基于机器视觉的太阳能网版缺陷检测
|
摘要
为了解决传统太阳能网版缺陷检测法效率低、检测速度慢和准确率低的问题,本文提出一种硬件与软件相结合的基于机器视觉的太阳能网版检测方法。根据测量精度的要求,本文还设计了一台适合太阳能网版图像采集的移动平台,软件部分主要包括缺陷检测、栅线宽度测量。利用质心检测和直线拟合测量栅线的宽度,在此基础上,通过支持向量机(SVM)图像分类法检测太阳能网版缺陷,利用已经分类好的样本进行训练生成一个分类器。经过实验验证,缺陷检测的准确率超过95%,栅线宽度测量误差为1μm左右,证明该方法不仅具有检测成本低、可靠性高、检测效率高等特点,而且具有实用推广价值。
In order to solve the problem of low efficiency,low detection speed and low accuracy of traditional solar panel defect detection method,this paper presents a method for the detection of solar panel based on machine vision,which adopts the method of combining hardware and software.According to the requirement of measurement precision,a mobile platform is designed which is suitable for shooting solar panel images,where the software part includes defect detection and measurement of grid line width.The width of the grid line is measured by centroid detecting and fitting a straight line.Based on this,the defect of the solar panel is detected by the support vector machine(SVM)image classification system,and a classifier is generated by using classified samples.Experimental verification shows that the defect detection accuracy is higher than 95%,and the grid line width measurement error is 1μm.It is proved that the method not only is of low cost,high reliability and high detection efficiency,but also has wide practical value.
In order to solve the problem of low efficiency,low detection speed and low accuracy of traditional solar panel defect detection method,this paper presents a method for the detection of solar panel based on machine vision,which adopts the method of combining hardware and software.According to the requirement of measurement precision,a mobile platform is designed which is suitable for shooting solar panel images,where the software part includes defect detection and measurement of grid line width.The width of the grid line is measured by centroid detecting and fitting a straight line.Based on this,the defect of the solar panel is detected by the support vector machine(SVM)image classification system,and a classifier is generated by using classified samples.Experimental verification shows that the defect detection accuracy is higher than 95%,and the grid line width measurement error is 1μm.It is proved that the method not only is of low cost,high reliability and high detection efficiency,but also has wide practical value.
引文
[1]王宪保,李洁,姚明海,等.基于深度学习的太阳能电池片表面缺陷检测方法[J].模式识别与人工智能,2014,27(6):517-523.DOI:10.16451/j.cnki.issn1003-6059.2014.06.006.
[2]JIA Hongbin,MURPHEY Y L,SHI Jianjun.An intelligent real-time vision system for surface defect detection[C]//Proceedings of the 17th International Conference on Pattern Recognition.Los Alamitos,CA:IEEE Computer Society Press,2004:239-242.DOI:10.1109/ICPR.2004.1334512.
[3]PENG Xiangqian,CHEN Youping,YU Wenyong,et al.An online defects inspection method for float glass fabrication based on machine vision[J].International Journal of Advanced Manufacturing Technology,2008,39(11/12):1180-1189.DOI:10.1007/s00170-007-1302-7.
[4]胡浩,梁晋,唐正宗,等.数字图像相关法测量金属薄板焊接的全场变形[J].光学精密工程,2012,20(7):1636-1644.DOI:10.3788/OPE.20122007.1636.
[5]卜雄洙,李桂娟,杨波,等.中心偏移的全景环形图像快速展开[J].光学精密工程,2012,20(9):2103-2109.DOI:10.3788/OPE.20122009.2103.
[6]JIAN Chuanxia,GAO Jian,AO Yinhui.Automatic surface defect detection for mobile phone screen glass based on machine vision[J].Applied Soft Computing,2017,52:348-358.DOI:10.1016/j.asoc.2016.10.030.
[7]LU Chijie,TSAI Duming.Independent component analysis-based defect detection in patterned liquid crystal display surfaces[J].Image and Vision Computing,2008,26(7):955-970.DOI:10.1016/j.imavis.2007.10.007.
[8]ZHANG Lei,GRIFT T E.A new approach to crop-row detection in corn[C]//American Society of Agricultural and Biological Engineers Annual International Meeting 2010.St.Joseph,MI:American Society of Agricultural and Biological Engineers,2010:3950-3964.DOI:10.13031/2013.29834.
[9]SUN T H,TSENG C C,CHEN M S.Electric contacts inspection using machine vision[J].Image and Vision Computing,2010,28(6):890-901.DOI:10.1016/j.imavis.2009.11.006.
[10]王震宇.基于机器视觉钢板表面缺陷检测技术研究[J].计算机与现代化,2013(7):130-134.DOI:10.3969/j.issn.1006-2475.2013.07.035.
[11]李超,孙俊.基于机器视觉方法的焊缝缺陷检测及分类算法[J].计算机工程与应用,2018,54(6):264-270.DOI:10.3778/j.issn.1002-8331.1609-0322.
[12]黄志鸿,毛建旭,王耀南,等.基于机器视觉的啤酒瓶口缺陷检测分类方法研究[J].电子测量与仪器学报,2016,30(6):873-879.DOI:10.13382/j.jemi.2016.06.006.
[13]杨洋.基于SVM的印刷品缺陷在线检测[D].武汉:华中科技大学,2012.
