触点零件形貌在线自学习视觉检测系统研究
|
摘要
触点零件是工业中重要的电器元件,完成关键的切换功能,是决定电器使用寿命的主要因素,触点零件的表面形貌对零件的接触性能起着相当重要的作用。目前国内检测手段落后,采用计算机视觉检测的方法代替人工具有很大的实际应用价值。针对触点零件形貌多样的特点,本文研究开发能够实现自适应学习功能的在线视觉检测系统。
论文通过对检测需求的分析,设计了系统的整体结构和自学习检测流程,详细讨论了系统各个组成部分的功能及选型,并设计了软件系统的功能模块及工作流程。
根据检测系统实时性和准确性的要求,本文在图像处理阶段研究了基于高斯滤波模板的快速有效图像预处理方法,针对触点零件图像特点提出了一种基于改进极小值点的自适应阈值分割算法和多图中值阈值确定算法,并构建了5大类图像特征库,共包含45个特征。
通过对多类别模式识别主要因素的分析研究,提出了特征预处理方法和基于典型变量的特征优化算法,设计了一种两类浮动搜索-多类后向选择算法实现多类别特征选择,并结合支持向量机多级二叉树的多类别分类策略,共同实现了多类别触点零件快速学习和分类检测。
在上述工作基础上,论文搭建了试验硬件平台,开发了实现上述算法的相关软件,对多组零件进行了试验,取得了较好的试验结果。论文也分析了误差来源,提出进一步研究改进的方向。
As an important industrial electrical component, contact component decides electric equipments’life. The surface appearance features of contact component play an important role in contact performance. The detection methods depending on manual stay low level, using computer vision technique to replace human vision in quality detection has great practical value. Aimed to different varieties of contact appearance features, the online vision detection system achieving self-learning function is researched in this thesis.
Firstly, the structure and self-learning process of detection system are designed through the detection requirements analysis. The function parts and their selection methods are discussed in detail. The function modules and workflow of software system are designed.
In image processing, according to real-time and accuracy requirements, the Gaussian filter template is studied as a quick and efficient image pre-processing method. An adaptive threshold segmentation method based on improved minimum point algorithm and multi-graph median algorithm is proposed. And the 5 major categories of image feature library is built, which includes 45 features.
Through the analysis of major factors that influence the multi-category pattern recognition, the feature pre-processing method and optimization method based on canonical variables are put forward. To realize multi-category feature selection, a two-category floating search and multi-category backward selection algorithm is designed. Combined these algorithms with SVM multi-level binary tree classification strategy, the system realizes the multi-category quick self-learning and classification detection.
Based on the work mentioned above, the experimental platform is developed. The experimental system is tested through several groups of contact components and gets good results. The future research direction is proposed after the error analysis.
论文通过对检测需求的分析,设计了系统的整体结构和自学习检测流程,详细讨论了系统各个组成部分的功能及选型,并设计了软件系统的功能模块及工作流程。
根据检测系统实时性和准确性的要求,本文在图像处理阶段研究了基于高斯滤波模板的快速有效图像预处理方法,针对触点零件图像特点提出了一种基于改进极小值点的自适应阈值分割算法和多图中值阈值确定算法,并构建了5大类图像特征库,共包含45个特征。
通过对多类别模式识别主要因素的分析研究,提出了特征预处理方法和基于典型变量的特征优化算法,设计了一种两类浮动搜索-多类后向选择算法实现多类别特征选择,并结合支持向量机多级二叉树的多类别分类策略,共同实现了多类别触点零件快速学习和分类检测。
在上述工作基础上,论文搭建了试验硬件平台,开发了实现上述算法的相关软件,对多组零件进行了试验,取得了较好的试验结果。论文也分析了误差来源,提出进一步研究改进的方向。
As an important industrial electrical component, contact component decides electric equipments’life. The surface appearance features of contact component play an important role in contact performance. The detection methods depending on manual stay low level, using computer vision technique to replace human vision in quality detection has great practical value. Aimed to different varieties of contact appearance features, the online vision detection system achieving self-learning function is researched in this thesis.
Firstly, the structure and self-learning process of detection system are designed through the detection requirements analysis. The function parts and their selection methods are discussed in detail. The function modules and workflow of software system are designed.
In image processing, according to real-time and accuracy requirements, the Gaussian filter template is studied as a quick and efficient image pre-processing method. An adaptive threshold segmentation method based on improved minimum point algorithm and multi-graph median algorithm is proposed. And the 5 major categories of image feature library is built, which includes 45 features.
Through the analysis of major factors that influence the multi-category pattern recognition, the feature pre-processing method and optimization method based on canonical variables are put forward. To realize multi-category feature selection, a two-category floating search and multi-category backward selection algorithm is designed. Combined these algorithms with SVM multi-level binary tree classification strategy, the system realizes the multi-category quick self-learning and classification detection.
