Spaceborne SAR Image Simulation And Its Application In Orthographic Correction

As a promising remote sensing technique, Synthetic Aperture Radar (SAR) has the advantages of all-weather, full-time, high resolution and accuracy. SAR technology is widely applied in survey and mapping, disaster monitoring, reconnaissance and so on.The purpose of SAR image simulation is to create a simulated image similar to the real SAR image in terms of geometric and radiometric characteristics. SAR image simulation technology plays a key role in the design of SAR systems, detection and recognition of target, and especially for SAR images orthorectification, etc. The inputs of SAR image simulation are only SAR system parameters and DEM of the area to be illuminated. For SAR image simulation technology, the Ground Control Points(GCPs) are not necessary for the orthorectification of SAR images. Compared with geocoding method, it’s more simple and efficient. With the rapid development of spaceborne SAR, the needs for SAR image simulation technology will become increasingly urgent.In view of this situation, we analyzed the principle of SAR and summarized some methods to derived necessary parameters from the metadata of SAR image. Moreover, we offered and made necessary improvements in the detailed method for SAR image simulation. On the basis of that, we performed SAR image simulation by using SAR system parameters, range-doppler model, Muhleman model and ASTER DEM. The simulation image indicates extremely similar geometric characteristics to the corresponding real SAR image. The experimental results have proved the efficiency and reliability of the applied method.Furthermore, based on our SAR image simulation method, we performed the orthorectification of SAR data and obtained satisfactory results. The geometric distortions including foreshortening and layover were effectively eliminated. There is a strong correlation between ortho-image and hill-shading map. Finally, we evaluated the RMSE between the feature points within ortho-image and the counterparts of hill-shading map.