阵地地貌反求测量中的相机标定

摘要

To effectively realize the camera calibration on a position topography measurement,a multi-camera parallel calibration method is proposed based on a camera intrinsic parameter model.According to an active vision theory,four cameras are installed at four linear independence positions on a measuring support,and a parallel device is applied to drive these cameras for shooting at the same time,so as to accomplish the shooting effect of three linear independence translations.The calibration experiments of position topography measurement are designed based on this calibration method,the experiments show that the relative error of the scale factor ratio in two directions is less than 3.87％,the absolute error of central point a coordinate is less than 4 pixels,and the variation range of distortion factor is in 3 times.This method is suitable for position topography measurement for its convenient and fast calibration.%针对阵地地貌测量重构中的相机参数标定问题,基于相机的内参数模型提出了一种多相机并联阵地地貌测量标定方法.该方法基于主动视觉原理,利用测量支架在线性独立位置安装4部相机,并采用并联装置驱动相机同时拍摄,从而实现与三次独立线性平移效果相同的拍摄.根据该标定方法设计了阵地地貌测量标定实验,实验结果表明两个方向尺度因子比值的相对误差＜3.87%,主点坐标绝对误差在4个像素的范围内线性,畸变因子的变化范围在3倍以内,该方法方便、快捷,适合于阵地地貌测量的标定.