Independent factor analysis for clutter reduction in GPR data for landmine detection

This paper illustrates clutter reduction in stepped-frequency ground penetrating radar (SFGPR) data for anti-personal landmines detection. Ground reflected clutter is often a performance limiting factor in ground-penetrating radar (GPR) detection of near-surface targets including antipersonnel mines. In this paper the independent factor analysis (IFA), as a method for blind source separation, is used to separate the clutter and the target signals. The algorithm has been experimentally evaluated and compared with factor analysis algorithm using different non-metallic AP landmines with different depths. The experimental data have been collected by using frequency stepped GPR system covering two ranges from 1.5GHz to 4GHz and from range 1GHz to 20GHz, from laboratory measurements.