Based on the work mentioned above, the experimental platform is developed. The experimental system is tested through several groups of contact components and gets good results. The future research direction is proposed after the error analysis.
引文
[1]姚芳.触点电接触失效预测的研究.博士学位论文,河北工业大学, 2004
[2]陆俭国.电器可靠性理论及其应用.北京:机械工业出版社, 1996
[3]李志华.接触电阻引发的飞机重复性故障.航空工程与维修, 1999(1): 32-33
[4]荣命哲.电接触理论.北京:机械工业出版社, 2004
[5]尹娜.新型银基导电陶瓷复合电接触材料研究.博士学位论文,山东大学, 2007
[6]马大国.原木表面结子的计算机视觉检测.硕士学位论文,东北林业大学, 2002
[7]孙慧.用双目立体视觉技术重建电器触头立体图像.硕士学位论文,河北工业大学, 2002
[8]成芳.稻种质量的机器视觉无损检测研究.博士学位论文,浙江大学, 2004
[9]吴健康,肖锦玉.计算机视觉基本理论和方法.合肥:中国科学技术大学出版社, 1993
[10]钟玉琢,乔秉新,李树青.机器人视觉技术.北京:国防工业出版社, 1994
[11]房超.机器视觉及其在工业检测中的应用.自动化博览, 2007(8): 46-48
[12] Yiannis Aloimonos, Aeds Rosenfield. Computer Vision. Science, 1991(253): 1181-1324
[13] Theo Pavilidis. Algorithms for Graphics and Image Processing. Computer Science, 1982
[14]赵荣椿.数字图像处理导论.西安:西北工业大学出版社, 1995
[15]吴健康.计算机视觉基本理论和实践.合肥:中国科学技术大学出版社, 1993
[16]夏良正,李久贤.数字图像处理.南京:东南大学出版社, 1999
[17]任潇潇.复杂曲面坐标测量的建模与应用.硕士学位论文,合肥工业大学, 2003
[18] Halioua M, Liu H C. Optical three-dimensional sensing by phase measuring profilometry. Optics and Lasers in Engineering, 1989,11(3): 185-215
[19]于起峰,陆宏伟,刘肖琳.基于图像的精密测量与运动测量.北京:科学出版社, 2002
[20]孙长库.激光测量技术.天津:天津大学出版社, 2000
[21] Anil K. Jain, Chitra Dorai. Practicing Vision: Integration, Evaluation and Application. Pattern Recognition, 1997(30): 183-196
[22] Wen Hsiang Tsai. Moment-preserving thresholding: A new approach. Computer Vision, Graphics and Image Processing, 1998(41): 11-14
[23]卞铭.机器视觉在在线检测中的应用.硕士学位论文,合肥工业大学, 2006
[24]孟然.基于机器视觉的工件特征识别与分类方法研究.硕士学位论文,天津科技大学, 2006
[25]谢立.双测点机器视觉玻壳产品动态识别系统的开发.硕士学位论文,天津科技大学, 2005
[26] David G. Lowe. Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision, 2004(60): 91-110
[27] Kunio Osada, Takahiko Furuya, Ryutarou Ohbuchi. SHREC’08 Entry: Local Volumetric Features for 3D Model Retrieval. Shape Modeling and Applications, 2008(6): 245-246
[28]廖强,蔡飞,吴存学.基于机器视觉技术的精确定位方法研究及应用.仪器仪表与检测, 2005(4): 101-103
[29]刘焕军,王耀南,段峰.空瓶的智能检测算法研究.湖南大学学报(自然科学版), 2005.32(l): 20-24
[30]马金奎,路长厚,程英.机器视觉测试系统.工具技术, 2004,38(9): 129-132
[31]郑威.基于嵌入式机器视觉的缺陷检测系统硬件设计.硕士学位论文,武汉理工大学, 2006
[32] CCD&CMOS图像和机器视觉产品手册.凌云光视数字图像公司, 2005
[33]尚会超.印刷图像在线检测的算法研究与系统实现.博士学位论文,华中科技大学, 2006
[34] Mahmoud Meribout, Mamoru Nakanishi, Takeshi Ogura. Accurate and Real-time Image Processing on a New PC-compatible Board. Real-time Imaging, 2002,8(1): 35-51
[35]周太明.光源原理与设计.上海:复旦大学出版社, 2006
[36]李福建,张元培.机器视觉系统组成研究.自动化博览, 2004(2): 61-63
[37]唐向阳,张勇,李江有等.机器视觉关键技术的现状及应用展望.昆明理工大学学报(理工版), 2004,29(2): 63-93
