圆度测量系统研究与实现
|
摘要
类圆物体的机器视觉检测在工农业生产、医疗卫生等领域有着广泛的应用,圆度是类圆性物体检测的重要指标之一。本文在对卷烟、滤棒在线圆度测量系统预研的背景下,研究了使用机器视觉进行圆度测量的技术。
本文设计了一整套搭建机器视觉圆度测量系统的执行流程,包括图像采集、摄像机标定、图像处理、图像分析等。从硬件、软件方面全面介绍了圆度测量系统的研究和实现。
本文介绍了硬件的选型知识,着重介绍了光源的选型。比较了多种经典的摄像机标定、二值化、距离变换、图像分割方法,并给出了实现效果图,为系统的实现奠定基础,也为其他圆度测量系统提供了参考。
本文提出了边缘权值变换的概念,并给出了实现算法和实现效果图。
在圆度计算方面,本文创造性地提出使用边缘检测图像和二值化图像相结合的方法求取直径。提出赋予搜索点向心力、边缘引力、边缘权值力三种能量,以能量最大找到物体实际轮廓点的方法。在对偶然误差的处理上,本文提出了使用最大值窗口扫描找到最长弦作为直径的方法。
根据本文提出的方法,使用LahVIEW平台和VC++平台相结合实现了卷烟、滤棒在线圆度测量系统。在实验室坏境下取得了较好的测量效果。
Machine Vision Measurement of circle-like object has a wide application in industry, agriculture, and medical treatment. Roundness is one of the most important items applied in circle-like object measurement. Based on the pre-research of roundness measurement system of the cigarette and filter stick, this paper has done some research on machine vision roundness measurement technology.
In this paper, the implementation method to build a machine vision roundness measurement system, including image acquisition, camera calibration, image process and image analysis, has been designed. A comprehensive introduction to the research and realization process of the system has been given from the aspects of both software and hardware.
This paper has introduced the selection of hardware, especially the selection of light source. And this paper has given a comparison of a variety of the algorithms such as binarization , distance transform ,image segmentation, which are the foundation for the realization of the system and also the reference for other roundness measurement systems.
An edge weight concept has been raised in this paper. Furthermore, the implementation algorithms and images of result have been showed and approached.
In the aspect of roundness calculation, a creative way to get diameter by using both edge detection image and binarization image has been applied to practice. Meanwhile, three types of energy named Centripetal force, edge force and edge weight force has been well considered. This paper has found the real edge of the object by maximizing the energy of the search points. In order to get rid of the accidental error, a maximum scanning window has been used to treat the maximum value as the diameter.
According to the method approached, this paper has combined the LabVIEW platform and VC ++ platform to realize the online roundness measurement system of Cigarette and filter stick. A preferable result has been obtained in laboratory environment.
本文设计了一整套搭建机器视觉圆度测量系统的执行流程,包括图像采集、摄像机标定、图像处理、图像分析等。从硬件、软件方面全面介绍了圆度测量系统的研究和实现。
本文介绍了硬件的选型知识,着重介绍了光源的选型。比较了多种经典的摄像机标定、二值化、距离变换、图像分割方法,并给出了实现效果图,为系统的实现奠定基础,也为其他圆度测量系统提供了参考。
本文提出了边缘权值变换的概念,并给出了实现算法和实现效果图。
在圆度计算方面,本文创造性地提出使用边缘检测图像和二值化图像相结合的方法求取直径。提出赋予搜索点向心力、边缘引力、边缘权值力三种能量,以能量最大找到物体实际轮廓点的方法。在对偶然误差的处理上,本文提出了使用最大值窗口扫描找到最长弦作为直径的方法。
根据本文提出的方法,使用LahVIEW平台和VC++平台相结合实现了卷烟、滤棒在线圆度测量系统。在实验室坏境下取得了较好的测量效果。
Machine Vision Measurement of circle-like object has a wide application in industry, agriculture, and medical treatment. Roundness is one of the most important items applied in circle-like object measurement. Based on the pre-research of roundness measurement system of the cigarette and filter stick, this paper has done some research on machine vision roundness measurement technology.
