A new RPCS refining method and application in geopositioning of QuickBird stereo imagery

Due to the independence of sensor or camera specifics when used to describe rigorous geometric model, the rational functional model has been widely used in photogrammetry for high resolution satellite imagery (HRSI). Some HRSI commercial companies such as Space Imaging and Digital Globe have provide RFM parameters (RPCs) to the end users as the standard auxiliary data attaching to the image product. This paper first briefly introduces the mathematical backgrounds and 3-D extraction applications of the RPCs. Then two main kinds of RPCs refining method, namely, the direct method and indirect method are reviewed and discussed. On the base of it, a new refining method which comprehensively combing the characteristics of past methods is addressed. Besides the initial RPCs and a certain number of ground control points (GCPs), the new method does not require any specific information so it is very convenient for the end users to apply. Finally, a pair of QuickBird stereo imagery has been tested. The test results fully demonstrate the feasibility of the new method.