[38]李洪宇.机器视觉图像采集设计及研究.硕士学位论文,大庆石油学院, 2007
[39]李俊.机器视觉照明光源关键技术研究.硕士学位论文,天津理工大学, 2006
[40] J Batlle, J Marti, P Ridao. A New FPGA/DSP-Based Parallel Architecture for Real-Time Image Processing. Real-Time Imaging, 2002,8(5): 345-356
[41]陈兵旗.实用数字图像处理与分析.北京:中国农业大学出版社,清华大学出版社, 2008
[42]何明一,卫保国.数字图像处理.北京:科学出版社, 2008
[43] Tinku Acharya, Ajoy K. Ray. Image processing: principles and applications.北京:清华大学出版社, 2007
[44] Rafael C. Gonzalez, Richard E. Woods. Digital image processing.北京:电子工业出版社, 2007
[45]许景波.高斯滤波器逼近理论与应用研究.博士学位论文,哈尔滨工业大学, 2007
[46]余松煜,周源华,张瑞.数字图像处理.上海:上海交通大学出版社, 2007
[47]张健.一种图像分割关键技术的研究和实现.硕士学位论文,北京邮电大学, 2008
[48]张新峰,沈兰荪.模式识别及其在图像处理中的应用.测控技术, 2004,23(5): 28-32
[49]范立南,韩晓微,张广渊.图像处理与模式识别.北京:科学出版社, 2007
[50] Sergios Theodoridis, Konstantinos Koutroumbas. Pattern recognition.北京:电子工业出版社, 2004
[51]钟珞,潘昊,封筠等.模式识别.武汉:武汉大学出版社, 2006
[52] Linda G. Shapiro, George C. Stockman. Computer vision.北京:机械工业出版社, 2005
[53] David A. Forsyth, Jean Ponce. Computer vision: a modern approach.北京:电子工业出版社, 2004
[54]曹玉萍,田学民.基于典型变量分析状态残差的故障检测方法.控制工程, 2007,7(14): 74-76
[55]吴换妮.基于支持向量机的多类模式识别模型.硕士学位论文,重庆大学, 2007
[56]袁亚湘.非线性优化计算方法.北京:科学出版社, 2008
[57]王言金.最优化的不可行内点算法研究.博士学位论文,武汉大学, 2004
[58]雍龙泉.二次规划的算法研究.硕士学位论文,西安电子科技大学, 2005
[59] Hu Fangxia, He Xin, Niu Tiangui. Study on the key image processing technology in the inspection of packing quality for small-pack cigarettes. Computer Science, 207,62: 67-71
[2]陆俭国.电器可靠性理论及其应用.北京:机械工业出版社, 1996
[3]李志华.接触电阻引发的飞机重复性故障.航空工程与维修, 1999(1): 32-33
[4]荣命哲.电接触理论.北京:机械工业出版社, 2004
[5]尹娜.新型银基导电陶瓷复合电接触材料研究.博士学位论文,山东大学, 2007
[6]马大国.原木表面结子的计算机视觉检测.硕士学位论文,东北林业大学, 2002
[7]孙慧.用双目立体视觉技术重建电器触头立体图像.硕士学位论文,河北工业大学, 2002
[8]成芳.稻种质量的机器视觉无损检测研究.博士学位论文,浙江大学, 2004
[9]吴健康,肖锦玉.计算机视觉基本理论和方法.合肥:中国科学技术大学出版社, 1993
[10]钟玉琢,乔秉新,李树青.机器人视觉技术.北京:国防工业出版社, 1994
[11]房超.机器视觉及其在工业检测中的应用.自动化博览, 2007(8): 46-48
[12] Yiannis Aloimonos, Aeds Rosenfield. Computer Vision. Science, 1991(253): 1181-1324
[13] Theo Pavilidis. Algorithms for Graphics and Image Processing. Computer Science, 1982
[14]赵荣椿.数字图像处理导论.西安:西北工业大学出版社, 1995
[15]吴健康.计算机视觉基本理论和实践.合肥:中国科学技术大学出版社, 1993
[16]夏良正,李久贤.数字图像处理.南京:东南大学出版社, 1999
[17]任潇潇.复杂曲面坐标测量的建模与应用.硕士学位论文,合肥工业大学, 2003
[18] Halioua M, Liu H C. Optical three-dimensional sensing by phase measuring profilometry. Optics and Lasers in Engineering, 1989,11(3): 185-215
[19]于起峰,陆宏伟,刘肖琳.基于图像的精密测量与运动测量.北京:科学出版社, 2002
[20]孙长库.激光测量技术.天津:天津大学出版社, 2000
[21] Anil K. Jain, Chitra Dorai. Practicing Vision: Integration, Evaluation and Application. Pattern Recognition, 1997(30): 183-196
