罗德里格矩阵在相机外参数初值求解中的应用

Application of Lodrigues matrix in solving the initial value of camera exterior orientation parameters

为了解决近景摄影测量中因方程的非线性特征及大角度现象所导致的外参数初值解算困难或精度不高的问题,本文引入了罗德里格矩阵,主要利用罗德里格矩阵根据角锥法所求解的公共点来解算相机的外参数初值。此方法不仅简化了其解算过程,避免了由角度函数引起的非线性特征,而且减弱了大角度对解算精度的影响。实验证明,罗德里格矩阵结合角锥法解算出的外方位线元素初值精度误差在0~10mm之间,角元素初值精度误差在0~0.3°之间,足够用于外参数精确值的解算,具有较高的实用价值。

In order to solve the problem that the initial value of external parameters are hard to be calculated or have a low precision because of the equation＇ s nonlinear characteristics and large-angle phenomenon, Lodrigues matrix has been introduced to calculate the initial values of external parameters. This approach not only simplifies the process of solution, avoided the nonlinearity caused by the angle function, but also reduces the influence of the large angle on the accuracy of solution. Experiments show that the precision of exterior orientation initial value is between 0 ~ lOmm, and the precision of angle initial value is between 0 - O. 3degree, which is accurate enough for solving exact value of the external parameters and has high practical value.