东北大学学报(自然科学版) ›› 2010, Vol. 31 ›› Issue (3): 330-333.DOI: -

• 论著 • 上一篇    下一篇

基于控制点鲁棒提取的视觉成像畸变校正方法

刘军;胡振伟;张达;谢植;   

  1. 东北大学信息科学与工程学院;
  • 收稿日期:2013-06-20 修回日期:2013-06-20 发布日期:2013-06-20
  • 通讯作者: -
  • 作者简介:-
  • 基金资助:
    国家高技术研究发展计划项目(2007AA04Z194)

Anamorphic visual image rectification method based on robust pick-up of control points

Liu, Jun (1); Hu, Zhen-Wei (1); Zhang, Da (1); Xie, Zhi (1)   

  1. (1) School of Information Science and Engineering, Northeastern University, Shenyang 110004, China
  • Received:2013-06-20 Revised:2013-06-20 Published:2013-06-20
  • Contact: Liu, J.
  • About author:-
  • Supported by:
    -

摘要: 基于控制点的摄像机视觉成像畸变校正过程中,控制点自动提取困难且误差较大,因此难以实现动态图像的在线准确校正.针对目前控制点提取校正方法的局限性,提出了基于鲁棒性好的控制点匹配提取校正方法.该方法在实现控制点的自动、精确、鲁棒提取的基础上,建立畸变图像与原始图像的最优次数多项式映射模型,并结合双线性插值运算对图像进行校正.该校正算法应用结果表明:该方法在图像识别系统中实现了快速、精确的图像畸变校正,校正误差为0.037像素.

关键词: 控制点, 图像畸变, 匹配, 校正

Abstract: It is hard to rectify accurately the dynamic images online because of the big errors resulting from the difficult automatic pick-up of control points in the rectification process of unavoidable anamorphic images by camera. Taking account of the limitation of existing pick-up methods, an approach to well-matched high robust pick-up of control points was proposed for the rectification, which can implement the automatic, accurate and robust pick-up of control points through developing a polynomial mapping model with optimum polynomial degree for both anamorphic and original images, thus rectifying those images via bilinear interpolation. The application results show that the method can rectify the image quickly and accurately. It can yield accurate correction results of which error is only 0.037 pixel.

中图分类号: