RPC校正方法研究  

Research on the approach to RPC emendation

作　　者：刘江[1] 岳庆兴[2] 邱振戈[2] 

机构地区：[1]四川省第三测绘工程院,成都610500 [2]国家测绘地理信息局卫星测绘应用中心,北京100830

出　　处：《国土资源遥感》2013年第1期61-65,共5页Remote Sensing for Land & Resources

摘　　要：传统校正RPC误差一般采用地面坐标经RPC投影到图像上的像点和对应的测量像点间的多项式变换进行,但这样做就增加了参数的数目;而参数的个数随多项式变换形式的不同也不固定,又增加了RPC利用的难度。本文认为RPC本身包含的多项式参数与附加多项式参数的性质类似,提出通过直接对部分RPC进行校正的方法来消除RPC的误差。以IKONOS卫星图像为例的实验证明,在不增加附加参数的前提下,使用本文方法可以取得不低于传统方法的较高定位精度。The conventional way to emend errors of rational polynomial coefficients （RPC） is carried out by a polynomial transform between metrical image point coordinates and projective point coordinates from ground points, but in such a performance, the number of parameters will increase, and the number of parameters changes along with the form of the polynomial, which increases the utilization difficulties. In this paper, the authors consider that the nature of RPC is analogous to that of the attached polynomial parameters, and therefore propose that the errors of RPC parameters can be eliminated by emending precision can be attained without using additional partial RPC. The experimental results show that the ideal point parameters, and the precision is better than that of the traditionalmeans.

关 键 词：有理多项式系数(RPC) IKONOS 校正 参数选取 精度 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]