基于小波变换和SSVME的PCB产品视觉检测中缺陷分类研究

被引:2
作者
任斌 [1 ]
程良伦 [2 ]
机构
[1] 东莞理工学院电子工程学院
[2] 广东工业大学自动化学院
基金
广东省自然科学基金;
关键词
PCB; 缺陷检测; 小波变换和光滑支持向量机集成算法; 分类;
D O I
暂无
中图分类号
TP391.41 []; TN41 [印刷电路];
学科分类号
080203 ; 080903 ; 1401 ;
摘要
针对PCB产品视觉检测中图像缺陷细微、形状复杂、特征难于提取、易受噪声影响的问题,提出基于小波变换和光滑支持向量机集成SSVME(Smooth Support Vector Machine Ensemble)的多分类方法,有效解决了细微、复杂缺陷难以识别分类的问题。实验表明,该方法六类缺陷混合识别率达到95.26%,高于BP神经网络的最优识别率90.35%和基于区域方法的80.67%,而且训练和分类时间短。从理论和实验中验证了该方法的有效性,是PCB产品视觉检测领域中缺陷识别分类的新方法,具有重要的应用价值。
引用
收藏
页码:167 / 171
页数:5
相关论文
共 17 条
[1]  
Application of Neural Networks in Optical Inspection and Classification of Solder Joints in Surface Mount Technology. A.Giuseppe,B.Gioacchino,F.Girolamo. IEEE Transactions on Industrial Informatics . 2006
[2]  
A new machine vision system for SMD PCB auto-inspection. Peng D.B,Yan C.J,Chou C.C. The international symposium on measurement Techonology & Intelligent Instrument . 2001
[3]  
Printed Circuit Board Inspection Using Image Analysis. Horng-Hai Loh,Ming-Sing Lu. IEEE Transactions on Industry Applications . 1999
[4]  
Automated Visual Inspection of Surface Mount PCBs. Teoh E.K,Mital D.P,Lee B.W.etal. 16th Annual Conference of IEEE on IE CON . 1990
[5]  
Automated visual inspection: 1981 to 1987. RT Chin. Computer Vision . 1988
[6]   多项式光滑的支持向量回归机 [J].
任斌 ;
程良伦 .
控制理论与应用, 2011, 28 (02) :261-265
[7]  
Automated inspection of PCB components using a genetic algorithm template-matching approach[J] . A. J. Crispin,V. Rankov. &nbspThe International Journal of Advanced Manufacturing Technology . 2007 (3)
[8]  
Non-destructive reverseengineering of trace maps in multilayered PCBs. Longbotham H G,Ping Y,Kothari H N,et al. Autotestcon . 1995
[9]  
Polynomial smooth epsilon-support vectorregression based on Hermite interpolation. Ren Bin,,Cheng Lianglun. Journal of Computation-al Information Systems . 2010
[10]  
Study of PCB automatic op-tical inspection system based on mathematical morphology. Lin Lizong,Zhou Luoshan,Wan Jianding. 2009International Conference on Computer Technology and Development . 2009