PCB视觉检测系统的研究
|
摘要
在现代电子设备中,印刷电路板(PCB)发挥着越来越重要的作用,其质量的好坏在一定程度上决定了电子产品的性能。因此,对印刷电路板进行缺陷检测在工业生产中具有极其重要的应用价值。为了实现PCB生产过程中的缺陷检测自动化,本文应用虚拟仪器及其机器视觉技术,开发了一种PCB视觉检测系统。
论文从图像处理和模式识别的角度出发,研究了不同PCB缺陷的特点,在总结前人所用的PCB缺陷识别方法的基础上,综合各自优点,提出了基于图像识别的PCB检测方法。为了提高印刷电路板数字图像的信噪比,采用了直方图均衡化、中值滤波等预处理算法优化图像质量;在电路板缺陷识别与检测方面,对元件丢失和元件极性反装,提出模板匹配算法,实现了印刷电路板图像与模板图像之间的快速对准与参考比较;对焊点缺陷,提出应用数学形态学处理方法提取焊点的特征,然后根据特征判别是否为缺陷焊点;对PCB短路、断路的缺陷,提出基于边缘检测法来识别方法。
论文应用NI公司的虚拟仪器软件LabVIEW及图像处理工具包IMAQ Vision进行PCB视觉检测系统开发,实现了图像采集,光板检测,元件匹配检测,表面焊点检测,检测结果分析等功能。该系统操作简单,界面友好,可靠性高。
Nowadays, Printed Circuit Board (PCB) plays an important role in electronic devices. The quality of electronic productions somewhat depends on that of Printed Circuit Board. Due to the fatigue, subjectivity and speed requirement, manual inspection is ineffective to inspect every complex board. To realize defect inspection autom atically, some vision detection system for the PCB by Virtual Instrument and computer vision are studied in this article.
This paper proposes a new method for automatic component detection by computer. Based on digital image processing, we have studied different features of the PCB defection . In order to identify the disfiguration, we colligate advantages of some other Image identify methods and develop a way belonged to this thesis which based on image recognition technology. Some image preprocesses algorithms such as Histogram equalization, Median Filter are used for enhancing the SNR (Signal Noise Ratio) of the PCB image. In the procedure of defect detection and identification, a series of digital image processing algorithms are adopted to detect the lost and polarity fault of components by the way of Template Matching Algorithm. This process achieve the fast alignment and reference comparison between the PCB image with the template image. In a case of solder inspection, Mathematical Morphology is used to extract features of solder joints and judge whether there is solder joint or not. At last,to detect the fault of short Circuit and open circuit, Edge detecting Algorithm is used.
In this paper,the software development adopts the NI Company's Virtual Instrument Labview and vision took it IMAQ Vision. The software realizes the function of complete detecting, image collection, image pre-process, the fault inspection of bared PCB, component matching inspection and solder joint inspection. It has the merit of easily operation, friendly interface, high reliability and so on.
论文从图像处理和模式识别的角度出发,研究了不同PCB缺陷的特点,在总结前人所用的PCB缺陷识别方法的基础上,综合各自优点,提出了基于图像识别的PCB检测方法。为了提高印刷电路板数字图像的信噪比,采用了直方图均衡化、中值滤波等预处理算法优化图像质量;在电路板缺陷识别与检测方面,对元件丢失和元件极性反装,提出模板匹配算法,实现了印刷电路板图像与模板图像之间的快速对准与参考比较;对焊点缺陷,提出应用数学形态学处理方法提取焊点的特征,然后根据特征判别是否为缺陷焊点;对PCB短路、断路的缺陷,提出基于边缘检测法来识别方法。
论文应用NI公司的虚拟仪器软件LabVIEW及图像处理工具包IMAQ Vision进行PCB视觉检测系统开发,实现了图像采集,光板检测,元件匹配检测,表面焊点检测,检测结果分析等功能。该系统操作简单,界面友好,可靠性高。
Nowadays, Printed Circuit Board (PCB) plays an important role in electronic devices. The quality of electronic productions somewhat depends on that of Printed Circuit Board. Due to the fatigue, subjectivity and speed requirement, manual inspection is ineffective to inspect every complex board. To realize defect inspection autom atically, some vision detection system for the PCB by Virtual Instrument and computer vision are studied in this article.
This paper proposes a new method for automatic component detection by computer. Based on digital image processing, we have studied different features of the PCB defection . In order to identify the disfiguration, we colligate advantages of some other Image identify methods and develop a way belonged to this thesis which based on image recognition technology. Some image preprocesses algorithms such as Histogram equalization, Median Filter are used for enhancing the SNR (Signal Noise Ratio) of the PCB image. In the procedure of defect detection and identification, a series of digital image processing algorithms are adopted to detect the lost and polarity fault of components by the way of Template Matching Algorithm. This process achieve the fast alignment and reference comparison between the PCB image with the template image. In a case of solder inspection, Mathematical Morphology is used to extract features of solder joints and judge whether there is solder joint or not. At last,to detect the fault of short Circuit and open circuit, Edge detecting Algorithm is used.
