Pre-processing of inkjet prints NIR spectral data for principal component analysis

Abstract

This paper presents a novel approach in non-destructive analysis of inkjet-printed documents. Our method is based on the combination of molecular spectroscopy in the Near Infrared Region (NIR) and a chemometric method - principal component analysis (PCA). The aim of this work was to prepare spectral data for the analysis of the interrelationships between 19 samples consisting of the same type of office paper on which black squares were full printed in black ink only. The spectra were obtained separately using the Ocean Optics System in two spectral regions, i.e., overtones: 1000-1600 nm and combination bands: 1600-2300 nm, with the paper base. Experimental results confirmed the high reliability of the proposed approach despite the sparse dataset.

Study of Vis-NIR and NIR spectral data of black inkjet prints using principal component analysis for forensic discrimination

“…Based on preliminary PCA tests, it was found out that this approach tends to be consistent in spectral data analysis. 18,20 The pre-processing methods were applied before exporting the spectra in PCA. Vis-NIR spectra were recalculated in optical densities, transformed by interpolation, SG filter and normalisation mean (Figure 1).…”

“…Application of the interpolation and SG filter to NIR spectra. 20 the same samples as group G2 of overtones range PCA model.…”

“…Based on preliminary PCA tests, it was found out that this approach tends to be consistent in spectral data analysis. 18,20…”

“…The optimisation of NIR spectra, before creating the PCA model, by the appropriate pre-processing method is an essential step of inkjet analysis. After preliminary tests and studies, 20 pre-processing transformation methods were implemented as follows: interpolation, smoothing SG filter (Figure 3), standard normal variate (SNV) (Figure 4), detrending, baseline correction and normalisation – mean (Figure 5). These methods were implemented to each spectrum.…”

“…Scores from the PCA for the printer inks from spectral region of 1000-1600 nm. 20 significantly extending the database of available samples of measured spectra. Moreover, the new multivariate classification methods will be implemented to increase the precision of the classification.…”