基于微焦点X射线的SMT焊点缺陷检测仿真  

SMT Solder Joint Defect Detection Simulation Based on Microfocus X-ray

作　　者：王付军[1] 刘兰英[1] WANG Fu-jun;LIU Lan-ying(Dongchang College,Liaocheng University,Liaocheng Shandong 252000,China)

机构地区：[1]聊城大学东昌学院,山东聊城252000

出　　处：《计算机仿真》2020年第9期428-431,共4页Computer Simulation

基　　金：山东省高校科技计划项目(J17KB145)。

摘　　要：针对传统的X射线的SMT焊点缺陷检测方法存在着检测所需时间较长、精确率较低等问题,基于微焦点X射线的SMT焊点缺陷检测方法,将基准图像的灰度均值作为初始背景,采集到新的SMT焊点图像后,对其背景进行更新,获取整体灰度校正后的图像。对各个子块进行直方图均匀化,获取部分对比度增强图像;对其做Gabor模板卷积,得到边缘图像。通过对增强图像的像素点进行加权,对马氏距离的协和方差重新估计,调整目标区域像素点权重,获取某个灰度值样本和全部样本加权的马氏距离,通过马氏距离增强目标区域的边缘特征;运用连通区域标记法,提取SMT焊点缺陷位置,对缺陷进行检测。实验结果表明,所提方法可检测焊点缺陷明显高于当前广泛应用的焊点缺陷检测方法,尤其在检测效率上有明显提升。The traditional method to detect X-ray SMT solder joint defect often leads to long detection time and low accuracy.Therefore,this paper focuses on a method to detect SMT solder joint defect based on microfocus X-ray.The grayscale mean value of reference image was used as the initial background.After new SMT solder joint image was acquired,the background was updated to get the image after overall grayscale correction.Then,the histogram homog⁃enization was applied to each sub-block,so that some images with contrast enhancement were obtained.Gabor tem⁃plate was carried out,and then the edge image was obtained.By weighting the pixels of the enhanced image,the co⁃variance of Mahalanobis distance was estimated again and the weight of pixel point in target area was adjusted.After that,Mahalanobis from a gray value sample to the weights of all samples.Moreover,the edge feature of target area was enhanced by Mahalanobis distance.Finally,the connected components labeling method was used to extract the position of SMT solder joint defect and detect the defect.Simulation results prove that the proposed method is signifi⁃cantly higher than the current solder joint defect detection methods,especially in the detection efficiency.

关 键 词：直方图均匀化 协方差矩阵 马氏距离 连通区域标记法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]