Terrorists conceal highly energetic materials (HEM) as Improvised Explosive Devices (IED) in various types of materials such as PVC, wood, Teflon, aluminum, acrylic, carton and rubber to disguise them from detection equipment used by military and security agency personnel. Infrared emissions (IREs) of substrates, with and without HEM, were measured to generate models for detection and discrimination. Multivariable analysis techniques such as principal component analysis (PCA), soft independent modeling by class analogy (SIMCA), partial least squares-discriminant analysis (PLS-DA), support vector machine (SVM) and neural networks (NN) were employed to generate models, in which the emission of IR light from heated samples was stimulated using a CO2 laser giving rise to laser induced thermal emission (LITE) of HEMs. Traces of a specific target threat chemical explosive: PETN in surface concentrations of 10 to 300 ug/cm2 were studied on the surfaces mentioned. Custom built experimental setup used a CO2 laser as a heating source positioned with a telescope, where a minimal loss in reflective optics was reported, for the Mid-IR at a distance of 4 m and 32 scans at 10 s. SVM-DA resulted in the best statistical technique for a discrimination performance of 97%. PLS-DA accurately predicted over 94% and NN 88%.