Tripwire detection in SAR images using a modified radon transform

Abstract

Anti-personnel fragmentation mines and improvised explosive devices often use metallic wires-so-called tripwires-for the trigger mechanism. Depending on environment and vegetation, these wires are hardly visible. Therefore, an airborne synthetic aperture radar (SAR) was developed to assist the process of mine clearance. In this work, an algorithm for the detection and localization of tripwires in SAR images is proposed. A modified Radon transform is employed that relies on the SAR phase information. Instead of simply integrating absolute values over a straight line, the complex-valued data is analyzed in the frequency domain. The result in the transformed image is then given by the maximum amplitude within a given frequency range, causing tripwires to clearly separate from clutter. The functionality of this algorithm is successfully demonstrated by measurements of tripwires attached to a dummy mine in wet grass.