Author Login Editor-in-Chief Peer Review Editor Work Office Work

Computer Engineering ›› 2007, Vol. 33 ›› Issue (17): 213-215. doi: 10.3969/j.issn.1000-3428.2007.17.073

• Artificial Intelligence and Recognition Technology • Previous Articles     Next Articles

Automatic X-corners Detection for Online Camera Calibration

HAO Ying-ming1,2, ZHU Feng1   

  1. (1. Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016; 2. Graduate School, Chinese Academy of Sciences, Bejing 100039)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-09-05 Published:2007-09-05

摄像机在线标定中的棋盘格角点自动检测方法

郝颖明1,2,朱 枫1   

  1. (1. 中国科学院沈阳自动化研究所,沈阳 110016;2. 中国科学院研究生院,北京 100039)

Abstract: This paper investigates the problem of automatic X-corners detection aiming at online camera calibration. Since X-corners are coplanar, a plane-to-plane transform is introduced and rough location of X-corners is realized. Within the area of rough location, X-corners detection operator and the energy center of the area to obtain the accurate location are utilized. This method is applicable in both whole and local illumination owing to integrating the extraction and arrangement of the corners. In addition, when some corners are unable to be detected, others can be detected and arranged correctly. The prosented method is automatic, accurate and robust in automatic online camera calibration.

Key words: camera calibration, X-corners, subpixel

摘要: 针对摄像机在线标定的特殊需要,研究棋盘格角点的自动检测问题。利用棋盘格角点共面特点,将平面到平面的变换引入到角点检测中,实现了角点的自动粗定位。在粗定位窗口内,通过角点检测算子与区域能量中心共同实现了角点精确定位。该算法将角点的检测与排序巧妙地结合起来,不仅对全局和局部光照变化都具有较强的适应性,还解决了图像中某些角点检测失败时,其他角点的正确检测与排序问题。实验结果表明,该方法不仅在鲁棒性和检测精度方面高于常用棋盘格角点检测方法,而且全部处理过程可自动实现,在摄像机在线标定中具有很高的实际应用价值。

关键词: 摄像机标定, 棋盘格角点, 亚像素

CLC Number: