钕铁硼小型工件表面缺陷自动检测系统
|
摘要
为了使钕铁硼工件的生产满足现代化制造业高效率、高质量、自动化与智能化的要求,钕铁硼小型工件表面缺陷自动检测技术已成为钕铁硼企业控制和提高产品质量的重要技术手段。本文在对国内外各种领域已有的表面缺陷检测方法研究的基础上,针对本课题不同金属镀层工件表面强反射程度不同和工件表面缺陷种类繁多、位置随机等的具体情况和客户提出的具体要求,结合表面检测技术发展并在技术上力图创新及结果上力图经济实用的目的,确立了钕铁硼小型工件表面缺陷自动检测系统整体方案。
论文的主要研究内容有:
1.根据用户提出的要求和工件本身强反射的特点,对比了其他检测方法后,设计了以黑白面阵CCD采集实现对钕铁硼小型工件表面缺陷进行自动检测的检测系统方案,实现了双面、高速、高分辨率的无损检测。
2.设计了亮度均匀可调的照明光源,实现了对不同批次和不同金属镀层的工件均可获得高质量的图像,简化了图像处理的难度和提高了检测的速度和准确率。
3.设计了精巧的工件翻面机构,使得实现工件表面缺陷的双面检测机构的结构精简紧凑。
4.研究钕铁硼工件表面缺陷的特点,对缺陷图像处理的流程进行了分析。
5.设计了控制机械执行机构动作的控制系统,控制电机的工作方式和剔除各种缺陷的执行机构的运行。结合自动上料机构、工件翻面机构、剔除各种缺陷的执行机构和对合格品的自动计数等,实现了集自动上、下料、表面缺陷自动检测、缺陷自动按类剔除、合格品自动计数为一体的表面缺陷自动检测机。
In order to satisfy the demand of modernized, automated and intellectualized production for high efficiency and high quality, the automatic surface defect inspection technology of Nd-Fe-B small metal part becomes the important ways for factory to control and raise quality. There are some difficulties in this case as follows: different metal surface can cause different strong reflection ,the kinds of different detect is so many and the position of detect is random. In order to innovate surface detect inspection technology and realize economical results, after analyzing various ready inspection methods and considering specific technical requirements, we design out this automatic surface detect inspection system, which adopts computer vision technology as inspection theory by area CCD gets image.
The central content of this paper is introduced as follows:
1. According to client’s demand and considering strong reflection characteristic of metal surface, by contrast with other inspection method, we design out this automatic inspection system, which acquires image of Nd-Fe-B small metal part by area CCD. The system realizes two-side, high speed and high resolution unbroken inspection.
2. Design out illuminating light with its luminance equably changing, which helps to get high quality image for different batch and different metal surface part. This reduces the difficulty of image processing and raises the speed and preciseness of defect inspection.
3. Design out one finished device to turn part, which makes two-side inspection machine more exact and more easily.
4. Research the surface defect characteristic of Nd-Fe-B metal part and analyze the surface defect detection image processing.
5. Design out the control system, which can control executing mechanism, control motor working mode and defect part eliminating device. With those devices, this system combines so many functions into this surface defect automatic inspection system, such as automatically delivering parts, surface defect automatically inspecting, sorting parts according to different surface detect and automatically counting etc.
论文的主要研究内容有:
1.根据用户提出的要求和工件本身强反射的特点,对比了其他检测方法后,设计了以黑白面阵CCD采集实现对钕铁硼小型工件表面缺陷进行自动检测的检测系统方案,实现了双面、高速、高分辨率的无损检测。
2.设计了亮度均匀可调的照明光源,实现了对不同批次和不同金属镀层的工件均可获得高质量的图像,简化了图像处理的难度和提高了检测的速度和准确率。
3.设计了精巧的工件翻面机构,使得实现工件表面缺陷的双面检测机构的结构精简紧凑。
4.研究钕铁硼工件表面缺陷的特点,对缺陷图像处理的流程进行了分析。
5.设计了控制机械执行机构动作的控制系统,控制电机的工作方式和剔除各种缺陷的执行机构的运行。结合自动上料机构、工件翻面机构、剔除各种缺陷的执行机构和对合格品的自动计数等,实现了集自动上、下料、表面缺陷自动检测、缺陷自动按类剔除、合格品自动计数为一体的表面缺陷自动检测机。
In order to satisfy the demand of modernized, automated and intellectualized production for high efficiency and high quality, the automatic surface defect inspection technology of Nd-Fe-B small metal part becomes the important ways for factory to control and raise quality. There are some difficulties in this case as follows: different metal surface can cause different strong reflection ,the kinds of different detect is so many and the position of detect is random. In order to innovate surface detect inspection technology and realize economical results, after analyzing various ready inspection methods and considering specific technical requirements, we design out this automatic surface detect inspection system, which adopts computer vision technology as inspection theory by area CCD gets image.
