A previous ELFT
project existed at NIST about a decade ago. This new iteration expands on the number, source, and types of probes, the type and size of reference databases, and the kinds of Extended Feature Set (EFS) data available. It also adds the option to extract
quality, EFS, and minutiae correlation data to help enable future latent friction ridge research. ELFT will begin this fall, accepting participation on an ongoing basis.
We are actively soliciting feedback
on the test plan
and application programming interface (API)
for the next several weeks as we prepare to launch the evaluation. If you have a few moments and interest in the evaluation, please
review these documents and send any comments to el...@nist.gov
. You may optionally give feedback publicly
on the ELFT GitHub
, especially about technical issues with the API. We are looking for feedback in all
areas, but are principally interested in feedback about:
- timing requirements,
- RAM and storage requirements,
- including anticipated reference database size per 100 000 identities with 10 fingerprints each,
- feasibility of search scenarios and database structures,
- feasibility of palm position localization in search responses,
- availability of useful EFS features,
- whether our procedure mimics your production code, and
- anticipated participation in optional data extraction methods.
We look forward to your comments and your future ELFT participation!