We conduct technology evaluations of fingerprint matching, identification, and verification systems.
IARPA has invited the biometrics research community to participate in the Nail-to-Nail (N2N) Fingerprint Capture Challenge. This official U.S. Government Challenge problem seeks to reward researchers for creating autonomous rolled capture devices whose images matched as good as or better than current operator-assisted rolled images.
The Fingerprint Vendor Technology Evaluation (FpVTE) is an independently administered technology evaluation of fingerprint matching, identification, and verification systems.
FpVTE was designed to assess the capability of fingerprint systems to meet requirements for both large-scale and small-scale real world applications.
NIST has conducted two separate slap fingerprint segmentation evaluations to assess the state of the art in fingerprint segmentation technology. SlapSeg04 evaluated a slap segmentation algorithm's ability to segment fingerprints that could be matched using high end fingerprint matching technology. SlapsegII started in 2009 and compares the slap segmentation algorithm's output to hand marked ground truth results. This assesses the segmentation algorithm's ability to preserve the entire fingerprint that is available and not just produce a "matchable" fingerprint. SlapSegII is an ongoing evaluation and allows algorithm developers to submit software at anytime.
The original PFT (2003) evaluation that only reported the matching algorithm's accuracy. The newer PFTII evaluation (ongoing) also reports matcher accuracy information. In addition PFTII will report template extraction times, template size information and matcher timings. PFTII will use both two finger and ten finger datasets to report results on slap-to-slap, slap-to-roll, and roll-to-roll matching. PFTII continues to be only a 1-to-1 verification evaluation it does not report 1-to-many matching results.
The approval of the INCITS 378 fingerprint template standard creates the possibility of a fully interoperable multivendor marketplace for applications involving fast, economic, and accurate interchange of compact biometric templates. Does the template give accuracy comparable with proprietary (image-based) implementations? Can template data be generated and matched by different vendors without attendant increase in error rates? The MINEX evaluation was designed to answer these questions.
NIST Evaluation of Latent Fingerprints Technologies, ELFT07 test was aimed at assessing the core capabilities of current automatic latent matching algorithms. The test is described in the the evaluation concept document. The evaluation will assess technologies submitted to NIST as software development kits which implement the Latent Testing API.
Fingerprint image quality can be improved, either by sensor design, by user interface design, or by standards compliance, better performance can be realized. For those aspects of quality that cannot be designed-in, an ability to analyze the quality of a live sample is needed.