[14]孙智权,周奇,陈震,等.基于CMOS图像传感器的太阳能电池缺陷检测系统设计[J].仪表技术与传感器,2018(1):60-63.DOI:10.3969/j.issn.1002-1841.2018.01.015.
[15]刘焕军,王耀南,段峰.基于支撑向量机的空瓶智能检测方法[J].控制与决策,2005,20(12):1434-1437.DOI:10.13195/j.cd.2005.12.116.liuhj.026.
[16]贾云得.机器视觉[M].北京:科学出版社,2000:48-81.
[17]贺秋伟,王龙山,于忠党,等.基于图像处理和支持向量机的微型齿轮缺陷检测[J].吉林大学学报(工学版),2008,38(3):565-569.DOI:10.13229/j.cnki.jdxbgxb2008.03.008.
[18]KEERTHIS S,LIN C J.Asymptotic behaviors of support vector machines with Gaussian kernel[J].Neural Computation,2003,15(7):1667-1689.DOI:10.1162/089976603321891855.
[19]吴彰良,孙长库,刘洁.基于支持向量机的油封缺陷图像检测方法[J].光电工程,2012,39(3):40-45.DOI:10.3969/j.issn.1003-501X.2012.03.008.
[2]JIA Hongbin,MURPHEY Y L,SHI Jianjun.An intelligent real-time vision system for surface defect detection[C]//Proceedings of the 17th International Conference on Pattern Recognition.Los Alamitos,CA:IEEE Computer Society Press,2004:239-242.DOI:10.1109/ICPR.2004.1334512.
[3]PENG Xiangqian,CHEN Youping,YU Wenyong,et al.An online defects inspection method for float glass fabrication based on machine vision[J].International Journal of Advanced Manufacturing Technology,2008,39(11/12):1180-1189.DOI:10.1007/s00170-007-1302-7.
[4]胡浩,梁晋,唐正宗,等.数字图像相关法测量金属薄板焊接的全场变形[J].光学精密工程,2012,20(7):1636-1644.DOI:10.3788/OPE.20122007.1636.
[5]卜雄洙,李桂娟,杨波,等.中心偏移的全景环形图像快速展开[J].光学精密工程,2012,20(9):2103-2109.DOI:10.3788/OPE.20122009.2103.
[6]JIAN Chuanxia,GAO Jian,AO Yinhui.Automatic surface defect detection for mobile phone screen glass based on machine vision[J].Applied Soft Computing,2017,52:348-358.DOI:10.1016/j.asoc.2016.10.030.
[7]LU Chijie,TSAI Duming.Independent component analysis-based defect detection in patterned liquid crystal display surfaces[J].Image and Vision Computing,2008,26(7):955-970.DOI:10.1016/j.imavis.2007.10.007.
[8]ZHANG Lei,GRIFT T E.A new approach to crop-row detection in corn[C]//American Society of Agricultural and Biological Engineers Annual International Meeting 2010.St.Joseph,MI:American Society of Agricultural and Biological Engineers,2010:3950-3964.DOI:10.13031/2013.29834.
[9]SUN T H,TSENG C C,CHEN M S.Electric contacts inspection using machine vision[J].Image and Vision Computing,2010,28(6):890-901.DOI:10.1016/j.imavis.2009.11.006.
[10]王震宇.基于机器视觉钢板表面缺陷检测技术研究[J].计算机与现代化,2013(7):130-134.DOI:10.3969/j.issn.1006-2475.2013.07.035.
[11]李超,孙俊.基于机器视觉方法的焊缝缺陷检测及分类算法[J].计算机工程与应用,2018,54(6):264-270.DOI:10.3778/j.issn.1002-8331.1609-0322.
[12]黄志鸿,毛建旭,王耀南,等.基于机器视觉的啤酒瓶口缺陷检测分类方法研究[J].电子测量与仪器学报,2016,30(6):873-879.DOI:10.13382/j.jemi.2016.06.006.
[13]杨洋.基于SVM的印刷品缺陷在线检测[D].武汉:华中科技大学,2012.
[14]孙智权,周奇,陈震,等.基于CMOS图像传感器的太阳能电池缺陷检测系统设计[J].仪表技术与传感器,2018(1):60-63.DOI:10.3969/j.issn.1002-1841.2018.01.015.
[15]刘焕军,王耀南,段峰.基于支撑向量机的空瓶智能检测方法[J].控制与决策,2005,20(12):1434-1437.DOI:10.13195/j.cd.2005.12.116.liuhj.026.
[16]贾云得.机器视觉[M].北京:科学出版社,2000:48-81.
[17]贺秋伟,王龙山,于忠党,等.基于图像处理和支持向量机的微型齿轮缺陷检测[J].吉林大学学报(工学版),2008,38(3):565-569.DOI:10.13229/j.cnki.jdxbgxb2008.03.008.
[18]KEERTHIS S,LIN C J.Asymptotic behaviors of support vector machines with Gaussian kernel[J].Neural Computation,2003,15(7):1667-1689.DOI:10.1162/089976603321891855.
[19]吴彰良,孙长库,刘洁.基于支持向量机的油封缺陷图像检测方法[J].光电工程,2012,39(3):40-45.DOI:10.3969/j.issn.1003-501X.2012.03.008.