NIST has conducted testing of one-to-one SDK (Software Development Kit) based COTS (Commercial Off-The-Shelf) fingerprint matching systems to evaluate the accuracy of one-to-one matching used in the US-VISIT program. Fingerprint matching systems from eight vendors not used in US-VISIT were also evaluated to insure that the accuracy of the matcher tested was comparable to the most accurate available COTS products. The SDK based matching application was tested on 12 different single finger data sets of varying difficulty. The average true accept rate (TAR) at a false accept rate (FAR) of 0.01% was better than 98% for the two most accurate systems while the worst TAR at a FAR of 0.01% was greater than 94%. The data sets used and the ranking of the systems are discussed in detail in the report.
The National Institute of Standards and Technology (NIST) Evaluation of Latent Fingerprint Technologies-Extended Feature Sets (ELFT-EFS) consists of multiple ongoing latent algorithm evaluations. This report describes the design, process, results, and conclusions of ELFT-EFS Evaluation #1; an accuracy test of latent fingerprint searches using features marked by experienced human latent fingerprint examiners, in addition to automatic feature extraction and matching (AFEM). There has never previously been an evaluation of latent fingerprint matchers of this scale in which systems from different vendors used a common, standardized feature set. The results show that searches using images plus manually marked Extended Features (EFS) demonstrated effectiveness as an interoperable feature set. The four most accurate matchers demonstrated benefit from manually marked features when provided along with the latent image. The latent image itself was shown to be the single most effective search component for improving accuracy, and was superior to features alone in most cases. For most matchers, the addition of new EFS features provided an improvement in accuracy. In several cases, some of the algorithms provided counterintuitive results that may be indicative of implementation issues; therefore, these results are preliminary, and broad conclusions on the efficacy of these features in improving performance should await the subsequent results from Evaluation #2, in which some known software issues are being corrected by the participants. The accuracy when searching with EFS features is promising considering the results are derived from early-development, first-generation matchers. Further studies using next-generation matchers are warranted (and underway) to determine the performance gains possible with EFS. Disclaimer In no case does identification of any commercial product, trade name, or vendor, used in order to perform the evaluations described in this document, imply recommendation or endorsement by the National Institute of Standards and Technology, nor does it imply that the products and equipment identified are necessarily the best available for the purpose.
This report is an extension of the NIST "Studies of one-to-one Fingerprint Matching with Vendor SDK Matchers" which evaluated the accuracy of SDK (Software Development Kit) based COTS (Commercial Off-The-Shelf) fingerprint matching systems for one-toone verification applications [1]. Fingerprint matching systems from twelve vendors were evaluated. The two finger matching evaluation is an extension of that testing used to evaluate the accuracy that can be achieved by combining the index finger scores to achieve a match. These results are based on the SDK matchers provided for the original single finger SDK testing. More details will be available from the Minutiae Exchange Test 2004 (MINEX04) http://fingerprint.nist.gov/MINEX04. The more accurate matchers in the two finger SDK scoring were able to achieve true accept rates (TAR) in the range of .985-.998 at a false accept rate (FAR) of 0.0001. A copy of this report and appendices is available at http://fingerprint.nist.gov/SDK.
No abstract
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.