In this paper, the implementation method to build a machine vision roundness measurement system, including image acquisition, camera calibration, image process and image analysis, has been designed. A comprehensive introduction to the research and realization process of the system has been given from the aspects of both software and hardware.
This paper has introduced the selection of hardware, especially the selection of light source. And this paper has given a comparison of a variety of the algorithms such as binarization , distance transform ,image segmentation, which are the foundation for the realization of the system and also the reference for other roundness measurement systems.
An edge weight concept has been raised in this paper. Furthermore, the implementation algorithms and images of result have been showed and approached.
In the aspect of roundness calculation, a creative way to get diameter by using both edge detection image and binarization image has been applied to practice. Meanwhile, three types of energy named Centripetal force, edge force and edge weight force has been well considered. This paper has found the real edge of the object by maximizing the energy of the search points. In order to get rid of the accidental error, a maximum scanning window has been used to treat the maximum value as the diameter.
According to the method approached, this paper has combined the LabVIEW platform and VC ++ platform to realize the online roundness measurement system of Cigarette and filter stick. A preferable result has been obtained in laboratory environment.
引文
[1]刘相滨.类圆性颗粒图像分割技术研究[D].湖南大学,2006
[2]陈浩,王景中,姚光明.基于类圆分割的棒材计数图像识别[J].工程图学学报2004,4:98-103
[3]辛颖.棒材图像计数系统研究[D].东北大学.2006
[4]杨蜀泰.大米外观品质计算机视觉检测的研究[D].西北农林科技大学.2005.
[5]刘今越.基于视觉的精密定位系统研究及应用[D].河北工业大学,2005
[6]http://www.china-vision.net/
[7]马颂德,张正友著.计算机视觉:计算理论与算法基础[M].北京:科学出版社.1998
[8]邱茂林,马颂德,李毅.计算机视觉中摄像机定标综述[J].自动化学报.Vol.26,NO.1.Jan 2000:/43-55
[9]李欣菊.摄像机标定算法研究[D].中国科学院自动化研究所.2003
[10]NI.NI Vision Concepts Manual[M].National Instruments,2007.
[11],ROGER Y.TSAI.A Versatile Camera Calibration Technique for High-Accuracy 3D Machine Vision Metrology Using Off-the-shelf TV Cameras and Lenses[J].IEEE Journal of Robotics and Automation,Vol.RA-3,323-344,NO.4,1987.8
[12]邹凤娇.摄像机标定及相关技术研究[D].四川大学.2005
[13]Zhang Z.A Flexible New Technique for Camera Calibration[J].IEEE Transaction on Pattern Analysis and Machine Intelligence,2000,22(11):1330-1334
[14]贾丹.摄像机现场标定算法研究[D].哈尔滨工程大学,2007
[15]吴冰,秦志远.自动确定图像二值化最佳阈值的新方法[J].测绘学院学报.18(4),2001:283-286
[16]钟远高,黄昶,倪国瑜.类圆粘连物体圆度测量[J].计算机系统应用.11.2008:113-115
[17]魏诚.车牌图像二值化效果的改进算法研究[D].浙江大学.2006
[18]Per-Erik DANIELSSON.Euclidean Distance Mapping[J].Computer Graphics and Image Processing.14,1980:227-248.
[19]Rosenfeld A.Pfaltz J L.Sequential operations in digital picture processing.Journal of the Association for Computing Machinery(ACM),1966,13(4);471-494.
