Proprietary Fingerprint Template Evaluations (PFT) Overview

Summary

Proprietary Fingerprint Template evaluations assess the accuracy of end-stage fingerprint matchers. These are the computationally-expensive algorithms used in the last stages of one-to-many Automated Fingerprint Identification System (AFIS) searches. These evaluations are intended to assess the core algorithmic capability of performing one-to-one fingerprint verification.

Description

The National Institute of Standards and Technology's Proprietary Fingerprint Template (PFT) program is an ongoing biometric technology evaluation program measuring the performance of fingerprint template generation and one-to-one verification software using vendor proprietary fingerprint templates. This is as opposed to standardized templates which are measured in NIST's MINEX program. There have been three distinct generations of the PFT evaluation.

PFT III

NIST's one-to-one proprietary fingerprint template generation and matching technology evaluation, PFT III, replaced PFT II in October 2019. PFT III continues to be a one-to-one evaluation of template generation and matching. The evaluation reports results on datasets present in both PFT II and PFT 2003, as well as new PFT III-specific datasets. Rolled and plain optical and ink impressions at resolutions of 500 and 1000 pixels per inch are evaluated as well as non-contact imagery. New datasets are added as they become available.

PFT II

PFT II reported template extraction times, template sizes, and match times. PFT II used both two finger and ten finger datasets to report results on plain-to-plain, plain-to-roll, and roll-to-roll matching. It was replaced by PFT III in 2019.

PFT 2003

The original PFT evaluation reported only one-to-one matching algorithm accuracy for a few small datasets. It was replaced by PFT II in 2010, adding larger datasets and expanded analysis.

NOTE

• High accuracy from the evaluations does not automatically imply that a provider has the capability to field a full-scale AFIS installation. The algorithms submitted for this evaluation are stand alone and not evaluated as part of an AFIS system. PFT is not necessarily representative of a provider's capability to field large-scale identification technologies, because AFIS engineering requires tradeoffs between efficiency, cost, accuracy and other resources, and exploits multi-stage matching techniques to expedite search.
• In no case does identification of any commercial product, trade name, or vendor, imply recommendation or endorsement by the National Institute of Standards and Technology, nor does it imply that the products and equipment identified in the evaluation are necessarily the best available for the purpose.