A preliminary study of fingerprint aging using near infrared hyperspectral imaging (NIR-HSI)

Abstract

Fingerprinting is one of the most commonly used techniques to obtain pieces of evidence for identification of individuals. An estimation of how long a trace has been left at a crime scene could represent an important improvement for criminal investigations. There is no reliable analytical method, however, to estimate the age of a fingerprint, since this is an uncontrolled process and changes are affected by factors such as environmental conditions. This study aims to better understand the aging process of fingerprints and identify the relevant variables and limitations of the fingerprint aging process using near infrared hyperspectral imaging (NIR-HSI). For this purpose, aging of the fingerprints of 13 volunteers was evaluated using partial least squares – discriminant analysis (PLS-DA) as a preliminary exploratory approach. Four different modelling approaches were evaluated. The percentage of correctly classified pixels varied from 20.92% to 66.67%. An analysis of the associated spectra found that during the first days of aging the degradation of fat-soluble components, as well as the elimination/absorption of water, seemed to follow non-uniform trends and vary in degradation rate from donor to donor. Better classification tended to occur over longer aging times.