A Novel Geometric Calibration Method of Spaceborne SAR for Rugged Terrain: A Case Study of LuTan-1 Images

Abstract

The geometric positioning accuracy of synthetic aperture radar (SAR) images is affected by the radar system measurement and processing parameters, platform ephemeris data and auxiliary DEM data. Geometric calibration can compensate for the SAR systematic parameter errors to produce high-precision SAR products. SAR images of mountainous areas are difficult to be used for geometric calibration due to severe geometric distortion and difficulty in deploying and extracting control points. In this article, a novel and effective geometric calibration method for rugged terrain is proposed. The proposed method exploits the unique geometric and radiometric distortion features of SAR images. Based on the energy variation of SAR orthophoto, the calibration of range and azimuth parameters is achieved by searching for the fast time and slow time offsets that can compensate for the systematic errors. Experiments based on real SAR images from LuTan-1 satellite were conducted. The proposed method shows good performance in the experiments. After geometric calibration, the root-mean-square error of LuTan-1 L1A images is reduced from 80.77 to 6.39 m, which verifies that the proposed method can improve the positioning accuracy of SAR images. The proposed method is simple to realize and does not rely on additional control point data, which provides a new solution for the maintenance and parameter calibration of SAR satellites.

Keywords

• Energy maximization
• geometric calibration
• LuTan-1
• orthophoto
• pixel intensity
• rugged terrain