Using Rational Polynomial Coefficients (RPC) to generate digital elevation models : a comparative study

Models based on Rational Polynomial Coefficients (RPC) have recently sparked considerable interest within the remote sensing community because of their simplicity and accuracy. Indeed, some commercial, high-resolution, satellite imagery data are now supplied with RPC even though they do not disclose their physical sensor model. RPC, with stereo pairs, enable full photogrammetric processing including 3-D reconstruction, generation of digital elevation models (DEMs), orthorectification, block adjustment and feature extraction. In the light of this we here present a complete methodology for generating a DEM from stereo satellite images by using rational polynomial coefficients of the imaging geometry. We also conduct a study of the accuracy and performance, in terms of generating a stereo images-based DEM using RPC within three well known software packages. Our results are evaluated using sample data that was captured by IKONOS.