基于虚拟双目视觉的高速摄像机自动标定系统
|
摘要
虚拟双目视觉测试是基于一台高速摄像机和两面平面镜实现物体的三维重建,重点需解决传统双目视觉测试存在的同步性和成本高的问题。摄像机内外部参数的标定是测试精度的重要保证,为解决标定过程需要人工干预、处理过程繁琐等问题,基于虚拟双目视觉原理和张正友标定法,设计了高速摄像机参数自动标定系统,通过外部触发电路控制安装在云台支架上的标定板自动转动,实现不同角度标定板图像的自动拍摄,通过OpenCV计算机视觉库对图像进行自动处理,得到高速摄像机的内外部参数,最后验证了高速摄像机自动标定系统的准确性和实用性,为后续三维重建和动态三维测试提供了基础。
The virtual binocular vision test is based on a high-speed camera and two planar mirrors to realize the 3 D reconstruction of objects, and it solves the problems of synchronization and high cost in traditional binocular vision test. The internal and external parameters of camera calibration are an important guarantee of test precision. In order to solve the issue of human intervention and the complicated process during the calibration, a kind of automatic calibration system of high-speed camera parameter is designed based on the virtual binocular vision and Zhang's calibration method. The external trigger circuit is designed to control the automatic rotation of the calibration board mounted on the platform bracket so as to realize automatic image shooting of different angle calibration board. The internal and external parameters of the high speed camera are obtained through the automatic processing of images by the open source computer vision library. Finally, the accuracy and practicability of the automatic calibration system of high speed camera are verified, It could provide the basis for subsequent 3 D reconstruction and dynamic 3 D testing.
The virtual binocular vision test is based on a high-speed camera and two planar mirrors to realize the 3 D reconstruction of objects, and it solves the problems of synchronization and high cost in traditional binocular vision test. The internal and external parameters of camera calibration are an important guarantee of test precision. In order to solve the issue of human intervention and the complicated process during the calibration, a kind of automatic calibration system of high-speed camera parameter is designed based on the virtual binocular vision and Zhang's calibration method. The external trigger circuit is designed to control the automatic rotation of the calibration board mounted on the platform bracket so as to realize automatic image shooting of different angle calibration board. The internal and external parameters of the high speed camera are obtained through the automatic processing of images by the open source computer vision library. Finally, the accuracy and practicability of the automatic calibration system of high speed camera are verified, It could provide the basis for subsequent 3 D reconstruction and dynamic 3 D testing.
引文
[1] 韩云生.基于双目立体视觉的移动机器人目标定位[D].无锡:江南大学,2009.
[2] 沈彤,刘文波,王京.基于双目立体视觉的目标测距系统[J].电子测量技术,2015,38(4):52-54.
[3] 王耀东,余祖俊,石井抱.高速立体视觉与机械运动异常动作三维检测[J].仪器仪表学报,2013,34(10):2161-2169.
[4] 李鹏飞,田瑜,张蕾,等.双目视觉运动目标跟踪定位系统的设计[J].测控技术,2018,37(4):72-77.
[5] 林碧华,张开淋,刘向军.基于虚拟双目视觉的电器电弧弧根三维运动特性的研究方法[J].电器与能效管理技术,2017(23):1-5,22.
[6] 邱才元,许雄,刘向军.中压断路器触头系统的三维动态测试[J].中国测试,2016,42(8):33-38.
[7] 陈德为,庄煜祺,张培铭,等.基于单目视觉技术的智能交流接触器三维动态测试与分析技术[J].中国电机工程学报,2014,34(12):1931-1937.
[8] 徐芳, 刘晶红, 王宣. 基于标定的CCD图像畸变校正方法研究[J]. 液晶与显示, 2013, 28(4):633-640.
[9] 胡小平,谢珂,彭涛.一种新的双目固定式机器人三维视觉定位方法[J].光电工程,2014,41(8):10-15.
[10] 许会, 洪峰, 张智福, 等. 一种鱼眼相机快速标定装置及标定方法[P]. 湖南:CN107610185A, 2018-01-19.
[11] 王平, 薛开, 何波贤, 等. 一种基于OpenCV摄像头标定的3D输入方法[P]. 黑龙江:CN103399652A, 2013-11-20.
[12] 高如新,朱烜甫.基于ORB算法的双目视觉定位[J].电子测量技术,2017,40(4):142-145.
[13] 徐嵩,孙秀霞,刘树光,等.摄像机畸变标定的模型参考逼近方法[J].光学学报,2013,33(7):231-242.
[14] 涂波, 刘璐, 刘一会, 等. 一种扩展小孔成像模型的鱼眼相机矫正与标定方法[J]. 动化学报, 2014, 40(4):653-659.
[15] KAEHLER A, BRADSKI G. Learning OpenCV 3: Computer Vision in C++ with the OpenCV Library[M]. O′Reilly Media, Inc, 2016.
[16] 毛星云. OpenCV3编程入门[M]. 北京:电子工业出版社, 2015.
[2] 沈彤,刘文波,王京.基于双目立体视觉的目标测距系统[J].电子测量技术,2015,38(4):52-54.
[3] 王耀东,余祖俊,石井抱.高速立体视觉与机械运动异常动作三维检测[J].仪器仪表学报,2013,34(10):2161-2169.
[4] 李鹏飞,田瑜,张蕾,等.双目视觉运动目标跟踪定位系统的设计[J].测控技术,2018,37(4):72-77.
[5] 林碧华,张开淋,刘向军.基于虚拟双目视觉的电器电弧弧根三维运动特性的研究方法[J].电器与能效管理技术,2017(23):1-5,22.
[6] 邱才元,许雄,刘向军.中压断路器触头系统的三维动态测试[J].中国测试,2016,42(8):33-38.
[7] 陈德为,庄煜祺,张培铭,等.基于单目视觉技术的智能交流接触器三维动态测试与分析技术[J].中国电机工程学报,2014,34(12):1931-1937.
[8] 徐芳, 刘晶红, 王宣. 基于标定的CCD图像畸变校正方法研究[J]. 液晶与显示, 2013, 28(4):633-640.
[9] 胡小平,谢珂,彭涛.一种新的双目固定式机器人三维视觉定位方法[J].光电工程,2014,41(8):10-15.
[10] 许会, 洪峰, 张智福, 等. 一种鱼眼相机快速标定装置及标定方法[P]. 湖南:CN107610185A, 2018-01-19.
[11] 王平, 薛开, 何波贤, 等. 一种基于OpenCV摄像头标定的3D输入方法[P]. 黑龙江:CN103399652A, 2013-11-20.
[12] 高如新,朱烜甫.基于ORB算法的双目视觉定位[J].电子测量技术,2017,40(4):142-145.
[13] 徐嵩,孙秀霞,刘树光,等.摄像机畸变标定的模型参考逼近方法[J].光学学报,2013,33(7):231-242.
[14] 涂波, 刘璐, 刘一会, 等. 一种扩展小孔成像模型的鱼眼相机矫正与标定方法[J]. 动化学报, 2014, 40(4):653-659.
[15] KAEHLER A, BRADSKI G. Learning OpenCV 3: Computer Vision in C++ with the OpenCV Library[M]. O′Reilly Media, Inc, 2016.
[16] 毛星云. OpenCV3编程入门[M]. 北京:电子工业出版社, 2015.