The MINEX II evaluation was conducted in accordance with the MINEX II Specification which has bee... more The MINEX II evaluation was conducted in accordance with the MINEX II Specification which has been released as a separate NIST Interagency Report, NISTIR 7485. The plan was developed by NIST in consultation with members of the fingerprint and smart card industries, and the general public. The document was drafted in April 2007, and finalized on August 15, 2007. The new NISTIR version adds only a note on context, a coversheet, and acknowledgements to it. It is being referred to here because the document is cited herein, and because it is a suitable protocol for other organizations wishing to evaluate MATCH-ON-CARD implementations. Thoughout this report the names of the vendors are associated with a single letter. This association was instantiated to support automated administration of the test and to effect a containment of the vendor identities within specific personnel at NIST. The letter codes were assigned in approximate order of receipt of the implementation and its passing of subsequent validation and conformance trials. The use of these letters is maintained in this report to conserve space in its many tables. For reference, the letters are associated with the providers' names in a running footnote. A glossary of terms and definitions is given on page 4. Much of the tabulated content in this report was produced automatically. This involved the use of scripting tools to generate directly typesettable L A T E X content. This improves timeliness, flexibility and maintainability, and reduces transcription errors.
The MINEX II trials have been conducted to evaluate the accuracy and speed of MATCH-ON-CARD verif... more The MINEX II trials have been conducted to evaluate the accuracy and speed of MATCH-ON-CARD verification algorithms. These run on ISO/IEC 7816 smartcards. They compare reference and verification data conformant to the ISO/IEC 19794-2 COMPACT CARD fingerprint minutia standard. The test is an assessment of the core viability of matching fingerprints (i.e. the de facto leading compact biometric data element) on personal identity credentials based on the industry-standard smart cards. The results are relevant to users of minutia templates as additional authentication factor. This document is the third publication of MINEX II results. The prior two releases 1 were published
Specific hardware and software products identified in this report were used in order to perform t... more Specific hardware and software products identified in this report were used in order to perform the evaluations described in this document. 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 are necessarily the best available for the purpose. 1 The formal CON-OPS and API specification is available at links/iris/irexII/IQCE conops API v44 26apr10.pdf.
This article describes how to configure Linux-based PCs and an asynchronous transfer mode (ATM) s... more This article describes how to configure Linux-based PCs and an asynchronous transfer mode (ATM) switch to build on ATM network
2007: ― In H.3.4, made inline explicit link to NIST's encryption key for submission of all m... more 2007: ― In H.3.4, made inline explicit link to NIST's encryption key for submission of all material to NIST. ― Deleted redundant "7F 2E xx " from Table 14's data field, which erroneously repeated Table 6 content. ― In H.3.2, added explicit HTML link to the password-protected area- Registered participants are required to download and process a small validation set of images. July 25, 2007: Clarified section H.2.1 on how to participate. July 12, 2007: Schedule deleted here, and moved to newly establish MINEX II homepage:
This article describes how to configure Linux-based Personal Computers (PC) and an Asynchronous T... more This article describes how to configure Linux-based Personal Computers (PC) and an Asynchronous Transfer Mode (ATM) switch to build an ATM network. I will tell you how to obtain and install the ATM support software and device drivers. After the installation steps are given, I describe how to configure the ATM connections on the PCs and the switch to be used for Internet Protocol (IP) network traffic. I give some references at the end of this article on ATM networks and where to get the Linux-ATM software. The ATM interface cards I used are ENI-155P ATM adapters produced by Efficient Networks, and PCA-200EPC adapters from Fore Systems. These cards are installed in standard Pentium or Pentium-Pro based PCs running Linux. The ATM switch I used for this article is a Fore ASX-1000, although the information I give applies to all of the Fore ATM switches. I'll describe how to set up the switch to allow the Linux workstations to use IP over both Switched Virtual Circuits (SVC) and Perm...
Status of this Document This is the final evaluation plan posted on December 24, 2009. NIST inten... more Status of this Document This is the final evaluation plan posted on December 24, 2009. NIST intends that the content of this document is fixed. However, NIST will update the document in response to specific technical issues. NIST may add background information. Comments and questions should be submitted to [email protected]. A FAQ will be maintained in Annex B (to be developed). UPDATEs
Many large scale identity management applications require storage and exchange of standardized mi... more Many large scale identity management applications require storage and exchange of standardized minutiae templates. Minutia templates offer a more spaceefficient, less resource intensive, and more cost effective alternative to raw images. Recent minutiae interoperability tests (ILO, MTIT, MINEX ) all reported variation in minutia selection and placement as the major factor affecting interoperability. This paper quantifies their effects and investigates how variation in selection and placement of minutia from different suppliers relates to loss of performance compared with proprietary templates. We concur with MTIT findings that conformance testing methodologies for evaluating the semantic content of minutia templates is essential and interoperability can be improved by closer adherence to the minutia placement requirement defined in a standard.
