Classification of FOD Targets based on Polarimetric Characteristics

Abstract

The invasion of foreign object debris (FOD) on the airport runway seriously threatens the safety of flight, and the accurate small radar targets classification used for the hazard level discrimination is a challenge. In this paper, a classification method based on polarization information is proposed to be used for FOD targets classification. We utilize electromagnetic simulation software for obtaining radar cross section (RCS) polarimetric data in various incident angles based on typical FOD (e.g. Spanners, Screwdrivers, PVC Pipe) targets. In addition, polarization invariant is applied to characterize targets for classification using support vector machines (SVM). Experimental results indicate that this method can classify the three objects effectively and the great potential of polarization information for FOD targets classification.