Stand-off explosive detection on surfaces using multispectral MIR-imaging

We present a system for the stand-off detection of solid explosive traces and precursors on surfaces. The system consists of a widely tunable quantum cascade laser (QCL) and a thermal imaging camera. The external cavity quantum cascade laser (EC-QCL) illuminates the surface of a distant object at different characteristic wavelengths. In synchronisation with the camera a hyperspectral data cube of the backscattered radiation is generated allowing a multivariate analysis of the scene. We demonstrate how multidimensional image processing is used in order to fast and sensitively detect traces of hazardous substances such as trinitrotoluene (TNT) or pentaerythritol tetranitrate (PETN). The recognition algorithm is developed to effectively suppress false alarms. Experiments are performed on real world like surfaces such as standard car paint, synthetic cloth or jeans fabric.