March 19, 2006 Table was produced without including vendor B in the computation of the interopera... more March 19, 2006 Table was produced without including vendor B in the computation of the interoperable matchers. This exclusion has been fixed. The tables of the supplemental documents erroneously excluded vendor B in some cases. These errors have been fixed. • Thoughout this report the names of the vendors are associated with a single letter. This association was instantiated to support automated administration of the test and to effect a containment of the vendor identities within NIST. The letter codes were assigned in approximate order of receipt of the implementation and its passing of subsequent shakedown and conformance trials. The ordering is separated by operating system used by the implementation. The use of these letters is maintained in this report to conserve space in its many tables. For reference, the letters are associated with the vendors' names in a permanent footnote. • A glossary of terms and definitions is given in section 2 • The files listed and hyperlinked below accompany this document. They contain tables too numerous to include in this report.
As with all biometric evaluations, the results of this test must be carefully interpreted before ... more As with all biometric evaluations, the results of this test must be carefully interpreted before any predictive conclusions can be made. Users should factor the following into policy, planning and operational decisions. 1. The absolute error rates quoted herein were measured by using the provided implementations on a fixed corpus of operational fingerprint images. However, error rates observed in real-world applications are strongly dependent on a number of factors legitimately not reflected in the experimental design of MINEX II .
July 11, 2007. The final set of changes are: ― Clarification that the BIT structure is a read-onl... more July 11, 2007. The final set of changes are: ― Clarification that the BIT structure is a read-only entity in MINEX II. NIST does not intend to update the BIT, nor require card to support updating of the BIT. This means that the finger position information in the BIT will often be incorrect. ― Fix datatype for PC-based similarity score with that specified for card i.e. use two byte unsigned Integer similarity scores both on-card and off-card. ― Fix match_templates() function – template sizes input values were missing. ― Only one BIT required, if the two would otherwise be identical. ― Tag '83 ' is now optional in BIT. ― Typos, clarifications. June 29, 2007. This MOC test specification is a heavily revised version of the May 24 draft. It has been produced in response to the many comments received, and the decisions made at a NIST workshop. A short final comment period will close on July 5, 2007, and a final plan released on July 9. Please note the main changes: ― Corrected c...
The MINEX II evaluation was conducted in accordance with the MINEX II Specification which has bee... more The MINEX II evaluation was conducted in accordance with the MINEX II Specification which has been released as a separate NIST Interagency Report, NISTIR 7485. The plan was developed by NIST in consultation with members of the fingerprint and smart card industries, and the general public. The document was drafted in April 2007, and finalized on August 15, 2007. The new NISTIR version adds only a note on context, a coversheet, and acknowledgements to it. It is being referred to here because the document is cited herein, and because it is a suitable protocol for other organizations wishing to evaluate MATCH-ON-CARD implementations. Thoughout this report the names of the vendors are associated with a single letter. This association was instantiated to support automated administration of the test and to effect a containment of the vendor identities within specific personnel at NIST. The letter codes were assigned in approximate order of receipt of the implementation and its passing of subsequent validation and conformance trials. The use of these letters is maintained in this report to conserve space in its many tables. For reference, the letters are associated with the providers' names in a running footnote. A glossary of terms and definitions is given on page 4. Much of the tabulated content in this report was produced automatically. This involved the use of scripting tools to generate directly typesettable L A T E X content. This improves timeliness, flexibility and maintainability, and reduces transcription errors.
The MINEX II trials have been conducted to evaluate the accuracy and speed of MATCH-ON-CARD verif... more The MINEX II trials have been conducted to evaluate the accuracy and speed of MATCH-ON-CARD verification algorithms. These run on ISO/IEC 7816 smartcards. They compare reference and verification data conformant to the ISO/IEC 19794-2 COMPACT CARD fingerprint minutia standard. The test is an assessment of the core viability of matching fingerprints (i.e. the de facto leading compact biometric data element) on personal identity credentials based on the industry-standard smart cards. The results are relevant to users of minutia templates as additional authentication factor. This document is the third publication of MINEX II results. The prior two releases 1 were published
Specific hardware and software products identified in this report were used in order to perform t... more Specific hardware and software products identified in this report were used in order to perform the evaluations described in this document. 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 are necessarily the best available for the purpose. 1 The formal CON-OPS and API specification is available at links/iris/irexII/IQCE conops API v44 26apr10.pdf.
