Spectral and Optimized Marks for Qualitative Material Discrimination

A method for automatic qualitative classification of liquid samples based on their absorption spectrum in the ultraviolet, visible and near-infrared regions is presented. A fiber-optic setup has been employed for material characterization. A simple technique based on image matching is proposed to perform spectra comparison. This alternative implementation of conventional spectrum matching methodologies can be easily translated to hardware platforms thus improving the time response of the classification system. The proposed method does not make any assumption on the probability density function of the data and it is also capable of automatic outlier removal. “Spectral marks” based on the polar representation of the absorption spectra and their “optimized marks” improved by means of Principal Component Analysis have been implemented. Preliminary discrimination results have been obtained on the classification of different oil samples from seeds and olives.