In this paper,the software development adopts the NI Company's Virtual Instrument Labview and vision took it IMAQ Vision. The software realizes the function of complete detecting, image collection, image pre-process, the fault inspection of bared PCB, component matching inspection and solder joint inspection. It has the merit of easily operation, friendly interface, high reliability and so on.
引文
[1]何钰.汽车电子:半导体产业发展的助推力.电子产品世界,2005.10(19):117-119
[2]邦伟良.汽车电子在冉冉升起.电子产品世界,2005(10A):26-26
[3]罗兵,章云.SMT产品质量AOI快速检测方案.电子质量,2007(02):8-10
[4]彭占勇.计算机线路板质量检测技术.计算机与数字工程,2005(8):80-81
[5]翁寿松.SMT设各与PCB检测设备的发展动态.电子工业专用设备,2002(2):63-68
[6]范小涛,夏雨人.一种墓于图像处理的PCB检测算法研究.计算机工程与应用,2004(13):91-92
[7]鲜飞.SMT测试技术综述.电子测试,2007(Z1)
[8]J.A.L.Mahon.20 Years ofSMTautomatedopficalinspecfion.Proc.of Eos Conf.on industrial imagine and machine vision,Berlin,Germany:Springer-Verlag,2005:109-113
[9]倪兆明.高密度表面封装.现代表面贴装资,2002(3):63-64
[10]翁寿松.SMT设各与PCB检测设备的发展动态.电子工业专用设备,2002(2):63-68
[11]夏庆观,路红,赵茜,王东霞.基IMAQ的零件图像特征提取和识别.机械设计与制造,2005.9(9):132-133
[12]S.Mal lat.Multifrequency Channel Decopositions of images and Wavelet.IEEE Trans.on ASSP,Vlo37,No.12,2091-2110,1989
[13]柏长冰,齐春,宋福民,杨莹.基于“虚拟匹配”的PCB矩形Mark视觉快速检测.电子工业专用设备,2005(4):20-24
[14]Lam L,Whan S.Thinning methodologies-A comprehensive survey.Transactions on Pattern Analysis and Machine intelligence,1992,14(9)
[15]何伟,李薇,张玲.基于计算机图像处理的电路印刷板缺陷检测.计算机测量与控制,2007.15(10):1295-1297
[16]谢佳豪.基于计算机视觉的印刷电路板自动检测系统的研究:[硕士学位论文].西安:西安理工大学.2004
[17]段海森,徐杜.机器视觉在PC B缺陷检测中的应用.电子工业专用设备,2007.8(151):33-35
[18]IMAQ Vision Concepts Manual.National Instruments Corporation.2002-2003.
[19]陈虎,周朝辉,王守尊.基于数学形态学的图像去噪方法研.工程图学学报,2004(02):116-119
[20]李宇华.虚拟仪器开发平台LabVIEW介绍.计算机自动测量与控制,1996(05):134-137
[21]裴忠发,赵敬斌,罗志增.基于LabVIEW的机器视觉实现.机电工程,2002(04):59-62
[22]路林吉,饶家明.虚拟仪器概论.电子技术,2000(01):44-45
[23]刘阳.虚拟仪器的现状及发展趋势.电子技术应用,1998(4):3-5.
