基于Zernike矩的圆形插针特征点提取算法研究
|
摘要
在电连接器插针缺陷检测系统中,插针图像特征点的获取是检测关键,为了实现插针位置的高精度检测,提出了一种基于Zernike矩的插针特征点定位算法。通过增大Zernike矩的算子模板尺寸,以及优化边缘阶跃模型,提高了亚像素边缘检测算法的精度。将最大熵阈值法应用于Zernike矩边缘检测算法,实现了自动选择最佳阈值,解决了传统算法中手动调节阈值的低效率问题。对提取的亚像素边缘点进行拟合获得椭圆目标中心,实现了插针的位置特征点提取。仿真实验与实际测试结果表明,该算法能够实现图像边缘的亚像素检测和插针特征点的准确定位,算法定位误差小于0.1个像素。
In the defect detection system for electronic connectors,it is a key point to acquire the feature points of pin images.Thus in order to realize the high accuracy measurment of pin locations,a feature point positioning algorithm based on Zernike moment was proposed.The precision of sub-pixel edge detection algorithm was improved by increasing the size of operator model of Zernike moment and optimizing the edge step model.The maximum entropy threshold method was applied to the Zernike moment edge detection algorithm,which realizes the automatic selection of the optimal threshold and overcomes the disadvantage of the low efficiency of the manual adjusting threshold in traditional algorithms.The simulation image and the actual test result show that the algorithm can realize sub-pixel detection of image edge,accurately identify the feature point of the pin,and the algorithm localization error is less than 0.1 pixel.
In the defect detection system for electronic connectors,it is a key point to acquire the feature points of pin images.Thus in order to realize the high accuracy measurment of pin locations,a feature point positioning algorithm based on Zernike moment was proposed.The precision of sub-pixel edge detection algorithm was improved by increasing the size of operator model of Zernike moment and optimizing the edge step model.The maximum entropy threshold method was applied to the Zernike moment edge detection algorithm,which realizes the automatic selection of the optimal threshold and overcomes the disadvantage of the low efficiency of the manual adjusting threshold in traditional algorithms.The simulation image and the actual test result show that the algorithm can realize sub-pixel detection of image edge,accurately identify the feature point of the pin,and the algorithm localization error is less than 0.1 pixel.
引文
[1]杨奋为.航天电连接器组件的常见故障检验[J].机电元件,2005(3):40-45,52.Yang Fenwei.Inspection on common defaults of space electrical connector[J].Electromechanical Components,2005(3):40-45,52.
[2]孙双花.视觉测量关键技术及在自动检测中的应用[D].天津:天津大学,2007.Sun Shuanghua.Key technologies of visual measurement and its application in automatic detection[D].Tianjin:Tianjin University,2007.
[3]杜福洲,赵德龙.基于双目视觉的多型号电连接器检测技术[J].航空精密制造技术,2016,52(5):23-28.Du Fuzhou,Zhao Delong.Research on multi-type electrical connectors detection based on binocular vision[J].Aviation Precision Manufacturing Technol.,2016,52(5):23-28.
[4]Mukhopadhyay P,Chaudhuri B B.A survey of Hough transform[J].Pattern Recognition,2015,48(3):993-1010.
[5]李元锋,李正生,王洋.CCD摄像头圆目标中心定位方法[J].应用光学,2010,31(6):956-960.Li Yuanfeng,Li Zhengsheng,Wang Yang.Center location of circle target captured by CCD camera[J].J.of Appl.Opt.,2010,31(6):956-960.
[6]李杰,耿学贤.一种基于视觉测量的亚像素边缘检测算法[J].科学技术与工程,2015,15(19):32-36.Li Jie,Geng Xuexian.A subpixel edge detection algorithm based on visual measurements[J].Science Technol.and Engin.,2015,15(19):32-36.
[7]Lyvers E P,Mitchell O R,Akey M L,et al.Subpixel measurements using a moment-based edge operator[J].IEEETrans.on Pattern Analysis and Machine Intelligence,1989,11(12):1293-1309.
[8]Ghosal S,Mehrotra R.Orthogonal moment operators for subpixel edge detection[J].Pattern Recognition,1993,26(2):295-306.
[9]Teague M R.Image analysis via the general theory of moments[J].J.of the Opt.Society of America,1980,70(8):1468.
[10]魏本征,赵志敏,华晋.基于改进形态学梯度和Zernike矩的亚像素边缘检测方法[J].仪器仪表学报,2010,31(4):838-844.Wei Benzheng,Zhao Zhimin,Hua Jin.Sub-pixel edge detection method based on improved morphological gradient and Zernike moment[J].Chinese J.of Scientific Instrument,2010,31(4):838-844.
[11]崔继文,谭久彬.基于Zernike矩的亚像素边缘定位算法[J].光学技术,2005,31(5):779-782.Cui Jiwen,Tan Jiubin.Algorithm for edge subpixel location based on Zernike moment[J].Opt.Technique,2005,31(5):779-782.
[2]孙双花.视觉测量关键技术及在自动检测中的应用[D].天津:天津大学,2007.Sun Shuanghua.Key technologies of visual measurement and its application in automatic detection[D].Tianjin:Tianjin University,2007.
[3]杜福洲,赵德龙.基于双目视觉的多型号电连接器检测技术[J].航空精密制造技术,2016,52(5):23-28.Du Fuzhou,Zhao Delong.Research on multi-type electrical connectors detection based on binocular vision[J].Aviation Precision Manufacturing Technol.,2016,52(5):23-28.
[4]Mukhopadhyay P,Chaudhuri B B.A survey of Hough transform[J].Pattern Recognition,2015,48(3):993-1010.
[5]李元锋,李正生,王洋.CCD摄像头圆目标中心定位方法[J].应用光学,2010,31(6):956-960.Li Yuanfeng,Li Zhengsheng,Wang Yang.Center location of circle target captured by CCD camera[J].J.of Appl.Opt.,2010,31(6):956-960.
[6]李杰,耿学贤.一种基于视觉测量的亚像素边缘检测算法[J].科学技术与工程,2015,15(19):32-36.Li Jie,Geng Xuexian.A subpixel edge detection algorithm based on visual measurements[J].Science Technol.and Engin.,2015,15(19):32-36.
[7]Lyvers E P,Mitchell O R,Akey M L,et al.Subpixel measurements using a moment-based edge operator[J].IEEETrans.on Pattern Analysis and Machine Intelligence,1989,11(12):1293-1309.
[8]Ghosal S,Mehrotra R.Orthogonal moment operators for subpixel edge detection[J].Pattern Recognition,1993,26(2):295-306.
[9]Teague M R.Image analysis via the general theory of moments[J].J.of the Opt.Society of America,1980,70(8):1468.
[10]魏本征,赵志敏,华晋.基于改进形态学梯度和Zernike矩的亚像素边缘检测方法[J].仪器仪表学报,2010,31(4):838-844.Wei Benzheng,Zhao Zhimin,Hua Jin.Sub-pixel edge detection method based on improved morphological gradient and Zernike moment[J].Chinese J.of Scientific Instrument,2010,31(4):838-844.
[11]崔继文,谭久彬.基于Zernike矩的亚像素边缘定位算法[J].光学技术,2005,31(5):779-782.Cui Jiwen,Tan Jiubin.Algorithm for edge subpixel location based on Zernike moment[J].Opt.Technique,2005,31(5):779-782.