣和.基于最小二乘拟合法的焊点形状检测[J].仪器仪表学报,2007,28(7):1255-1258. 被引量:16
  • 5KIM J H, CHO H S, KIM S. Pattern classification of solder joint images using a correlation neural network [J]. Eng. Applic. Intell, 1996,9(6):655-669.
  • 6KIM T H, CHO T H, MOON Y S, et al. Visual inspection system for the classification of solder joints[J]. Pattern Recognition, 1999,32 ( 4 ) : 565-575.
  • 7ACCIANI G, BRUNETTI G, CHIARANTONI E, et al. An automatic method to detect missing components in manufactured products[ C]. Proceedings of International Joint Conference on Neural Networks, 2005:2324-2329.
  • 8MICHAEL E Z, STEFANOS K G, GEORGE A R. A bayesian framework for multilead SMD post-placement quality inspection [ C ]. IEEE Transactions on systems, man, and cybernetics-part B : cybernetics, 2004,34 ( 1 ) : 440-453.
  • 9WU H H, ZHANG X M, KUANG Y C. A real-time machine vision system for solder paste inspection [ C ]. IEEE/-ASME international conference on advanced intelligent mechatronics, 2008:205-210.
  • 10ROBERT R E, STOUGH J P, STEPHEN M C, et al. Statistical appearance model based on intensity quantile histograms[ C ]. IEEE International Symposium on Biomedical Imaging: From Nano to Micro Proceedings, 2006:422-425.

共引文献40

同被引文献20

引证文献3

二级引证文献20

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部