This article describes how to configure Linux-based PCs and an asynchronous transfer mode (ATM) s... more This article describes how to configure Linux-based PCs and an asynchronous transfer mode (ATM) switch to build on ATM network
2007: ― In H.3.4, made inline explicit link to NIST's encryption key for submission of all m... more 2007: ― In H.3.4, made inline explicit link to NIST's encryption key for submission of all material to NIST. ― Deleted redundant "7F 2E xx " from Table 14's data field, which erroneously repeated Table 6 content. ― In H.3.2, added explicit HTML link to the password-protected area- Registered participants are required to download and process a small validation set of images. July 25, 2007: Clarified section H.2.1 on how to participate. July 12, 2007: Schedule deleted here, and moved to newly establish MINEX II homepage:
This article describes how to configure Linux-based Personal Computers (PC) and an Asynchronous T... more This article describes how to configure Linux-based Personal Computers (PC) and an Asynchronous Transfer Mode (ATM) switch to build an ATM network. I will tell you how to obtain and install the ATM support software and device drivers. After the installation steps are given, I describe how to configure the ATM connections on the PCs and the switch to be used for Internet Protocol (IP) network traffic. I give some references at the end of this article on ATM networks and where to get the Linux-ATM software. The ATM interface cards I used are ENI-155P ATM adapters produced by Efficient Networks, and PCA-200EPC adapters from Fore Systems. These cards are installed in standard Pentium or Pentium-Pro based PCs running Linux. The ATM switch I used for this article is a Fore ASX-1000, although the information I give applies to all of the Fore ATM switches. I'll describe how to set up the switch to allow the Linux workstations to use IP over both Switched Virtual Circuits (SVC) and Perm...
Status of this Document This is the final evaluation plan posted on December 24, 2009. NIST inten... more Status of this Document This is the final evaluation plan posted on December 24, 2009. NIST intends that the content of this document is fixed. However, NIST will update the document in response to specific technical issues. NIST may add background information. Comments and questions should be submitted to [email protected]. A FAQ will be maintained in Annex B (to be developed). UPDATEs
Many large scale identity management applications require storage and exchange of standardized mi... more Many large scale identity management applications require storage and exchange of standardized minutiae templates. Minutia templates offer a more spaceefficient, less resource intensive, and more cost effective alternative to raw images. Recent minutiae interoperability tests (ILO, MTIT, MINEX ) all reported variation in minutia selection and placement as the major factor affecting interoperability. This paper quantifies their effects and investigates how variation in selection and placement of minutia from different suppliers relates to loss of performance compared with proprietary templates. We concur with MTIT findings that conformance testing methodologies for evaluating the semantic content of minutia templates is essential and interoperability can be improved by closer adherence to the minutia placement requirement defined in a standard.
March 19, 2006 Table was produced without including vendor B in the computation of the interopera... more March 19, 2006 Table was produced without including vendor B in the computation of the interoperable matchers. This exclusion has been fixed. The tables of the supplemental documents erroneously excluded vendor B in some cases. These errors have been fixed. • Thoughout this report the names of the vendors are associated with a single letter. This association was instantiated to support automated administration of the test and to effect a containment of the vendor identities within NIST. The letter codes were assigned in approximate order of receipt of the implementation and its passing of subsequent shakedown and conformance trials. The ordering is separated by operating system used by the implementation. The use of these letters is maintained in this report to conserve space in its many tables. For reference, the letters are associated with the vendors' names in a permanent footnote. • A glossary of terms and definitions is given in section 2 • The files listed and hyperlinked below accompany this document. They contain tables too numerous to include in this report.
As with all biometric evaluations, the results of this test must be carefully interpreted before ... more As with all biometric evaluations, the results of this test must be carefully interpreted before any predictive conclusions can be made. Users should factor the following into policy, planning and operational decisions. 1. The absolute error rates quoted herein were measured by using the provided implementations on a fixed corpus of operational fingerprint images. However, error rates observed in real-world applications are strongly dependent on a number of factors legitimately not reflected in the experimental design of MINEX II .
July 11, 2007. The final set of changes are: ― Clarification that the BIT structure is a read-onl... more July 11, 2007. The final set of changes are: ― Clarification that the BIT structure is a read-only entity in MINEX II. NIST does not intend to update the BIT, nor require card to support updating of the BIT. This means that the finger position information in the BIT will often be incorrect. ― Fix datatype for PC-based similarity score with that specified for card i.e. use two byte unsigned Integer similarity scores both on-card and off-card. ― Fix match_templates() function – template sizes input values were missing. ― Only one BIT required, if the two would otherwise be identical. ― Tag '83 ' is now optional in BIT. ― Typos, clarifications. June 29, 2007. This MOC test specification is a heavily revised version of the May 24 draft. It has been produced in response to the many comments received, and the decisions made at a NIST workshop. A short final comment period will close on July 5, 2007, and a final plan released on July 9. Please note the main changes: ― Corrected c...
Uploads
Papers by Wayne Salamon