Discrimination of Black Pen Inks on Writing Documents Using Visible Reflectance Spectroscopy and PLS-DA

Abstract

The analysis of inks is performed to verify the occurrence of forgery in documents. Spectroscopic methods are attractive techniques for use in forensic document analysis as they, in most instances, preserve the integrity of the document. In this work, it is proposed a discrimination method of black pen inks of different types, brands and models for application in cursive handwriting. The visible reflectance spectra was obtained by the video spectral comparator VSC®6000 and discrimination analysis performed by partial least squares (PLS-DA). The method was validated with an independent test set and with a blind test. The bias in the results was corrected for the outlier identification. The PLS-DA models presented low root mean squared error of predictions (RMSEPs) and allowed a fast, non-destructive and an efficient discrimination of all pen inks evaluated. The method has proved to be accurate and robust regarding the handwriting of different individuals and capable of identifying the pen type, brand and pen model in a forensic case.