A review of calibration methods for biometric systems in forensic applications
Tauseef Ali, Luuk Spreeuwers, Raymond N. J. Veldhuis · University of Twente Research Information · 2012
When, in a criminal case there are traces from a crime scene - e.g., finger marks or facial recordings from a surveillance camera - as well as a suspect, the judge has to accept either the hypothesis \\emph{$H_{p}$} of the prosecution, stating that the trace originates from the subject, or the hypothesis of the defense \\emph{$H_d$}, stating the opposite. The current practice is that forensic experts provide a degree of support for either of the two hypotheses, based on their examinations of the trace and reference data - e.g., fingerprints or photos - taken from the suspect. There is a growing interest in a more objective quantitative support for these hypotheses based on the output of biometric systems instead of manual comparison. However, the output of a score-based biometric system is not directly suitable for quantifying the evidential value contained in a trace. A suitable measure that is gradually becoming accepted in the forensic community is the Likelihood Ratio (LR) which is the ratio of the probability of evidence given \\emph{$H_p$} and the probability of evidence given \\emph{$H_d$}. In this paper we study and compare different score-to-LR conversion methods (called calibration methods). We include four methods in this comparative study: Kernel Density Estimation (KDE), Logistic Regression (Log Reg), Histogram Binning (HB), and Pool Adjacent Violators (PAV). Useful statistics such as mean and bias of the bootstrap distribution of \\emph{LRs} for a single score value are calculated for each method varying population sizes and score location.