In the forensic examination of automotive paint, each layer is analyzed individually using IR spectroscopy. Typically, each layer of paint is hand sectioned and presented to the spectrometer one layer at a time for analysis, which can be time consuming. In addition, sampling too close to the boundary between adjacent layers can pose a problem as it may produce an IR spectrum that is a mixture of the two layers. Not having a “pure” spectrum of each layer can prevent a meaningful comparison between each paint layer or in the situation of searching an automotive database can prevent the forensic paint examiner from developing an accurate hit list of potential suspect vehicles. These two problems have been addressed by collecting concatenated IR data from all paint layers in a single analysis of the paint chip after scanning across the cross sectioned layers of the paint sample using a FTIR imaging microscope. Decatenation of the IR spectral data is achieved by multivariate curve resolution using a Varimax extended rotation to select the initial estimates of the reconstructed IR spectra of each layer for the alternating least squares algorithm. By comparing the reconstructed IR spectrum of each layer against the IR spectral library of the a forensic automotive paint database using search prefilters (i.e., discriminants) to reduce the size of the IR spectral library to a specific assembly plant from which the automotive paint sample originated and a cross correlation library search algorithm to identify library spectra most similar to the unknown in the truncated library, we will show that it is possible to identify the correct make and model of the vehicle using these reconstructed spectra. This approach to automotive paint analysis, not only saves time and eliminates the need to analyze each layer separately, but ensures that the recovered IR spectrum of each layer is “pure” and not a mixture.

Multivariate curve resolution and pattern recognition applied to infrared images of paint chips to facilitate the forensic examination of automotive paints

2019 Fall National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.