ARTIFICIAL NEURAL NETWORK ANALYSIS IN INTERFEROMETRIC THz IMAGING FOR DETECTION OF LETHAL AGENTS

In the context of a non-invasive, non-contact method to detect concealed lethal agents employing stand-off imaging in the Terahertz (THz) range using an interferometric detector array, the techniques of image analysis are discussed. To complement the experimental testing, extensive modeling simulates reconstructed images of lethal agents at different THz frequencies applying interferometric techniques. The composite images are consequently analyzed using artificial neural network to make positive identification of the lethal agent. This work also attempts to address issues related to the presence of barriers and isolate some suitable numerical techniques required for rapid and successful recognition of lethal agents during security screening in THz range.