[24]程虎.虚拟仪器的现状和发展趋势.现代科学仪器,1999(4):6-9
[25]林正盛.虚拟仪器技术及其应用.微型机与应用,1997(8):8-10
[26]陕西海泰电子有限责任公司.虚拟仪器产品与技术,2002
[27]李亚奎,李亚伟.基于分形理论的高效机器视觉检测系统.基于计算机的测试测量和自动化应用方案,2005:17-19
[28]J.Y.-Z Wang.LabView in engineering laboratory courses.Frontiers in Education,2003(02):5-8
[29]王健.代替人眼的机器视觉系统.机电信息,2000(6):3-5
[30]李春森.基于LabVIEW和IMAQ的LCD机器视觉精确检测系统.电子技术,2003(3):52-56
[31]L.Chugani,Mahesh,R.Samant Abhag,Michael Cerra.LabVIEW Signal Processing.Prentice Hall PTR,1998:133-168
[32]U.S.National Instruments Inc.INSTRUPEDIA~(TM)(仪器百科全书~(TM)).U.S.A,2000
[33]李华,王非,张哲.基于LabVIEW平台的光度测量与校正系统.计算机测量与控制,2004,12(4):316-321
[34]谭二,盛诩智.复杂彩色图像的二值分割及初步处理.华中科技大学学报,2001,12(12):64-66
[35]章毓晋.图像处理和分析基础.第l版.北京:高等教育出版社,2002
[36]Kenneth R Castlcman.数字图象处理.北京:电子工业出版社,2002
[37]谷口庆治.数字图像处理.北京:科学出版社,2002
[38]邓继忠,张泰岭.基于虚拟仪器的图像处理与分析系统.农机化研究,2001,(3):30-32
[39]H.G:.Senel.R.A.Peters and B.Dawant.Topological Median Filters.IEEE TRANSACTIONS ON IMACE.PROCESSING:.2001.Vol.11.No.1209-1227
[40]张小洪,杨丹,刘亚威.基于Canny算子的改进型边缘检测算法.计算机工程与应用,2003.29:113-115
[41]H.Y.Francis,Chan,F.K.Lam,Hui Zhu.Adaptive Threshold by Variational Method.IEEE Trans.Image Processing.1998,(3)3:89-96
[42]李刚,韩建国.PCB图像检测中阈值化分割的研究.北京化工大学学报,2002.4(29):72-75
[43]Kittler J,Illingworth J.On Threshold Selection Using Clustering Criteria.IEEE Trans.1985,MC-15:652-655
[44]张杨,李凯.非均匀光照条件下PCB图像的一种阈值分割方法[J].北京服装学院学报,2006.6(2):1-6
[45]J.Mreinhardt,W.E.Higgins.Comparison between the morphology skeletonand morphology shape decomposition.IEEE Trans.Pattern Anal.Machineintel 1.1996(18):951-957
[46]夏庆观,路红邹.基于MAQ的模式识别在零件检测中的应用.仪器仪表与检测,2005(09):73-75
[47]胡仲波.基于虚拟仪器技术的PCB视觉检测系统:[硕士学位论文].南京:南京理工大学,2006
[48]Z.les.The Processing method as a set of the Image transformation in shape understanding,2001.No.2(25):223-234
[49]H.Moon.R.Chellappa.A.Rosenfeld.Optimal Edge-Based Detection.TANSACTIONS ON IMAGE PROCESSING,2002.No.11(11):1209-1227
[50]郑伟:图像法检测印刷电路板缺陷:[硕士学位论文].西安:西安理工大学,2002
[51]张道勇.基于线阵CCD的大幅面扫描技术研究:[硕士学位论文].广州:广东工业大学,2005
[52]陈向伟,王龙山,刘庆民.微小塑料零件图像边缘检测方法的研究.轻工机械,2004(01):88-90
[53]National Instruments,IMAQ User Manual Concepts,1999:11-4-11-12
[2]邦伟良.汽车电子在冉冉升起.电子产品世界,2005(10A):26-26
[3]罗兵,章云.SMT产品质量AOI快速检测方案.电子质量,2007(02):8-10
[4]彭占勇.计算机线路板质量检测技术.计算机与数字工程,2005(8):80-81
[5]翁寿松.SMT设各与PCB检测设备的发展动态.电子工业专用设备,2002(2):63-68
[6]范小涛,夏雨人.一种墓于图像处理的PCB检测算法研究.计算机工程与应用,2004(13):91-92
[7]鲜飞.SMT测试技术综述.电子测试,2007(Z1)
[8]J.A.L.Mahon.20 Years ofSMTautomatedopficalinspecfion.Proc.of Eos Conf.on industrial imagine and machine vision,Berlin,Germany:Springer-Verlag,2005:109-113
[9]倪兆明.高密度表面封装.现代表面贴装资,2002(3):63-64
[10]翁寿松.SMT设各与PCB检测设备的发展动态.电子工业专用设备,2002(2):63-68
[11]夏庆观,路红,赵茜,王东霞.基IMAQ的零件图像特征提取和识别.机械设计与制造,2005.9(9):132-133
[12]S.Mal lat.Multifrequency Channel Decopositions of images and Wavelet.IEEE Trans.on ASSP,Vlo37,No.12,2091-2110,1989
[13]柏长冰,齐春,宋福民,杨莹.基于“虚拟匹配”的PCB矩形Mark视觉快速检测.电子工业专用设备,2005(4):20-24
[14]Lam L,Whan S.Thinning methodologies-A comprehensive survey.Transactions on Pattern Analysis and Machine intelligence,1992,14(9)
[15]何伟,李薇,张玲.基于计算机图像处理的电路印刷板缺陷检测.计算机测量与控制,2007.15(10):1295-1297
[16]谢佳豪.基于计算机视觉的印刷电路板自动检测系统的研究:[硕士学位论文].西安:西安理工大学.2004
[17]段海森,徐杜.机器视觉在PC B缺陷检测中的应用.电子工业专用设备,2007.8(151):33-35
[18]IMAQ Vision Concepts Manual.National Instruments Corporation.2002-2003.
