2018-04-30: DET accuracy results from the first public report are being publicly released in tabular form. The tar file is 99 MB.
2018-04-18: NIST is pleased to announce the release of Part 1 of IREX IX. The document, NIST Interagency Report 8207, is a performance evaluation of iris recognition algorithms.
2017-07-11: A third submission phase has been added to IREX IX. The deadline for submission is September 1st, 2017
2017-01-06: The Phase II submission deadline has been extended to January 21st, 2017.
2016-10-03: The final validation package is available for download. The package includes a set of purely synthetic iris images. These images are not at all representative of the actual test data NIST will use for evaluation.
2016-08-25: The source code in the validation package has been updated. For Class B submissions, identify_template() should only be called if an identification template was successfully generated.
2016-08-22: A new validation package is available. It requires downloading CASIA's synthetic iris dataset (http://biometrics.idealtest.org/downloadDB.do?id=4&subset=6).
2016-08-12: An updated version of the IREX IX Concept, Evaluation Plan, and API Specification has been released. It extends the submission deadline for Phase I by one month. The new deadline is October 7th, 2016.
2016-07-27: An updated version of the IREX IX Concept, Evaluation Plan, and API Specification has been released. It includes an updated Participation Agreement .
2016-07-19: A new version of the validation package is available. It now includes an example of pupil and iris boundary points.
2016-06-13: An updated version of the IREX IX Concept, Evaluation Plan, and API Specification has been released. It clarifies that submitted libraries must link according to the C++11 linkage specification, not the C specification.
2016-06-13: A validation package is now available. The validation process ensures the correct operation of participants' software before and after delivery to NIST. The package also include header files for both Class A and Class B libraries. For further instructions, see the readme.txt.
2016-05-23: NIST announces the availability of the final IREX IX Concept, Evaluation Plan, and API Specification . Comments should be sent to email@example.com.
2016-04-19: The IREX IX Concept, Evaluation Plan, and API Specification, Version 0.1 is now available. The document is open for public comment until May 18, 2016. Comments should be sent to firstname.lastname@example.org.
2016-04-17: Plans for the IREX IX activity will be presented at the Iris Experts Group's (IEG-II) meeting on 2016-06-15 at NIST in Gaithersburg, MD.
2016-04-15: NIST is pleased to announce the latest in its series of quantitative activities supporting interoperable high performance iris recognition, IREX IX.
- Verification: One-to-one matching
- Identification: One-to-many matching
- Effects of illumination wavelength: visible through the near infrared, both within and between wavelengths
- Effects of off-angle images: such as might be expected from captures on handheld mobile devices
- Forensic applications: images from cameras not designed for iris recognition
As before, the primary evaluations will be carried out on sequestered data from US government sources – this assures a level playing field for all. Secondary evaluations on publicly available datasets will also be carried out – this aids developers by enabling reproduction of evaluations.
Background: NIST initiated the Iris Exchange (IREX) program in 2008 in order to give quantitative support to the iris recognition marketplace, including support of the ISO/IEC 19794-6 and 29794-6 standards. The program has quantified accuracy of compressed ISO/IEC 19794-6 images (IREX I), image quality assessment algorithms (IREX II), and 1:N algorithms (IREX III and IV). It has guided compression (IREX I and IREX IV), and capture (IREX V). It has projects for iris equipment abstraction (IREX VII) and optical certification (IREX VIII). Other projects may be conducted in the future; please direct suggestions and inquires to email@example.com.