The central content of this paper is introduced as follows:
1. According to client’s demand and considering strong reflection characteristic of metal surface, by contrast with other inspection method, we design out this automatic inspection system, which acquires image of Nd-Fe-B small metal part by area CCD. The system realizes two-side, high speed and high resolution unbroken inspection.
2. Design out illuminating light with its luminance equably changing, which helps to get high quality image for different batch and different metal surface part. This reduces the difficulty of image processing and raises the speed and preciseness of defect inspection.
3. Design out one finished device to turn part, which makes two-side inspection machine more exact and more easily.
4. Research the surface defect characteristic of Nd-Fe-B metal part and analyze the surface defect detection image processing.
5. Design out the control system, which can control executing mechanism, control motor working mode and defect part eliminating device. With those devices, this system combines so many functions into this surface defect automatic inspection system, such as automatically delivering parts, surface defect automatically inspecting, sorting parts according to different surface detect and automatically counting etc.
引文
[1] 陈国华,钕铁硼磁体的市场与发展,第七届中国小电机技术研讨会论文集,2002,107~109
[2] 李飞,钕铁硼永磁材料及其应用前景,中国有色金属学会第三届青年学术会议论文集,1998.9,第 8 卷,增刊 2,120-123
[3] 何叶青,高性能烧结 Nd-Fe-B 合金的成分、铸锭组织、磁场取向与磁性能关系的研究,博士学位论文,北京科技大学,2000
[4] 栾景飞,烧结钕铁硼永磁体性能的研究,博士后研究工作报告,浙江大学,2002
[5] 林 翠、李晓刚、王光雍,金属材料在污染大气环境中初期腐蚀行为和机理研究进展,腐蚀科学与防腐技术,2004.3 第 16 卷 第 2 期:89-95
[6] 赵荣椿、赵忠明、崔甦生,数字图像处理导论,西安,西北工业大学出版社,1996
[7] Ben Dawson,Simon Melikian,Application Machine Vision to Verification and Testing,SME Technical Paper,Second Quarter 2004:1-13/190
[8] Thomas F. Laurin , Machine Vision Lighting: A First-order Consideration, THE PHOTONICS HANDBOOK~(TM),2004 Book 3:269-274
[9] A.R,Future Direction in Machine Vision,Topical Meeting on Machine Vision,March 1985,WA3-1～WA3-4
[10] W.A. Perkins,INSPECTOR:A Computer Vision System which Learn to Inspect Parts,IEEE Trans,PAMI-5:584-592
[11] K. Wiltschi,A. Pinz,T. Lindberg,Automatic assessment scheme for steel quality inspection,Machine Vision and Application ,2000(12):113-128
[12] 乔建军、李占青、徐科,冷轧带钢表面缺陷自动检测的研究与应用,金属材料研究,2004.3 第 30 卷 第 1 期:32-34
[13] 倪征、苏光大、马惠敏等,玻璃瓶在线检测系统设计,计算机工程与应用,2004,第 3 期:218-220
[14] 李炜、黄心汉、王敏等,基于机器视觉的带钢表面缺陷检测系统,华中科技大学学报(自然科学版),2003.2 第 31 卷 第 2 期:72-74
[15] 石岭、王克奇、白雪冰等,板材表面缺陷检测技术,林业机械与木工设备,2005,第 33 卷 第 3 期:40-42
[16] 胡亮、段发阶、丁克勤等,钢板表面缺陷计算机视觉在线检测系统的研制,2005.2 第 40 卷 第 2 期:59-61
