[1]刘航铖,严小军,惠宏超.基于单目视觉的轴孔零件定位系统设计[J].机械与电子,2021,(06):70-75.
　LIU Hangcheng,YAN Xiaojun,HUI Hongchao.Design of Axis Hole Parts Positioning System Based on Monocular Vision[J].Machinery & Electronics,2021,(06):70-75.
点击复制

基于单目视觉的轴孔零件定位系统设计()
分享到：

机械与电子[ISSN:1001-2257/CN:52-1052/TH]

卷:
期数:
2021年06期
页码:
70-75
栏目:
智能工程
出版日期:
2021-06-23

文章信息/Info

Title:
Design of Axis Hole Parts Positioning System Based on Monocular Vision
文章编号:
:１００１-２２５７ ( ２０２１ ) ０６-００７０-０６
作者:
刘航铖; 严小军; 惠宏超
北京航天控制仪器研究所,北京 １０００３９
Author(s):
LIU Hangcheng ;  YAN Xiaojun ;  HUI Hongchao
( Beijing Institute of Aerospace Control Devices , Beijing １０００３９ , China )
关键词:
单目视觉; 零件定位; 椭圆拟合; 二义性
Keywords:
monocular vision ;  part positioning ;  ellipse fitting ;  duality
分类号:
TP２４９ ; TP３９１.４１
文献标志码:
A
摘要:
针对轴孔零件精密装配过程中单目视觉定位系统的使用困难,设计了一种基于单目视觉的轴孔零件定位系统,提出了基于图像特征的零件倾斜角计算方法,可直接判定零件倾斜姿态.通过对椭圆快速检测算法进行优化,减小噪声影响,实现了高精度检测.最终通过该系统实现了轴孔零件的准确定位,提高了轴孔零件自动装配的成功率.实验证明,提出的方法能够在单目视觉条件下准确得到待装配零件的空间位置信息,满足同轴度要求 １５ μ m 级的轴孔精密装配需求,具有较好的工程应用价值.
Abstract:
Aiming at the difficulty of using monocular vision positioning system in the process of precision assembly of shaft hole parts , a positioning system of shaft hole parts based on monocular vision is designed , and a method of calculating the inclination angle of parts based on image features is proposed ,which can directly determine the inclination attitude of parts．By optimizing the fast ellipse detection algorithm , the influence of noise is reduced , and the high-precision detection is realized．Finally , through the system , the accurate positioning of the shaft hole parts is realized , and the success rate of the automatic assembly of the shaft hole parts is improved．The experimental results show that the proposed method can accurately obtain the spatial position information of the parts to be assembled under the condition of monocular vision , and meet the requirement of precision assembly of shaft hole with coaxiality of １５ μ m ,which has good engineering application value．

参考文献/References:

[１] 朱博文,马立,陈韦男,等 ． 自动轴孔装配系统的现状与关键技术[J] ． 现代制造工程, ２０１９ ( ５ ): １５６-１６１．

[２] THOMAS G , KAYE J E , JAYAS R , et al．Cylinder rotational orientation based on circle detection [ C ]// International Conference Image Analysis and Recognition , ２００５ : ４９９-５０４．
[３] 魏东辰,严小军,惠宏超 ． 基于目标纹理的机器视觉转角检测方法[J] ． 导航与控制, ２０２０ , １９ ( ３ ): ８６-９４．
[４] 陈至坤,胡连军,王一,等 ． 基于角度约束的目标位姿测量二 义 性 剔 除 方 法 [ J ] ． 应 用 光 学, ２０１８ , ３９ (１ ):１０７-１１１．
[５] 王婷婷,李戈,赵杰,等 ． 基于双目视觉的运动目标检测跟踪与定位[J] ． 机械与电子, ２０１５ ( ６ ): ７３-７６．
[６] 陈彦宇,田东庄,许翠华,等 ． 基于双目视觉的钻杆接头定位系统设计[J] ． 组合机床与自动化加工技术,２０２０(２ ): １４３-１４４ , １６０．
[７] 张李俊,黄学祥,冯渭春,等 ． 消除圆位姿二义性的三线构型法[J] ． 光学学报, ２０１６ , ３６ ( ５ ): １１２-１２０．
[８] 张磊,徐科军,赵锐,等 ． 一种基于圆特征的单目视觉位姿测量算法的改进[J] ． 合肥工业大学学报(自然科学版),２００９ , ３２ ( １１ ): １６６９-１６７３．
[９] 唐永龙 ． 微器件高精度装配精确定位关键技术研究[D] ． 北京:北京理工大学,２０１４．
[１０] SUZUKI A , ABE K．Topological structural analysis of digitized binary images by border following [ J ] ．Computer vision , graphics , and image processing , １９８５ ,３０(１ ): ３２-４６．
[１１] FITZGIBBON A , PILU M , FISHER R B．Direct least square fitting of ellipses [ J ] ．IEEE Transactions on pattern analysis and machine intelligence , １９９９ , ２１(５ ): ４７６-４８０．
[１２] CHEN Z , LIAO H Y , ZHANG X M．Telecentric stereo micro-vision system : calibration method and experiments [ J ] ．Optics and lasers in engineering ,２０１４ , ５７ : ８２-９２．

备注/Memo

备注/Memo:
收稿日期: ２０２１-０２-０５
基金项目:装备预研航天科技联合基金( ６１４１B０６１４０２ )
作者简介:刘航铖 ( １９９６ ),男,四川成都人,硕士研究生,研究方向为机器视觉检测及智能制造等,通信作者;严小军 ( １９７２ ),男,江苏无锡人,研究员,博士,研究方向为仪表工艺.
更新日期/Last Update: 2021-06-22