[22] Wen Hsiang Tsai. Moment-preserving thresholding: A new approach. Computer Vision, Graphics and Image Processing, 1998(41): 11-14
[23]卞铭.机器视觉在在线检测中的应用.硕士学位论文,合肥工业大学, 2006
[24]孟然.基于机器视觉的工件特征识别与分类方法研究.硕士学位论文,天津科技大学, 2006
[25]谢立.双测点机器视觉玻壳产品动态识别系统的开发.硕士学位论文,天津科技大学, 2005
[26] David G. Lowe. Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision, 2004(60): 91-110
[27] Kunio Osada, Takahiko Furuya, Ryutarou Ohbuchi. SHREC’08 Entry: Local Volumetric Features for 3D Model Retrieval. Shape Modeling and Applications, 2008(6): 245-246
[28]廖强,蔡飞,吴存学.基于机器视觉技术的精确定位方法研究及应用.仪器仪表与检测, 2005(4): 101-103
[29]刘焕军,王耀南,段峰.空瓶的智能检测算法研究.湖南大学学报(自然科学版), 2005.32(l): 20-24
[30]马金奎,路长厚,程英.机器视觉测试系统.工具技术, 2004,38(9): 129-132
[31]郑威.基于嵌入式机器视觉的缺陷检测系统硬件设计.硕士学位论文,武汉理工大学, 2006
[32] CCD&CMOS图像和机器视觉产品手册.凌云光视数字图像公司, 2005
[33]尚会超.印刷图像在线检测的算法研究与系统实现.博士学位论文,华中科技大学, 2006
[34] Mahmoud Meribout, Mamoru Nakanishi, Takeshi Ogura. Accurate and Real-time Image Processing on a New PC-compatible Board. Real-time Imaging, 2002,8(1): 35-51
[35]周太明.光源原理与设计.上海:复旦大学出版社, 2006
[36]李福建,张元培.机器视觉系统组成研究.自动化博览, 2004(2): 61-63
[37]唐向阳,张勇,李江有等.机器视觉关键技术的现状及应用展望.昆明理工大学学报(理工版), 2004,29(2): 63-93
[38]李洪宇.机器视觉图像采集设计及研究.硕士学位论文,大庆石油学院, 2007
[39]李俊.机器视觉照明光源关键技术研究.硕士学位论文,天津理工大学, 2006
[40] J Batlle, J Marti, P Ridao. A New FPGA/DSP-Based Parallel Architecture for Real-Time Image Processing. Real-Time Imaging, 2002,8(5): 345-356
[41]陈兵旗.实用数字图像处理与分析.北京:中国农业大学出版社,清华大学出版社, 2008
[42]何明一,卫保国.数字图像处理.北京:科学出版社, 2008
[43] Tinku Acharya, Ajoy K. Ray. Image processing: principles and applications.北京:清华大学出版社, 2007
[44] Rafael C. Gonzalez, Richard E. Woods. Digital image processing.北京:电子工业出版社, 2007
[45]许景波.高斯滤波器逼近理论与应用研究.博士学位论文,哈尔滨工业大学, 2007
[46]余松煜,周源华,张瑞.数字图像处理.上海:上海交通大学出版社, 2007
[47]张健.一种图像分割关键技术的研究和实现.硕士学位论文,北京邮电大学, 2008
[48]张新峰,沈兰荪.模式识别及其在图像处理中的应用.测控技术, 2004,23(5): 28-32
[49]范立南,韩晓微,张广渊.图像处理与模式识别.北京:科学出版社, 2007
[50] Sergios Theodoridis, Konstantinos Koutroumbas. Pattern recognition.北京:电子工业出版社, 2004
[51]钟珞,潘昊,封筠等.模式识别.武汉:武汉大学出版社, 2006
[52] Linda G. Shapiro, George C. Stockman. Computer vision.北京:机械工业出版社, 2005
[53] David A. Forsyth, Jean Ponce. Computer vision: a modern approach.北京:电子工业出版社, 2004
[54]曹玉萍,田学民.基于典型变量分析状态残差的故障检测方法.控制工程, 2007,7(14): 74-76
[55]吴换妮.基于支持向量机的多类模式识别模型.硕士学位论文,重庆大学, 2007
[56]袁亚湘.非线性优化计算方法.北京:科学出版社, 2008
[57]王言金.最优化的不可行内点算法研究.博士学位论文,武汉大学, 2004
[58]雍龙泉.二次规划的算法研究.硕士学位论文,西安电子科技大学, 2005
[59] Hu Fangxia, He Xin, Niu Tiangui. Study on the key image processing technology in the inspection of packing quality for small-pack cigarettes. Computer Science, 207,62: 67-71