Fast SCFBP Algorithm for GPR-SAR Imaging Integrated With Underground Medium Permittivity Inversion

Due to the complex detection environment of ground-penetrating radar (GPR), time-domain algorithm (TDA) can accurately accommodate the wave path variations in layered medium, which presents significant potentials to achieve high focusing quality for GPR-synthetic aperture radar (GPR-SAR) imaging. However, the current TDAs will involve huge interpolation operations and bring redundant computational burden in the processing, which will inevitably degrade the algorithm performance. In this article, a fast subsurface Cartesian factorized back-projection (SCFBP) algorithm is proposed for GPR-SAR imaging, where the sub-image spectrum properties of the layered medium are investigated particularly and a two-step spectrum compression is developed according to the permittivity. As the sub-image spectrums are effectively aligned and compressed, only a very low Nyquist sampling rate (NSR) is required for the noninterpolated sub-image merging process, which will dramatically decrease the computational burden to achieve high performance both in efficiency and accuracy. Moreover, the inaccurate permittivity of the underground medium is particularly considered in GPR-SAR applications. By establishing the mapping relationship between the inaccurate permittivity of the underground medium and the introduced phase error in the imaging process, a fast echo-based compensation is developed with SCFBP, which can effectively eliminate the effects of the inaccurate permittivity on the imaging quality and accurately invert the permittivity of the underground medium for subsequent medium classification and identification. Promising results from both simulation and raw data experiments are presented and analyzed to validate the performance advantages of the proposed algorithm.