On the importance of rare features in AFIS-ranked latent fingerprint matched templates

In this paper, we introduce an algorithm to generate a score from the matched templates derived by the forensic examiner at the ACE-V stage. Such a score can be viewed quantitatively as a measure of confidence of the forensic examiner for the given latent and impression prints. This quantitative measure can be used in statistics-based evidence evaluation frameworks. Together with the description and evaluation of new realistic forensic casework driven score computation, we also exploit this experimental framework to show the importance of type attributes for minutiae in terms of its discriminating ability in forensic scenarios. We derive the conclusion that together with reliably extracted typical minutiae features, the presence of rare minutiae features helps to improve the measure of confidence of the forensic examiner at the ACE-V stage.