Investigation of 3D Polynomial Model for the Geometric Correction of GeoEye Imagery

GeoEye satellite has provided the world with high spatial resolution 0.50 meter panchromatic images. Not only the high spatial resolution, but also the high spectral, radiometric and temporal resolutions of GeoEye imagery make it ideally suited for mapping applications. Currently, the empirical sensor models are traditionally employed, instead of the physical models, to establish the mathematical relationships between the image space and the ground space. In this study, the geometric accuracy of pan GeoEye images was determined using three dimensional (3D) polynomial model. The obtained planimetric accuracy was compared to that obtained using the rational function model (RFM). The implementation of the 3D polynomial model (PM) experiments was performed using only the new standalone software EMAN, since the 3D PMs are not available in other software packages such as ERDAS imagine or PCI from Geomatica. However, both EMAN and PCI software were used to carry out the experiments for the RFMs. A GeoEye panchromatic image covering the city of Tanta, El Gharbiya, Egypt was used in this study. The results revealed that the 3D polynomial models have the capability for geometrically correcting GeoEye images since it provides competitive geometric accuracies, compared to those obtained using 3D RFMs.