[19]陈虎,周朝辉,王守尊.基于数学形态学的图像去噪方法研.工程图学学报,2004(02):116-119
[20]李宇华.虚拟仪器开发平台LabVIEW介绍.计算机自动测量与控制,1996(05):134-137
[21]裴忠发,赵敬斌,罗志增.基于LabVIEW的机器视觉实现.机电工程,2002(04):59-62
[22]路林吉,饶家明.虚拟仪器概论.电子技术,2000(01):44-45
[23]刘阳.虚拟仪器的现状及发展趋势.电子技术应用,1998(4):3-5.
[24]程虎.虚拟仪器的现状和发展趋势.现代科学仪器,1999(4):6-9
[25]林正盛.虚拟仪器技术及其应用.微型机与应用,1997(8):8-10
[26]陕西海泰电子有限责任公司.虚拟仪器产品与技术,2002
[27]李亚奎,李亚伟.基于分形理论的高效机器视觉检测系统.基于计算机的测试测量和自动化应用方案,2005:17-19
[28]J.Y.-Z Wang.LabView in engineering laboratory courses.Frontiers in Education,2003(02):5-8
[29]王健.代替人眼的机器视觉系统.机电信息,2000(6):3-5
[30]李春森.基于LabVIEW和IMAQ的LCD机器视觉精确检测系统.电子技术,2003(3):52-56
[31]L.Chugani,Mahesh,R.Samant Abhag,Michael Cerra.LabVIEW Signal Processing.Prentice Hall PTR,1998:133-168
[32]U.S.National Instruments Inc.INSTRUPEDIA~(TM)(仪器百科全书~(TM)).U.S.A,2000
[33]李华,王非,张哲.基于LabVIEW平台的光度测量与校正系统.计算机测量与控制,2004,12(4):316-321
[34]谭二,盛诩智.复杂彩色图像的二值分割及初步处理.华中科技大学学报,2001,12(12):64-66
[35]章毓晋.图像处理和分析基础.第l版.北京:高等教育出版社,2002
[36]Kenneth R Castlcman.数字图象处理.北京:电子工业出版社,2002
[37]谷口庆治.数字图像处理.北京:科学出版社,2002
[38]邓继忠,张泰岭.基于虚拟仪器的图像处理与分析系统.农机化研究,2001,(3):30-32
[39]H.G:.Senel.R.A.Peters and B.Dawant.Topological Median Filters.IEEE TRANSACTIONS ON IMACE.PROCESSING:.2001.Vol.11.No.1209-1227
[40]张小洪,杨丹,刘亚威.基于Canny算子的改进型边缘检测算法.计算机工程与应用,2003.29:113-115
[41]H.Y.Francis,Chan,F.K.Lam,Hui Zhu.Adaptive Threshold by Variational Method.IEEE Trans.Image Processing.1998,(3)3:89-96
[42]李刚,韩建国.PCB图像检测中阈值化分割的研究.北京化工大学学报,2002.4(29):72-75
[43]Kittler J,Illingworth J.On Threshold Selection Using Clustering Criteria.IEEE Trans.1985,MC-15:652-655
[44]张杨,李凯.非均匀光照条件下PCB图像的一种阈值分割方法[J].北京服装学院学报,2006.6(2):1-6
[45]J.Mreinhardt,W.E.Higgins.Comparison between the morphology skeletonand morphology shape decomposition.IEEE Trans.Pattern Anal.Machineintel 1.1996(18):951-957
[46]夏庆观,路红邹.基于MAQ的模式识别在零件检测中的应用.仪器仪表与检测,2005(09):73-75
[47]胡仲波.基于虚拟仪器技术的PCB视觉检测系统:[硕士学位论文].南京:南京理工大学,2006
[48]Z.les.The Processing method as a set of the Image transformation in shape understanding,2001.No.2(25):223-234
[49]H.Moon.R.Chellappa.A.Rosenfeld.Optimal Edge-Based Detection.TANSACTIONS ON IMAGE PROCESSING,2002.No.11(11):1209-1227
[50]郑伟:图像法检测印刷电路板缺陷:[硕士学位论文].西安:西安理工大学,2002
[51]张道勇.基于线阵CCD的大幅面扫描技术研究:[硕士学位论文].广州:广东工业大学,2005
[52]陈向伟,王龙山,刘庆民.微小塑料零件图像边缘检测方法的研究.轻工机械,2004(01):88-90
[53]National Instruments,IMAQ User Manual Concepts,1999:11-4-11-12
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |