Comparison of Clutter Reduction Methods for Buried Object Detection in Heterogeneous Soil

Abstract

A GPR was conducted in this study using a Vector Network Analyzer (VNA) connected to two antennas to detect buried metals. The target was located 17 cm below the ground surface. The detection process was carried out to obtain radar images, namely A-scan and B-scan. Clutter reduction was performed using three methods: weighting, averaging, and singular value decomposition (SVD). This study reviews the performance of clutter-reduction methods under heterogeneous soil conditions. Qualitatively, all methods clarified the target and reduced clutter in the radar image. Quantitatively, the Signal-to-Clutter Ratio (SCR) is calculated after clutter reduction. In terms of results, the weighting method increased the SCR to 28.81 dB, while the averaging method increased it to 25.31 dB. Meanwhile, the SVD method only provided a small increase to 4.45 dB.