论文部分内容阅读
在大视场交通监控摄像机的标定中,由于不便于放置大尺寸标定靶标,常采用交通标线构成的特定几何图形替代标定靶标,但许多交通场景仅有一组平行的车道分界交通标线,难以获得传统标定方法所需的标定约束条件。针对此类交通场景,选取两平行交通标线和其中一条线上的三点作为标定参考物,依靠其中一条线上三点在单幅图像中的像素坐标、三点间的两个距离值及平行线间距作为约束条件来完成摄像机的标定。建立了图像坐标系与世界坐标系之间新的换算关系,推导出了摄像机内外参数的求解公式,探讨了旋转角对测量误差的影响。实验结果表明,在约束条件不足的情况下能够获得较好的摄像机参数近似解,沿道路方向上的测距精度优于对比文献的结果,具有标定参考物选取容易、操作简便、几何约束条件少等优点,适用于交通监控系统中大视场摄像机的快速标定场合。
In the calibration of large-field traffic surveillance cameras, because of the inconvenience of placing large-size calibration targets, specific geometric patterns composed of traffic marking lines are often used instead of the calibration targets. However, many traffic scenarios have only one set of parallel traffic lane marking, which is difficult Get the calibration constraints required by traditional calibration methods. For such traffic scenarios, two parallel traffic lines and three points on one of the lines are selected as the calibration reference. Based on the pixel coordinates of one of the three lines in a single image, two distance values between three points and Parallel line spacing as a constraint to complete the camera calibration. A new conversion relation between image coordinate system and world coordinate system is established. The formula for solving the internal and external parameters of the camera is deduced. The influence of the rotation angle on the measurement error is discussed. The experimental results show that the approximate solution to the camera parameters can be obtained under the condition of insufficient constraint. The accuracy of ranging along the road is better than that of the comparative literature. The calibration reference has the advantages of easy selection, simple operation and less geometric constraints And other advantages, for traffic monitoring system in large field cameras fast calibration occasions.