基于极坐标的摄像机标定

Camera calibration based on polar coordinate

为了改进多摄像机标定实现的方便性和参考标定板的多样性,用极坐标设计标定板并以此来进行摄像机参数的标定.根据设计好的标定板确定世界坐标下的特征点,识别相应的特征点在图像坐标的位置,匹配特征点在图像坐标和世界坐标的位置关系后完成摄像机的标定.该标定方法只需大约20点对左右数据,同时保证标定的结果和鲁棒性;较少的点对和图像的多样性避免了遮挡和公共部分不足的问题,有利于完成多摄像机标定,较传统多摄像机标定方法在分配摄像机和标定板之间的位置关系上更加灵活.与传统的精确标定板的结果比较,在保证精度情况下,标定板设计和整个标定系统的实现则更加简单易行.

Abstract： In order to improve the flexibility of multicamera calibration and the diversity of calibra tion boards, the polar coordinate system is used to design the calibration board, and the camera cali bration is based on the board. The feature points in the world coordinate are assigned by the designed board, and the positions of the feature points in the image coordinate are identified accordingly. The camera are calibrated after mapping the positions of the feature points in the image coordinate and the world coordinate. The calibration method only requires small sets of correspondences （ about 20 pairs）, and, meanwhile, it can ensure the accuracy and robustness. Small sets of correspondences and the diversity of images can avoid the blocking and inadequate public spaces for multicamera cal ibration, which is helpful for the multicamera calibration. It is more flexible than the traditional multicamera calibration on the distribution of cameras and boards. Compared with the traditional method with accurate board, the calibration board design and the calibration System are simpler to achieve and they can also ensure the accuracy.