[17] 应义斌、景寒松、马俊福等,黄花梨品质检测机器视觉系统,农业机械学报,2000.3 第 31 卷 第 2 期:113-115
[18] 黎明、马聪、杨小芹,机械加工零件表面纹理缺陷检测,中国图像图形学报,2004.3 第 9 卷 第 3 期:318-322
[19] 赵海洋、冯心海、薛钧义等,基于机器视觉的瓷砖表面质量分析与检测,西安建筑科技大学学报,2000.3 第 32 卷 第 1 期:43-46
[20] 何滢,强反射表面缺陷检测技术,硕士学位论文,天津大学,2002
[21] 丁金明,金属镀层工件表面缺陷自动检测系统的研究,硕士学位论文,天津大学,2004
[22] Carlos Lopez,Daniel F. Garcia,Ruben Usamentiaga,etc. ,Real-time system for flatness inspection of steel strips,Machine Vision Applications in Industrial Inspection,2005 Vol.5679:228-238
[23] Dirk Sischka,Reimund Bisek,Detection of Defects on the surface of Microelectronic Structures , IEEE TRANSACTIONS ON ELECTRON DEVICES,1989,VOL.36 NO.1:8-13
[24] John C Badger,Sean T Enright.,Automated surface inspection system,Iron and Steel Engineer,1996,VOL.73 NO.3:48-51
[25] Fernandez C.,Platero C.,Campoy P.,etc.,Vision System for on-line Surface Inspection in Aluminum Casting Process,Industrial Electronics Control and Instrumentation,1993 VOL.3:1854-1859
[26] 张凤林、孙学珠,工程光学,天津,天津大学出版社,1996
[27] 徐科、徐金梧,基于图像处理的冷轧带钢表面在线检测技术,钢铁,2002.12 第 37 卷 第 12 期:61-64
[28] 王庆有,CCD 应用技术,天津,天津大学出版社,2000
[29] 孙双花,在线式光电自动计数方法及应用,光电工程,2004.3 第 31 卷 第3 期:37-40
[30] 孙双花,磁性零件在线式激光光电计数仪的设计,仪器仪表学报,2005.1 第 26 卷 第 1 期:59-62
[31] Pakkanen J.,Iivarinen J.,A novel self-organizing neural network for defect image classification , 2004 IEEE International Joint Conference Neural Networks Proceedings .,2004.7 VOL.4:2553-2556
[32] 徐科、徐金梧、梁治国等,冷轧带钢表面质量自动检测系统的在线应用研究,冶金自动化,2003,第 1 期:51-53
[33] 林茂,料斗式电磁振动供料器的理论与实验研究,轻工机械,1994,第 2期:6-9
[34] 贺相、王亚军,日本振动料斗的调查,组合机床与自动化加工技术,1994,第 11 期:17-21
[35] 章毓晋,图像工程:图像理解与计算机视觉,北京,清华大学出版社。2000
[36] 周党群、程宪平、李刚健,PCB 视觉检测系统中目标图像提取的研究,电子测量与仪器学报,2001.9 第 15 卷 第 3 期:10-13
[37] 娄小平、吕乃光、邓文怡等,基于机器视觉复杂形状工件的非接触测量方法,工具技术,2005,第 39 卷 第 7 期:82-85
[38] 赵宏伟、张小清、冯裕钊等,基于数字图像处理的啤酒瓶静态智能检测技术,机电产品开发与创新,2005.7 第 18 卷 第 4 期:88-90
[39] 岳晓锋、赵敏、栾宝剑等,计算机视觉技术在发动机缺陷检测系统中的应用,计算机测量与控制,2005,第 13 卷 第 6 期:516-518
[40] 李艳梅、吴湘济,数字图像处理技术在纺织测试分析中的应用,上海工程技术大学学报,2005.6 第 19 卷 第 2 期:189-192
[41] 杨淑莹,VC++图像处理程序设计,北京,清华大学出版社 交通大学出版社,2005
[2] 李飞,钕铁硼永磁材料及其应用前景,中国有色金属学会第三届青年学术会议论文集,1998.9,第 8 卷,增刊 2,120-123
[3] 何叶青,高性能烧结 Nd-Fe-B 合金的成分、铸锭组织、磁场取向与磁性能关系的研究,博士学位论文,北京科技大学,2000
[4] 栾景飞,烧结钕铁硼永磁体性能的研究,博士后研究工作报告,浙江大学,2002
[5] 林 翠、李晓刚、王光雍,金属材料在污染大气环境中初期腐蚀行为和机理研究进展,腐蚀科学与防腐技术,2004.3 第 16 卷 第 2 期:89-95
[6] 赵荣椿、赵忠明、崔甦生,数字图像处理导论,西安,西北工业大学出版社,1996
[7] Ben Dawson,Simon Melikian,Application Machine Vision to Verification and Testing,SME Technical Paper,Second Quarter 2004:1-13/190
[8] Thomas F. Laurin , Machine Vision Lighting: A First-order Consideration, THE PHOTONICS HANDBOOK~(TM),2004 Book 3:269-274
[9] A.R,Future Direction in Machine Vision,Topical Meeting on Machine Vision,March 1985,WA3-1～WA3-4
[10] W.A. Perkins,INSPECTOR:A Computer Vision System which Learn to Inspect Parts,IEEE Trans,PAMI-5:584-592
[11] K. Wiltschi,A. Pinz,T. Lindberg,Automatic assessment scheme for steel quality inspection,Machine Vision and Application ,2000(12):113-128
[12] 乔建军、李占青、徐科,冷轧带钢表面缺陷自动检测的研究与应用,金属材料研究,2004.3 第 30 卷 第 1 期:32-34
[13] 倪征、苏光大、马惠敏等,玻璃瓶在线检测系统设计,计算机工程与应用,2004,第 3 期:218-220
[14] 李炜、黄心汉、王敏等,基于机器视觉的带钢表面缺陷检测系统,华中科技大学学报(自然科学版),2003.2 第 31 卷 第 2 期:72-74
[15] 石岭、王克奇、白雪冰等,板材表面缺陷检测技术,林业机械与木工设备,2005,第 33 卷 第 3 期:40-42
[16] 胡亮、段发阶、丁克勤等,钢板表面缺陷计算机视觉在线检测系统的研制,2005.2 第 40 卷 第 2 期:59-61
[17] 应义斌、景寒松、马俊福等,黄花梨品质检测机器视觉系统,农业机械学报,2000.3 第 31 卷 第 2 期:113-115
[18] 黎明、马聪、杨小芹,机械加工零件表面纹理缺陷检测,中国图像图形学报,2004.3 第 9 卷 第 3 期:318-322
[19] 赵海洋、冯心海、薛钧义等,基于机器视觉的瓷砖表面质量分析与检测,西安建筑科技大学学报,2000.3 第 32 卷 第 1 期:43-46
[20] 何滢,强反射表面缺陷检测技术,硕士学位论文,天津大学,2002
[21] 丁金明,金属镀层工件表面缺陷自动检测系统的研究,硕士学位论文,天津大学,2004
[22] Carlos Lopez,Daniel F. Garcia,Ruben Usamentiaga,etc. ,Real-time system for flatness inspection of steel strips,Machine Vision Applications in Industrial Inspection,2005 Vol.5679:228-238
[23] Dirk Sischka,Reimund Bisek,Detection of Defects on the surface of Microelectronic Structures , IEEE TRANSACTIONS ON ELECTRON DEVICES,1989,VOL.36 NO.1:8-13
[24] John C Badger,Sean T Enright.,Automated surface inspection system,Iron and Steel Engineer,1996,VOL.73 NO.3:48-51
[25] Fernandez C.,Platero C.,Campoy P.,etc.,Vision System for on-line Surface Inspection in Aluminum Casting Process,Industrial Electronics Control and Instrumentation,1993 VOL.3:1854-1859
[26] 张凤林、孙学珠,工程光学,天津,天津大学出版社,1996
[27] 徐科、徐金梧,基于图像处理的冷轧带钢表面在线检测技术,钢铁,2002.12 第 37 卷 第 12 期:61-64
[28] 王庆有,CCD 应用技术,天津,天津大学出版社,2000
[29] 孙双花,在线式光电自动计数方法及应用,光电工程,2004.3 第 31 卷 第3 期:37-40
[30] 孙双花,磁性零件在线式激光光电计数仪的设计,仪器仪表学报,2005.1 第 26 卷 第 1 期:59-62
[31] Pakkanen J.,Iivarinen J.,A novel self-organizing neural network for defect image classification , 2004 IEEE International Joint Conference Neural Networks Proceedings .,2004.7 VOL.4:2553-2556
[32] 徐科、徐金梧、梁治国等,冷轧带钢表面质量自动检测系统的在线应用研究,冶金自动化,2003,第 1 期:51-53
[33] 林茂,料斗式电磁振动供料器的理论与实验研究,轻工机械,1994,第 2期:6-9
[34] 贺相、王亚军,日本振动料斗的调查,组合机床与自动化加工技术,1994,第 11 期:17-21
[35] 章毓晋,图像工程:图像理解与计算机视觉,北京,清华大学出版社。2000
[36] 周党群、程宪平、李刚健,PCB 视觉检测系统中目标图像提取的研究,电子测量与仪器学报,2001.9 第 15 卷 第 3 期:10-13
[37] 娄小平、吕乃光、邓文怡等,基于机器视觉复杂形状工件的非接触测量方法,工具技术,2005,第 39 卷 第 7 期:82-85
[38] 赵宏伟、张小清、冯裕钊等,基于数字图像处理的啤酒瓶静态智能检测技术,机电产品开发与创新,2005.7 第 18 卷 第 4 期:88-90
[39] 岳晓锋、赵敏、栾宝剑等,计算机视觉技术在发动机缺陷检测系统中的应用,计算机测量与控制,2005,第 13 卷 第 6 期:516-518
[40] 李艳梅、吴湘济,数字图像处理技术在纺织测试分析中的应用,上海工程技术大学学报,2005.6 第 19 卷 第 2 期:189-192
[41] 杨淑莹,VC++图像处理程序设计,北京,清华大学出版社 交通大学出版社,2005