EXPRESS: Spectral Considerations for Standoff Infrared Detection of RDX on Reflective Aluminum

This paper examines infrared spectroscopic effects for the standoff detection of an explosive material, hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), inkjet printed on an aluminum surface. Results of a spectroscopic study are described, using multiple optical setups. These setups were selected to explore how variations in the angles of incidence and collection from the surface of the material result in corresponding variations in the spectral signatures. The goal of these studies is to provide an understanding of these spectral changes, since it affects standoff detection of hazardous materials on a reflective substrate. We demonstrate that variations in spectral effects are dependent on the relative surface concentration of the deposited RDX. We also show that it is reasonable to use spectroscopic data collected in a standard laboratory infrared spectrometer outfitted with a variable angle reflectometer set at 0 as reference spectra for data collected in a standoff configuration. These results are important to provide a systematic approach to understanding IR spectra collection using standoff systems in the field, and to allow for comparison between such data, and data collected in the laboratory. Though the precise results are constrained to a specific material system (thin layers on a reflective substrate), the approach and general discussion provided are applicable to a broad range of IR standoff sensing techniques and applications.

Denoising Autoencoder based Noise Reduction Technique for Raman Spectrometers for Standoff Detection of Chemical Warfare Agents

Raman spectrometers are studied and developed for the military purposes because of their nondestructive inspection capability to capture unique spectral features induced by molecular structures of colorless and odorless chemical warfare agents(CWAs) in any phase. Raman spectrometers often suffer from random noise caused by their detector inherent noise, background signal, etc. Thus, reducing the random noise in a measured Raman spectrum can help detection algorithms to find spectral features of CWAs and effectively detect them. In this paper, we propose a denoising autoencoder for Raman spectra with a loss function for sample efficient learning using noisy dataset. We conduct experiments to compare its effect on the measured spectra and detection performance with several existing noise reduction algorithms. The experimental results show that the denoising autoencoder is the most effective noise reduction algorithm among existing noise reduction algorithms for Raman spectrum based standoff detection of CWAs.