利用直线特征控制优化RPC模型

Optimizing RPC model Using Linear Feature Control

本文在常规RPC系统误差补偿模型的基础上，依据空间直线在卫星影像上的投影仍然是直线的投影特点，通过对直线特征的参数化表达，推导出了基于直线控制的RPC模型系统误差的物方和像方优化模型，并利用ZY-3卫星影像进行了对比实验。结果表明，利用直线特征对RPC模型进行精度优化是可行的。在利用1个控制点计算平移系数的情况下，当控制线数量达到4条时，优化后RPC模型的定位精度可以达到与同等数量控制点相当的水平。

According to the principle of RPC model error compensation based on ground control point and the projection characteristic that straight line feature is still a straight line in satellite image, this paper deduces RPC model error compensation based on hnear features in image space and object space by parametric expression of linear features, and carry out experiments by ZY - 3 satellite images. The results show that it is feasible by using the straight line features for accuracy optimization of the RPC model. In the case of using only 1 GCP to calculate translation coe^cients and if the number of GCLs reaches 4, the accuracy can achieve the same level as with equal number of GCPs.