[20]章毓晋.图像理解与计算机视觉[M].北京.清华大学出版社.2001
[21]唐娉.基于多分辨率分析的分水岭分割技术研究[D].中国科学院研究生院.2006
[22]游迎荣,范影乐,庞全.基于距离变换的粘连细胞分割方法[J].计算机工程与应用.2005.20:206-208
[23]Digabel H,and Lantuejoul C.Iterative algorithms In:Cheimant J L,Ed,Proc.2~(nd)European Symp.Quantitative Analysis Micro-structures in Material Science,Biology and Medicine.Stuttgart West Germany.Riedcrcr Verlag,1978:85-99
[24]Bcucher S,and Lantu_cjoul C,Use of watersheds in contour detection In Proc.International Workshop on Image Processing.Real-Time Edge and Motion Detection/Estimation,Rennes,September(1979)
[25]Vincent L,and Soille P.Watersheds in digital spaces:an efficient algorithm based on immersion simulations.IEEE.Trans.Patt.Anal.Mach.Intell.13,6(1991),583-598
[26]Meyer F.Topographic distance and watershed lines.Signal Processing 38(1994),113-125
[27]国家烟草专卖局,全国烟草标准化技术委员会.烟草技术标准标准汇编(三).中国标准出版社.2001.
[28]Roberts L G.Machine Perception of Three-Dimensional Solids Optical and Electro-Optical Information Processing[M].Cambridge:MIT Press,1965:157-197.
[29]Prewitt J M S,Menclelsohn M L.The Analysis of Cell Images[J].Annals of th N.Y.Academy of Sciences.1966,128;1035-1053.
[30]JOHN CANNY.A Computational Approach to Edge Detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence.PAM-8(6),November,1986:679-698
[31]李宏贵,李兴国.一种改进的梯度算子[J].中国图象图形学报.Vol.8(A),No.3Mar.2003:253-255
[32]华继钊,尹红梅,李志军,郭振民.Canny算法的改进[J],微机发展,Vol.15,No.4Apr.2005:24-26.
[33]王文豪.图象边缘检测中边界闭合性的分析与探讨[J].计算机与信息技 术.2005,12:90-91
[34]李了了,邓善熙,丁兴号,基于天津法的图像分块二值化算法[J].微计算机信息,2005,21(8-3):76-77.
[35]沈晗.非均匀光照文本图像的分割[D].上海,华东师范大学.2006
[36]USB2.0高分辨率数字摄像机(MVC-II一1M/1MM/2M/3M)用户手册.微视图像.
[37]NI IMAQ~(TM)VI Reference Manual.National Instruments.June 1998.
[38]种伟亮.基于分水岭算法的医学图像分析[D].上海,上海交通大学,2007
[39]魏杰,穆雷,汪伟,李郝林.基丁NI-LabVIEW和IMAQ开发平台的虚拟测量系统[J].精密制造与自动化.2005,2:43-44.
[40]林锐,韩永泉.高质量的C++编程指南[M].电子工业出版社.2007,5.
[2]陈浩,王景中,姚光明.基于类圆分割的棒材计数图像识别[J].工程图学学报2004,4:98-103
[3]辛颖.棒材图像计数系统研究[D].东北大学.2006
[4]杨蜀泰.大米外观品质计算机视觉检测的研究[D].西北农林科技大学.2005.
[5]刘今越.基于视觉的精密定位系统研究及应用[D].河北工业大学,2005
[6]http://www.china-vision.net/
[7]马颂德,张正友著.计算机视觉:计算理论与算法基础[M].北京:科学出版社.1998
[8]邱茂林,马颂德,李毅.计算机视觉中摄像机定标综述[J].自动化学报.Vol.26,NO.1.Jan 2000:/43-55
[9]李欣菊.摄像机标定算法研究[D].中国科学院自动化研究所.2003
[10]NI.NI Vision Concepts Manual[M].National Instruments,2007.
