Detection of residues from explosive manipulation by near infrared hyperspectral imaging: a promising forensic tool

Forensic Sci Int. 2014 Sep:242:228-235. doi: 10.1016/j.forsciint.2014.06.023. Epub 2014 Jun 30.

Abstract

In this study near infrared hyperspectral imaging (NIR-HSI) is used to provide a fast, non-contact, non-invasive and non-destructive method for the analysis of explosive residues on human handprints. Volunteers manipulated individually each of these explosives and after deposited their handprints on plastic sheets. For this purpose, classical explosives, potentially used as part of improvised explosive devices (IEDs) as ammonium nitrate, blackpowder, single- and double-base smokeless gunpowders and dynamite were studied. A partial-least squares discriminant analysis (PLS-DA) model was built to detect and classify the presence of explosive residues in handprints. High levels of sensitivity and specificity for the PLS-DA classification model created to identify ammonium nitrate, blackpowder, single- and double-base smokeless gunpowders and dynamite residues were obtained, allowing the development of a preliminary library and facilitating the direct and in situ detection of explosives by NIR-HSI. Consequently, this technique is showed as a promising forensic tool for the detection of explosive residues and other related samples.

Keywords: Explosive residues; Handprint; Hyperspectral imaging; IED; Near infrared; PLS-DA.

Publication types

  • Research Support, Non-U.S. Gov't