Data analysis of low signal to noise Raman spectra for detecting traces of explosives on surfaces

Kurzfassung

Remote, real-time threat detection is of great interest for security applications. For the detection of hazardous substances, techniques that are chemically selective and highly sensitive need to be available. Among various methods, UV Raman spectroscopy is especially promising, as it is far more sensitive than conventional Raman spectroscopy using visible or near-infrared light sources. When scanning for explosives at security check points an inherently low detection time is accompanied by the need for sensitive trace detection. These prerequisites generally result in very low signal-to-noise spectra and difficult data analysis. Averaging multiple spectra in order to improve the signal-to-noise ratio, however, entails the risk of clouding important spectral signatures present only in a few of the spectra. Therefore, we investigate the use of secondary statistical analysis, such as variance or skewness, to improve the data quality and increase the probability of a successful threat detection. For a first test, measurements of the explosive pentaerythritol tetranitrate (PETN) on a limited number of fabrics as well as the bare materials are performed. This dataset is used to test and compare different methods of statistical analysis and maximise the detection accuracy of an explosive. References [1] Lu, T., et al., Simultaneous determination of multiple components in explosives using ultraviolet spectrophotometry and a partial least squares method. RSC Advances, (2015). 5 (17): p. 13021-13027. [2] Ehlerding, A., et al., Resonance-Enhanced Raman Spectroscopy on Explosives Vapor at Standoff Distances. International Journal of Spectroscopy, (2012). 2012: p. 1-9. [3] Harmon, R.S., et al., Stand-off detection of explosive particles by imaging Raman spectroscopy. Detection and Sensing of Mines, Explosive Objects, and Obscured Targets XVI, (2011). [4] A. Köhntopp, et al., Data Analysis of Weak Raman Signals for the Trace Detection of Explosives, Optica Sensing Congress 2023 (AIS, FTS, HISE, Sensors, ES), Technical Digest Series (Optica Publishing Group, 2023), paper SM3B.4.