[11],ROGER Y.TSAI.A Versatile Camera Calibration Technique for High-Accuracy 3D Machine Vision Metrology Using Off-the-shelf TV Cameras and Lenses[J].IEEE Journal of Robotics and Automation,Vol.RA-3,323-344,NO.4,1987.8
[12]邹凤娇.摄像机标定及相关技术研究[D].四川大学.2005
[13]Zhang Z.A Flexible New Technique for Camera Calibration[J].IEEE Transaction on Pattern Analysis and Machine Intelligence,2000,22(11):1330-1334
[14]贾丹.摄像机现场标定算法研究[D].哈尔滨工程大学,2007
[15]吴冰,秦志远.自动确定图像二值化最佳阈值的新方法[J].测绘学院学报.18(4),2001:283-286
[16]钟远高,黄昶,倪国瑜.类圆粘连物体圆度测量[J].计算机系统应用.11.2008:113-115
[17]魏诚.车牌图像二值化效果的改进算法研究[D].浙江大学.2006
[18]Per-Erik DANIELSSON.Euclidean Distance Mapping[J].Computer Graphics and Image Processing.14,1980:227-248.
[19]Rosenfeld A.Pfaltz J L.Sequential operations in digital picture processing.Journal of the Association for Computing Machinery(ACM),1966,13(4);471-494.
[20]章毓晋.图像理解与计算机视觉[M].北京.清华大学出版社.2001
[21]唐娉.基于多分辨率分析的分水岭分割技术研究[D].中国科学院研究生院.2006
[22]游迎荣,范影乐,庞全.基于距离变换的粘连细胞分割方法[J].计算机工程与应用.2005.20:206-208
[23]Digabel H,and Lantuejoul C.Iterative algorithms In:Cheimant J L,Ed,Proc.2~(nd)European Symp.Quantitative Analysis Micro-structures in Material Science,Biology and Medicine.Stuttgart West Germany.Riedcrcr Verlag,1978:85-99
[24]Bcucher S,and Lantu_cjoul C,Use of watersheds in contour detection In Proc.International Workshop on Image Processing.Real-Time Edge and Motion Detection/Estimation,Rennes,September(1979)
[25]Vincent L,and Soille P.Watersheds in digital spaces:an efficient algorithm based on immersion simulations.IEEE.Trans.Patt.Anal.Mach.Intell.13,6(1991),583-598
[26]Meyer F.Topographic distance and watershed lines.Signal Processing 38(1994),113-125
[27]国家烟草专卖局,全国烟草标准化技术委员会.烟草技术标准标准汇编(三).中国标准出版社.2001.
[28]Roberts L G.Machine Perception of Three-Dimensional Solids Optical and Electro-Optical Information Processing[M].Cambridge:MIT Press,1965:157-197.
[29]Prewitt J M S,Menclelsohn M L.The Analysis of Cell Images[J].Annals of th N.Y.Academy of Sciences.1966,128;1035-1053.
[30]JOHN CANNY.A Computational Approach to Edge Detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence.PAM-8(6),November,1986:679-698
[31]李宏贵,李兴国.一种改进的梯度算子[J].中国图象图形学报.Vol.8(A),No.3Mar.2003:253-255
[32]华继钊,尹红梅,李志军,郭振民.Canny算法的改进[J],微机发展,Vol.15,No.4Apr.2005:24-26.
[33]王文豪.图象边缘检测中边界闭合性的分析与探讨[J].计算机与信息技 术.2005,12:90-91
[34]李了了,邓善熙,丁兴号,基于天津法的图像分块二值化算法[J].微计算机信息,2005,21(8-3):76-77.
[35]沈晗.非均匀光照文本图像的分割[D].上海,华东师范大学.2006
[36]USB2.0高分辨率数字摄像机(MVC-II一1M/1MM/2M/3M)用户手册.微视图像.
[37]NI IMAQ~(TM)VI Reference Manual.National Instruments.June 1998.
[38]种伟亮.基于分水岭算法的医学图像分析[D].上海,上海交通大学,2007
[39]魏杰,穆雷,汪伟,李郝林.基丁NI-LabVIEW和IMAQ开发平台的虚拟测量系统[J].精密制造与自动化.2005,2:43-44.
[40]林锐,韩永泉.高质量的C++编程指南[M].电子工业出版社.2007,5.