A novel preprocessing method for vehicular-mounted forward-looking UWB SAR system

The forward-looking ground penetrating radar (FLGPR) has been widely considered as an applicable landmines detection technology, but the self-signature and the strong ground clutter will severely degrade the SAR image. We present a novel method for interference and ground clutter suppression in the image domain based on the distance-band partition and iterative operation techniques, which can realize the parallel processing and adaptive update of the parameters on line. By using the property of FLGPR systems that the same spatial region is observed several times at different standoff distances, we exploit a range multi-look processing based on the iterative technique to obtain the high quality image. The proposed method in this paper is adaptive and efficient enough to work in a real-time operational scenario, so it can effectively meet the needs of the preprocessing in the vehicular-mounted forward-looking (VMFL) detection system. The effectiveness of our approaches is demonstrated with an experimental data set.